There wasn't a guide when we started out in hospitality. So we built one.

We've been in hospitality for over 25 years. We've supplied over 200 businesses. Cafes, restaurants, mobile vans, food trucks, industrial canteens, school tuckshops, IGA stores. We've watched first-time owners turn a shipping container into a six-figure business. We've watched experienced operators burn through $200K and fold within a year.

We've also built and failed businesses ourselves. Not coffee. Other ventures where we learned the hard way what happens when you don't plan, don't know your numbers, and don't build systems. Those failures became the foundation for how we run BOM today, and how we help every wholesale partner who walks through our door.

What Has Changed, And Why It Helps You

The model in WA used to be simple. Open a decent shop and you would probably do alright, because there was such a short supply of good cafes. A lot of what was out there was chain and franchise, built to be the same everywhere. Any independent that opened had a market crying out for it.

Two things shifted. There are more venues than there were, and people changed their thinking about how often they go out and where the money goes. On top of that there is the constant talk of a cost of living crisis. We don't think it is quite there in the way it gets described, but the talk itself has made customers more deliberate about what they spend, and deliberate is what you now design for.

Here is why that is good news, and we mean that genuinely.

Deliberate customers reward the operator who is also deliberate. When somebody is choosing where two coffees a week go instead of five, they choose the place that gets it right every single time, knows their name, and is worth the walk. That is a game you can win on purpose. It is a far better game than being one of four average cafes that happen to be open on the same street, because effort actually counts for something now.

None of that is a reason not to open. It is a reason to open deliberately, and that is the whole point of what follows.

We are not going to tell you it is easy, because it is not, and anyone who does is selling you something. It is harder than it was ten years ago and the margins are thinner. What we will tell you is that the difficulty is now concentrated in things you can actually control, which was not always true. That is a fairer fight than the one we started in.

Decide What You Are Building

Before anything else, answer one question. Are you building a business to keep, or a business to sell?

They are not the same build, and most people work that out far too late to act on it. A business to keep is designed around your life. The hours you want, the income you need, the days off you actually intend to take. A business to sell is designed around somebody else running it without you, which means documented systems, clean books, and a set of numbers a buyer can believe.

Plenty of the work overlaps. But the decisions where they differ are expensive to reverse, and they start at the lease. Know which one you are doing before you sign anything.

One warning on that, because people take it the wrong way. Deciding to keep the business is not permission to skip the systems. Build it as though you were going to hand it over either way. If you are selling, that is what a buyer is actually paying for. And if you are keeping it, that is the only thing that ever buys you a Saturday off. Systems are not paperwork for an exit. They are the difference between owning a business and owning a job.

This isn't an overview. It's the operating manual. Every section contains the specific detail, the exact platforms, and the real numbers that would have saved us years of trial and error. Trial and error is expensive. Working with people who've already made the mistakes is one of the best investments you can make.

Before you sign anything. Before you tell your mates. Before you start looking at tiles on Pinterest. Answer these honestly.

• Do you have 6 months of personal living expenses saved? Separate from business money. If your mortgage stops getting paid because the cafe had a slow month, you're in trouble before you start. We've seen owners lose their houses. It happens.
• Have you worked in a cafe for at least 6 months? Not as a customer. Behind the machine. The 5am starts, the 40-degree kitchen, the Saturday you missed your kid's birthday because you couldn't get cover. If you haven't lived it, you don't know what you're signing up for.
• Can you wake up at 4:30am for five years? Not a few months. Years. The alarm doesn't care if you were up late. Your customers don't care if you're tired. The machine needs to be on at 5:30.
• Are you okay earning less than your barista for the first year? Probably longer. You're the last one to get paid. That's not a warning. That's the deal. We know operators in their second year still paying themselves below minimum wage.
• Is your family on board? This will affect every relationship you have. Weekends disappear. Date nights get replaced by stocktakes. Your partner needs to genuinely understand what this means, not just say they do.
• Can you handle a customer being rude to your face while you smile? An unfair Google review at 11pm that keeps you up all night. A staff member quitting via text the morning of your busiest day. A supplier letting you down on a Friday. You need thick skin and a short memory.
• Do you have someone you can actually talk to? Not about the business plan. About the 2am stress spiral when the numbers aren't working. Hospo has one of the highest rates of mental health issues of any industry. Have a counsellor, a mentor, a mate who gets it. This isn't soft. This is survival.

If you answered no to more than one of those, you're not ready yet.

That's not discouragement. It's the conversation we have with every new account before we talk about beans. The cafes that survive aren't the ones with the best concept. They're the ones who did the boring work upfront and didn't quit when month three came in under the business plan.

And here's the part that catches people out. Busy is not the same as safe.

We have watched it happen to venues we supply. Queue out the door, phone ringing, and the bank account going the wrong way. Volume hides a lot until it doesn't.

That net margin line is the one people argue with. Operators who've actually run the numbers don't.

Mind Share Over Market Share

The old game was market share. Get the site, get the fit-out up, be the option in the area. In a short market that was enough, because being available was most of the job.

The area is not short any more. There are more venues than the spend supports, and the people walking past you have already decided how many times a week they're going out. You are not competing for a bigger slice of a growing market. You're competing to be the one they think of when they've decided they're only doing it twice this week instead of five.

That's mind share. And it's won differently.

Market share is bought with rent and fit-out. Mind share is built with relationships, consistency, and showing up in your local area on purpose. It costs less money and a lot more discipline. Which is exactly why most venues never do it.

Everything from Chapter 12 onward is about that. Not marketing in the agency sense. The work of being known.

Your Real Bottleneck Is Time

We have run venues at serious volume. Money wasn't the constraint. Staff weren't the constraint.

The bottleneck was time. It was always time.

You get home after a busy day and the admin is still sitting there, and most nights it does not get done. Not because you are lazy or careless. Because you are finished. That is not a character flaw, it is what eleven hours on your feet does to a person.

So the books don't get done. The follow-up doesn't get sent. The idea you had at 5am stays an idea.

This is why systems and automation run through this whole guide instead of sitting in one chapter at the back. Not because AI is exciting. Because time is the only thing you can't order more of on Monday.

Every automation in here has one job: give you back hours so you can put your time into the things that actually convert to money, and into a life outside the shop. Chapter 17 is the toolkit. But the argument for it starts here, on page one, because it changes what you build in Chapter 5 and who you hire in Chapter 10.

The Hard Stuff Everybody Skips

Here's what we see over and over. Operators are so fundamentally focused on the day to day running of the business that the work that would actually change their position never gets touched.

Nurturing your existing customers. Getting out into the local market. Exposing yourself, which means the uncomfortable bit, walking into the gym next door and the real estate office up the road and introducing yourself. Building relationships that have nothing to do with a transaction that day.

None of it is complicated. All of it is hard, because it's the first thing to fall off when service runs long.

That is the whole trap. The modern edge isn't a better bean or a nicer chair. It's being the operator who still does the relationship work in week 40, when everyone else has gone back to just surviving the shift.

Nobody opens a cafe because they love compliance. But every item in this chapter is a potential $5,000 to $50,000 mistake if you skip it. We've watched operators pay the dumb tax on every single one of these. Here's everything you need sorted before you serve a single cup.

This is the chapter that dates fastest. Every fee, rate and penalty below was checked in August 2026, and most of them are indexed or reviewed on 1 July each year. Use them to budget and to know what to ask about. Confirm the current number with ASIC, Fair Work, the ATO or your accountant before you rely on it, and get a lawyer across anything you are signing.

Business Structure

Sole trader is the simplest. You are the business. Easy to set up, cheap to run, but you're personally liable for everything. If the business goes under, your personal assets are on the line. Most cafes start here because it's fast.

Company (Pty Ltd) separates you from the business. Costs more to set up and has an annual ASIC review fee of $342 for a proprietary company. Registering the company with ASIC is $636 on its own, and your accountant's fee sits on top of that, so budget more than the $500 people quote you. What you get for it is protection of your personal assets. If you're investing $150K+, this is worth the conversation with your accountant.

Partnership. Going into business with a friend or a partner? Get a partnership agreement. In writing. Before anything else. Cover: who does what, how profits split, what happens when one person wants out. We've seen friendships destroyed over a cafe. A $1,500 legal agreement prevents most of it.

Then split the roles properly and write those down too. Not vaguely, and not "we'll both do everything", because both doing everything means both assuming the other one did it. Who owns the floor and the food? Who owns the team? Who owns marketing? And the important one: who owns the numbers?

Somebody has to be the person who knows the numbers cold. If neither of you is that person, and plenty of good operators are not, then do not hope one of you grows into it. Find a bookkeeper or accountant you trust and make it explicitly their job. The failure mode is a partnership where both people assume the other is watching the money, and nobody is.

Registrations

• ABN. Free. Apply online at abr.gov.au. Takes 5 minutes. You need this for everything.
• GST registration. Mandatory once you expect to turn over $75K+ (you will). Register when you get your ABN. You'll charge 10% GST and remit quarterly via BAS.
• Business name. If trading under anything other than your personal name. $47 for one year or $108 for three years via ASIC. Take the three years, it's cheaper and it's one less renewal to forget.
• Food business registration. Register with your local council and notify the Department of Health. Required before you serve anything. $200-400 depending on council.
• Employer registration. Once you hire anyone. Tax file number withholding, super guarantee, workers comp. Your accountant sets this up.

Insurance

These are the covers we typically see in a venue, with the sort of numbers people quote. Treat them as the list of things to raise with a broker, not as a recommendation about what you should buy. What you actually need depends on your site, your lease, your equipment and your risk, and only a licensed broker can tell you that. Your landlord will also dictate some of it before you get a choice.

• Public liability ($10-20M). Someone slips on your floor. Someone burns themselves. This is non-negotiable. Most landlords require it before you sign the lease. ~$800-1,500/year.
• Product liability. Someone gets sick from your food. Often bundled with public liability. Make sure it's there.
• Workers compensation. Mandatory in WA the moment you hire anyone. Through WorkCover WA. Cost depends on your industry classification and payroll. Budget ~2-3% of wages.
• Business insurance (contents, equipment, stock). Your $20K espresso machine, your fridges, your fit-out. What happens if there's a fire, flood, or break-in? ~$1,500-3,000/year.
• Business interruption. Optional but smart. If you can't trade for 3 months due to a flood or fire, this covers your fixed costs. We know operators who wished they had this.
• Management liability. If you're a company director. Covers you for things like wrongful dismissal claims, workplace disputes. Optional for small operations.

Food Safety & Compliance

• Food Safety Supervisor certificate. Under Standard 3.2.2A your business must appoint a certified Food Safety Supervisor before you handle unpackaged high-risk food. It's one appointed FSS for the business, not one on every shift. The rule is that they're reasonably available to advise and supervise your food handlers, meaning onsite while that food is being handled or easily contactable. Appointing someone who is never in the building doesn't count. It's a one-day course, ~$150, and the certificate has to have been issued in the past 5 years. In practice, certify two or three people so a sick day doesn't leave you exposed.
• Food handler training. All staff who handle food. Can be done online in 2-3 hours. ~$25-50 per person.
• Food Safety Plan (HACCP-based). Document your critical control points. Temperature logs for fridges. Cleaning schedules. Allergen management. Your Environmental Health Officer will check this.
• Allergen management. You must be able to tell customers what allergens are in your products. Display allergen info. Train every single staff member on this. An anaphylactic reaction in your venue is a nightmare scenario.
• Temperature logs. Fridge temps checked and recorded daily. Hot food held above 60°C, cold food below 5°C. These logs get inspected.

Employment Law Basics

Get this wrong and Fair Work will find you. They actively audit hospitality. For an individual, the maximum civil penalty a court can impose is $21,840 per contravention, rising to $218,400 where the court finds a serious contravention. Those are ceilings, not starting points, and the court decides where you land. Since 1 January 2025, intentionally underpaying someone is also a criminal offence, carrying up to 10 years in prison for an individual. And when it hits the news, every potential employee sees it.

• Work out which award covers you before anything else. A standalone cafe or restaurant sits under the Restaurant Industry Award 2020 (MA000119). But if your cafe operates in, or in connection with, a hotel, motel or resort, MA000119 specifically excludes you and you're under the Hospitality Industry (General) Award 2020 (MA000009) instead. A food and drink counter that's part of a retail business can land under the General Retail Industry Award 2020 (MA000004). Different award, different rates, different penalties. Read the coverage clause of each one at fairwork.gov.au, and if your setup is unusual, ring the Fair Work Infoline on 13 13 94 and get it confirmed before your first pay run.
• Your award is your bible for pay rates, penalty rates, overtime, breaks, and allowances. Read it. Your accountant and Employment Hero will help you apply it, but you need to understand it.
• National Employment Standards (NES). 11 minimum employment conditions that apply to every employee. Maximum weekly hours, leave entitlements, notice of termination, redundancy pay.
• Fair Work Information Statement. You must give this to every new employee. Before they start.
• Superannuation. 12% on top of wages for all eligible employees. Since 1 July 2026 super is paid every pay cycle, not quarterly, and it's calculated on qualifying earnings rather than ordinary time earnings. Do not fall behind on this. The penalties are brutal.
• Single Touch Payroll. You report every pay run to the ATO in real time. Your payroll software handles this (Employment Hero does it automatically).
• Record keeping. 7 years minimum. Timesheets, pay records, leave records. All of it. Again, Employment Hero keeps it all.

Other Licences

• Liquor licence. If serving alcohol. Apply through the WA Department of Local Government, Sport and Cultural Industries. Takes 4-8 weeks. Budget $400-1,000 for application fees. You'll also need an Approved Manager certificate ($150, online course).
• Music licence. If you play music (you will). Two licences needed: APRA AMCOS (for the songwriters) and PPCA (for the recording artists). Combined cost ~$300-600/year for a small cafe. Yes, Spotify in a commercial venue requires this.
• Outdoor dining permit. Tables on the footpath. Each council is different. Some charge per square metre. Some require bollards and barriers. Check before you plan your seating.

This is the chapter that saves you from yourself.

The Fit-Out Trap

The number one killer of new cafes in Perth is not bad coffee. It's not bad location. It's spending too much money before you pour a single cup.

We've seen cafes spend $250K+ on fit-outs. Custom joinery. Imported tiles. Bespoke lighting.

They look incredible. For about eight months. Which is roughly how long the money lasts when you've burned the budget on the build and forgotten about operating costs.

The cafes that survive spend the minimum to be clean, functional and inviting, then let the coffee and the service build the reputation. You can always upgrade later when you're making money. You can't unfurnish a cafe after you've blown the budget.

Startup Cost Breakdown

Monthly P&L Reality

Here's what a realistic month looks like for a small cafe doing okay. Not killing it. Just okay.

Read the staff line carefully, because it's the one most guides lie to you about.

That $3,000 is before tax. Before loan repayments if you borrowed to start. Before the espresso machine throws an error and costs $2K.

That's 7.5% net margin in a month that went fine. Add a broken grinder and a sick barista on a long weekend and you're at zero. That's not a reason not to do it. It's a reason to know your numbers down to the cent.

About That 45%

You will read, in a lot of places, that labour should be 30% of revenue. Some guides stretch to 35% and call it the pass mark.

Anyone quoting you 30 to 35% is not describing 2026.

Award rates have moved every July. Super went to 12% and is now paid every cycle. Casual loading sits on top of a base that has climbed for years running. Do that arithmetic against the prices a suburban cafe can actually charge and you get 40 to 50%. That's the real band. We plan to it, our accounts run in it, and the operators we know who claim 32% are either not paying themselves for the hours they work or not counting super and workers comp.

We're telling you this because the alternative is worse. If you hold yourself to a number that isn't real, every single week reads as a failure, and after six months of failing an imaginary test most people stop looking at the numbers altogether. That's how operators go blind to their own business.

Sitting at 45% is not a crisis. Sitting at 45% and not knowing it is.

None of that is bad management. It is the actual arithmetic of running an independent in 2026. The lever isn't heroic rostering. It's the price on the board, the average spend per customer, and how many hours of your own life the business needs to function.

Design Your P&L Before You Need It

Most operators inherit a profit and loss from their accountant, glance at the bottom line, and never touch the structure. That is a mistake, because a P&L you designed tells you where a problem is. A generic one only tells you that you have one.

Two rules make it useful.

Break your sales into the areas you actually run, and mirror them in cost of sales. Coffee, food, retail, catering. If coffee sales sit in one line and every ingredient in the building sits in a single "purchases" line underneath, you can never answer the only question that matters: which part of this business is making money and which part is being carried? Split the revenue, split the costs to match, and the answer falls out on its own. This is the single most useful hour you will ever spend with your accountant.

Put your own labour where it truly belongs. If you are on the machine four days a week, your wage is a cost of sale, the same as anyone else's, and it belongs in there. But the hours you spend on strategy, marketing, admin and building the business are not production. They are overheads.

How To Cost A Coffee

We ran a venue for years before we could have told you this number off the top of our heads. Here's the breakdown for a standard flat white, so you don't have to wait as long as we did:

Two notes on the inputs. We have costed a 12oz because that is what most venues actually pour, and the milk is the thing that scales with cup size. And we have used $1.88 a litre, which is roughly what a new site pays before it has any volume behind it. If your roaster gets you a better dairy rate, brilliant, that is margin you did not have to fight for. But build the model on the number that is true on day one, or every projection you make is quietly optimistic.

Two things trip people up here, and both of them flatter the number.

First, the GST. A $6.50 coffee is not $6.50 of revenue. Fifty nine cents of it belongs to the tax office and you are only holding it. Every margin above is worked on the $5.91 you actually keep. Cost against the sticker price instead and every drink on your board looks about 9% healthier than it is.

Second, which costs you count. On ingredients alone, coffee and milk, this cup runs at 76.2%. Put the cup, the lid and the sugar back in and it is 71.3%. Neither is wrong. Just know which one you are saying out loud, because the Chapter 11 benchmark is the ingredient number and your P&L only ever sees the loaded one.

And note where that leaves you against the 77% we push people towards. At $6.50 you are close but not quite there. To clear 77% on ingredients this cup wants to be about $6.75, which is a 25 cent move that no regular will mention and most will not notice. That is the conversation in Chapter 11 about not being afraid to price properly, arriving with a number attached.

At $44/kg beans: $0.88 per dose. At $39/kg: $0.78. That 10-cent saving across 200 cups a day is $20. Sounds like something. Over a year it's $5,200. Now ask yourself what you're giving up for that. Consistency. Training. A roaster who picks up the phone. Your customer pays $6.50 either way. Don't cheap out on the one thing they consume 260 times a year.

How To Cost A Dish

Every menu item needs a costed recipe card. For each dish:

1. List every ingredient with the quantity used per serve
2. Calculate the cost per ingredient (purchase price / portions per pack)
3. Total all ingredients = your COGS per dish
4. Target food cost: 28-35% of sale price
5. Sale price = COGS / target food cost percentage

Example: If a toastie costs you $2.80 in ingredients and you want a 30% food cost, your minimum sale price is $2.80 / 0.30 = $9.33. Round to $9.50 or $10.

Do this for every item on your menu. The items that don't hit your margin target either need re-engineering (cheaper ingredients, smaller portions) or a price increase. No ego. The numbers don't lie.

Then rank them, because this is the part we got to embarrassingly late. A small handful of your menu items will generate most of your profit. Your top 5 or 6 by profit contribution are the ones worth obsessing over. Better ingredients, faster prep, prominent placement.

The ones nobody orders that barely break even? Cut them. A smaller, sharper menu is faster to execute, easier to train and more profitable.

Dishes like that only ever turn up because somebody went looking. That is the whole reason this chapter exists.

Opening a cafe does not make it popular. Where you open it does most of that work before you have poured anything.

Think about a hot dog stand. Put one in what looks like the right part of town, good street, nice foot traffic, and you can stand there all day selling almost nothing. Push the same cart outside a nightclub at two in the morning and you sell out. Same product. Right location, right time.

That is the whole chapter. Everything below is how to find out which one you are about to sign a lease on.

Go And Count

Before we committed to our own site, we stood outside it at 5am. Not once. Every morning, for months, before we had decided anything.

We counted. Who walks past and when. Which direction they are heading. What they are wearing, because a hi-vis crew at 5:30 and a school run at 8:15 are two completely different businesses. Where they park. How long they stay. Which way they look as they come around the corner.

It costs nothing but early mornings, and it is the only research that has never once lied to us. A leasing agent will tell you about foot traffic. A spreadsheet will give you a catchment population. Neither of them tells you that everybody walking that footpath at 7am is heading to the station and will not stop for anything that takes longer than ninety seconds.

Do it on a weekday and a weekend. Do it in the rain. If the numbers only work in your head, they do not work.

Know Who Actually Lives There

Once you know who walks past, work out who they are. The area decides your menu, your hours, your pricing and your staffing far more than your taste does.

• Next to a primary school. Parents dropping kids before work. That is a hard, narrow, brutal morning peak and then quiet. Speed is everything. A pram has to fit. If you are slow at 8:15 you will not get a second chance, because they have a meeting.
• Blue collar and industrial. Tradies with real disposable income who want it strong, fast and every single day. They are the most loyal customers in hospitality and they do not care about your tiles.
• Office and professional. Two peaks, a lunch trade, and almost nothing on a weekend. Great Monday to Friday, and you need a plan for Saturday or you carry rent for nothing.
• Tourist and high traffic. Excellent while it lasts, and it does not last all year. Which brings us to the thing people forget.

The Postcode Trap

Everyone wants Fremantle. Or Leederville. Or Mount Lawley.

The rent in those areas will eat you alive unless you're doing serious volume from day one. And the saturation is worst exactly where the postcode is best, which means you're paying the highest rent in the state to fight the hardest fight for attention.

Some of the best-performing cafes we supply are in suburbs you wouldn't expect. Rockingham. Baldivis. Secret Harbour. Mandurah. Places where there's demand but not saturation. Where parking exists. Where families actually live and work.

The Three Things That Actually Matter

1. Foot traffic or destination traffic. You're either on a walking path (train station, shopping strip, school run route) or you're good enough that people drive to you. Count foot traffic at the site. Stand there at 7am on a Tuesday and count people. Do it again on a Saturday. If the numbers don't work on paper, they won't work in reality.
2. Parking within 50 metres. Perth is a driving city. If people can't park nearby, most won't stop. Count the parking spots. Check if they're time-limited. Check if they're shared with other businesses that peak at the same time as you.
3. Rent under 10% of projected revenue. If your rent is $4K/month, you need $40K+/month in revenue. Be honest about whether that's realistic for the location, for a new cafe, with no established customer base.

Suburb Types That Work

Growth corridors (Baldivis, Byford, Ellenbrook, Wellard): high demand, not yet saturated. Families moving in faster than cafes can open. Lower rent than inner city. Some of our highest-volume accounts are in these areas.

Industrial estates: tradies are the most loyal customers on earth. They want it strong, they want it fast, and they'll come every single day. No weekday competition. Often zero other options nearby.

Near schools and sports clubs: weekend trade goldmine. Parents dropping kids at sport need coffee and have 45 minutes to kill. Captive audience.

Shopping centre kiosk: lower fit-out, guaranteed foot traffic, higher rent percentage. Good way to test a concept before a standalone site.

Lease Negotiation

Your lease is probably the most important document you'll sign. Get a commercial lease lawyer. Not optional. $500-1,000 that will save you tens of thousands. And remember: the landlord needs a tenant as much as you need a premises. Everything in a commercial lease is negotiable. Most first-time operators don't know that.

Key terms to negotiate:

• Initial term. Push for a 3-year initial with two 3-year options, rather than a 5-year lock-in. If the business doesn't work, a 5-year lease with personal guarantee will follow you.
• Rent-free period. Ask for 2-4 weeks rent-free for fit-out. This is standard. Some landlords offer more for longer leases.
• Fit-out contribution. The landlord wants a tenant. Ask if they'll contribute $10-30K toward the fit-out. More common than people think, especially in newer developments or empty sites.
• Outgoings. Understand exactly what you're paying beyond base rent. Council rates, water rates, centre management fees, marketing levies, insurance. Ask for a breakdown of the last 12 months of actual outgoings.
• Annual increases. Negotiate a fixed percentage (3-4%) rather than CPI-linked, which can spike. Or a market review at option exercise.
• Personal guarantee. If you're operating as a company, the landlord will probably ask for a personal guarantee. Try to limit it to 6 months rent rather than the full term. This is negotiable.
• Make-good clause. What happens when you leave? Some leases require you to return the premises to its original condition. That can cost $20-50K. Negotiate a reasonable make-good, or a cash settlement amount.

• Count foot traffic at the site at three different times of day (7am, 12pm, 3pm) on two different days (weekday and weekend)
• Count available parking within 50m and check time limits
• Visit every other cafe within 1km. Note quality, pricing, busy periods, Google ratings
• Check Google Maps reviews for all competitors. Read the 1-star reviews. That's your opportunity
• Talk to neighbouring businesses about the area. How's foot traffic? Any issues with the landlord?
• Check council website for any planned developments nearby (new housing = future customers, new shopping centre = future competition)
• Get a building inspection. Check for plumbing, electrical, ventilation, asbestos, structural issues
• Verify the site has existing food premises approval (or what it'll take to get it)
• Check the flood zone and bushfire risk rating with local council
• Confirm the zoning allows food service (some retail zones have restrictions)
• Check if the site has an existing grease trap. If not, get a plumber quote before signing. Retrofitting one can cost $5-15K and needs council approval

Think about that before you sign, because it is the thing that turns a good year into a reason to leave. Every hour of work you put into the business gets valued at review time, and you are the one who created the value being repriced.

Which is the argument for two things. Push hard on the increase mechanism now, when the landlord still needs a tenant and you still have something they want. And build the kind of following that walks to your next site if it ever comes to that.

Sometimes things sound good and look good. The question nobody asks early enough is whether they actually work operationally and get you the return you want.

A pretty cafe is important. Massively important, especially these days. People choose with their eyes long before they taste anything, and a room that photographs well does real work for you on a Saturday.

But pretty on its own is an expensive way to be busy and broke.

The layout decides your speed of service, your revenue per hour, your average order value and how many people you need on the floor to get through a rush. Which means it decides whether the business gives you the lifestyle you wanted from it, or quietly takes that lifestyle instead. Because it is a lifestyle. Hospitality is a commitment. It really is.

Every metre matters. Design for both.

Floor Plan Principles

POS faces the entrance. Your point of sale should be positioned so that the person taking orders is always looking up at guests as they walk in. First eye contact. First smile. First impression. This one detail changes the energy of every single visit.

Sellable materials at the POS. Retail coffee bags, keep cups, pastries, bottled drinks, merchandise. Everything within arm's reach of the person paying. This is where average order value lives. A customer ordering a $6.50 flat white who also grabs a $22 bag of beans has more than tripled their transaction. Position these items at eye level, within the natural sightline of someone waiting for their order.

Customer flow is one-directional. Enter, order, wait, collect, sit or leave. No cross-traffic. No bottlenecks. No customers standing in the way of other customers. Map this on paper before you build anything.

Service positions are defined. In a two-person cafe, Position 1 is POS and customer interaction. Position 2 is the machine and drinks. They don't swap mid-service. Each person has a defined zone and a defined role. When you add a third person, they handle food and clean-up. When you add a fourth, you split drinks (one on machine, one on alternative brew).

Here's why the layout matters to you personally, not just to the service. A well-designed floor plan means your team can run a rush without you standing in it.

If the cafe only works when you're behind the machine, you haven't built a business. You've bought a job that pays below minimum wage and can't be sold. The goal is a floor plan good enough that you can step away and the quality doesn't drop. That is the first hour you ever buy back.

Count The Footsteps

Engineer every part of the bench around functionality and consistency. Put things where they need to be, not where they look good in a photograph.

Walk the job before it is built. How far is the thing you reach for two hundred times a day? Is anything you touch once a week sitting in prime position? Every unnecessary step is a second, every second is repeated all morning, and by the end of a year it is days of your life and a slower queue.

Bench space is the most expensive real estate you own. Treat it that way.

The Things Everybody Forgets

These are the ones that come up six weeks after opening, when it is expensive to fix.

• Ice. Almost nobody plans for ice, and almost everybody runs out. Iced lattes, iced long blacks, milkshakes, smoothies, cold water on a 40-degree day. You will use far more than you think and summer is exactly when you cannot afford to be scooping it out of a bag from the servo. Get a machine, size it for January and not for June, and put it where the person making cold drinks is standing. Not in the back.
• Freezer capacity. Always bigger than you planned. Ice, frozen goods, backup stock, batch prep. An undersized freezer forces small frequent orders, and small frequent orders cost more per unit and eat your week.
• Where the cold drinks get made. If your ice, your blender and your cold cups are three separate trips from the till, every cold order slows your whole line down. Group them.
• Power and water in the right places. Decided by your machine, which is why the machine gets chosen before the bench gets built, not after.
• Somewhere for deliveries to land. A milk delivery at 6am with nowhere to go ends up in the customer walkway.

How It Looks At Night

Think about what your venue looks like after dark, even if you never intend to open then.

A room lit for 7am service is flat and bright, and it looks like a waiting room at 6pm. If there is any chance you might one day do evening trade, wine, food, events, a private booking, then the time to sort the lighting is while the electrician is already on site. Retro-fitting it later means opening up walls you just paid to close.

The cheapest version of this is dimmers. Put them in. They cost very little during a fit-out and they hand you a completely different room at the flick of a switch. Plenty of venues have found a second revenue stream simply by being somewhere people want to sit once the sun goes down.

Speed of Service Design

Every second of service time is revenue. A cafe that takes 3 minutes per customer serves 20 per hour. A cafe that takes 2 minutes serves 30. Same staff, same rent, 50% more revenue. At a $5 average spend, 10 extra drinks per hour across a 10-hour day is $500/day. Over a year, that's $130K. From a better layout.

• Minimise POS button presses. Your most popular items should be one tap. Flat white, latte, cappuccino, long black. If your barista is navigating three menus to enter a flat white, you're losing 5 seconds per order. Over 200 orders, that's 16 minutes of wasted time. Build your top 8 drinks as quick-order buttons on the home screen and leave everything else a level down.
• Bump screens. Orders go from POS to a bump screen at the machine. No dockets falling on the floor. No shouting. No mistakes. Barista taps the order off when it's done. We use Pulse Systems. It changed our speed of service immediately.
• Mise en place. Everything in its place before service starts. Cups stacked. Lids ready. Milk pitchers positioned. Syrups within reach. A barista should be able to make a flat white without taking a single step.
• The 30-second handoff. From espresso extraction to customer handoff should be under 30 seconds for a standard milk drink. If it's longer, something in your setup is wrong.

Permits and Build Timeline

Allow 8-12 weeks for permits. Everyone underestimates this.

1. Food business registration. Council + Department of Health. Required before you open.
2. Development approval. If changing use class (retail to food). Check with council first.
3. Building permit. Any structural, plumbing, or electrical work.
4. Health inspection. Environmental Health Officer inspects before opening. Stainless surfaces, temp control, handwash basins, pest management.
5. Liquor licence. 4-8 weeks if applicable. Start early.
6. Outdoor dining permit. Each council has different rules and fees.

Here's something nobody in the equipment business will tell you.

Your grinder matters more than your machine. Your water filter matters more than your grinder. And your scale matters more than all of it.

The most expensive mistake we see is a $25K machine paired with a $1,500 grinder. That's a race car on cheap tyres. It happens because the machine is the thing customers can see, and the grinder is the thing that decides what's in the cup.

Priority Order

1. Grinder. The single most important piece of equipment. A good grinder with a mediocre machine will produce better coffee than a bad grinder with a $30K machine. Budget 60-70% of your machine cost. You need two: house blend and guest/single origin. Mazzer, Mahlkonig, Eureka Atom range. Get grinders with on-demand dosing, not timed dosing.
2. Water filtration, and go reverse osmosis. Perth water is hard. It will scale your machine, shorten its life and flatten every cup you make. A basic carbon filter improves the taste a little and does very little about the scale. Our position is that RO is worth it every time. It strips the water back to nothing, which means no scale, and it means the water tastes the same in January as it does in July instead of drifting with the supply. Because RO removes the good minerals along with the bad, run a system with blending or remineralisation so you can dial it back to around 150 parts per million, which is where espresso actually wants to be. Straight RO with nothing added back makes flat, hollow coffee, so this second half is not optional. Chapter 8 goes into why. Budget more than a carbon filter and less than a service call on a scaled-up boiler. It protects a machine worth many times the price and it is the cheapest consistency you will ever buy.
3. Scales. Consistency comes from measuring. Acaia, Felicita, or Timemore. At least two. One per grinder station. Dose every shot. Every single one.
4. Espresso machine. Yes, fourth. Entry ($5-12K): the ACM Evolve is where we point most new sites, a 2 group lands around $6,000 to $6,600 with a 13.5L boiler, it is built in Italy and it is a genuinely reliable workhorse. Expobar and a refurbished La Marzocco Linea Classic also belong in this bracket. Mid ($12-22K): La Marzocco Linea PB, Victoria Arduino Eagle One. Premium ($22-40K): KB90, Synesso, Slayer. Match to your volume. A $30K machine doing 100 cups/day is a waste. A $10K machine doing 400 cups/day will die.
5. Two group or three? Price the three before you decide, even if you are certain two will do. Two groups will run most small sites comfortably. The problem is the site that works, because outgrowing your machine at month six is a very expensive way to find out you were right. The gap in price between two and three is far smaller than the gap between a queue you can serve and one you cannot. Check your bench will take it before you fall in love with the idea. Also watch the compact versions: they are usually about 10cm narrower with a much smaller boiler, and that boiler is what carries you through a rush.

Everything Else

• Commercial dishwasher. Hobart, Washtech. Under-bench. You'll run it 50+ times daily.
• Under-bench fridges (x2 minimum). Milk (you'll use more than you think) and food.
• Display fridge or ambient cabinet. If doing food/pastries.
• Commercial blender. For frappes, smoothies, iced drinks. Vitamix or equivalent.
• Knock box, tamper, WDT tool, milk pitchers (x4 minimum).
• Cleaning: Cafetto, group head brush, milk jug rinser, sanitiser. Budget $50/month.
• Thermometers (digital probe for food safety compliance).
• Signage (A-frame, window graphics, menu boards).

For the full breakdown on choosing the right machine for your volume and budget, just ask us. We've seen every machine in every price bracket in real cafes. We'll tell you what works and what breaks. Book a chat and bring your budget number.

One last thing on gear, and it's the same argument as the fit-out.

Equipment is the most photogenic money you will ever spend, and the least persuasive. Nobody has ever chosen a cafe because of the badge on the machine. They choose it because the coffee was the same on Tuesday as it was on Friday, and somebody knew their name. The gear only exists to make that repeatable on your worst-staffed morning.

Every tool in this chapter is judged on one thing: does it hand you back time.

The right stack saves you hours every week. The wrong one quietly costs more than it saves, because someone still has to feed it. Here's what we use and why.

Point of Sale: Square

For a new site, Square is what we recommend, and we say that as a Square partner so take the disclosure with it.

It sets up in an afternoon rather than a project. The hardware is cheap enough that a second terminal is not a big decision. Online ordering, gift cards and loyalty are already in the box instead of being three more subscriptions. And because it is the system most people have already used somewhere, training a new starter on it takes minutes.

Being partners means we can usually help with the terminals, get online ordering and loyalty switched on properly, and walk you through onboarding rather than leaving you with a login and a shrug. Ask us before you buy anything.

The heavier options. Revel and the other enterprise platforms give you deeper inventory and multi-site reporting, and there is a point where that genuinely earns its cost. That point is not your first site. Start on Square, learn what you actually need from a POS by running one, and move later if the business demands it. Choosing a complex system on day one mostly means paying for reports nobody opens.

Whichever you choose, set it up with your real menu before you open. Test it with your staff. Time them entering orders. Your most popular items should be one tap. If a barista is navigating three menus to ring in a flat white you are losing five seconds an order, and across 200 orders that is sixteen minutes of standing still. Rearrange the buttons until the common orders are the fastest.

Bump Screens: Pulse Systems

Orders from POS appear on a screen at the machine. No paper dockets. No shouting. The barista sees the queue, makes the drink, taps it off. It changed our speed of service overnight. We went from 3-minute average to under 2 minutes per drink.

The screen also tracks order times, so you can see where your bottlenecks are. If lattes are averaging 2:30 and flat whites are averaging 1:45, you know milk steaming is slowing you down.

HR & Payroll: Employment Hero

We use both Employment Hero HR and Employment Hero Payroll. Two separate products from the same company, and they talk to each other properly, which is not something you can assume.

HR platform: onboarding (new starters fill out everything digitally before day one), document storage, policies, leave management, performance reviews. Every employee has their tax file declaration, super choice form, and Fair Work Information Statement stored digitally. No paper.

Payroll: automatically applies the Restaurant Industry Award rates including penalty rates for weekends, public holidays, and overtime. Calculates super. Handles Single Touch Payroll reporting to the ATO. Does everything that used to take 3 hours per week in 20 minutes.

Rostering: Employment Hero has a built-in roster that shows you the cost of each shift as you build it. You can see what the week will cost before you publish it. That's the whole value. Not hitting some magic percentage, but finding out on Wednesday that the week is heading for 52% while you can still move an hour.

Accounting: Xero

Industry standard for small business. Your accountant will almost certainly recommend it. Connect your bank feeds, POS, and payroll. Your bookkeeper (or you) does weekly reconciliation. Your accountant does quarterly BAS and annual tax return. Everything is in one place.

Business Wiki: Notion or Obsidian

Every recipe, every process, every checklist, every supplier contact, every training document. All in one place. When a new staff member starts, they get access to the wiki. Opening procedures, closing procedures, how to clean the grinder, how to handle a complaint, allergen information, supplier phone numbers. Everything.

Notion is the easier start and the better one if the team needs to get in there too. Free to begin with, works on a phone, and anyone can be shown how to use it in ten minutes. Obsidian is the pick if you want your notes as plain files you own outright on your own machine, with no company between you and your own knowledge. We run both, for different jobs. Either is right. Having neither is the problem.

This is how you build a business that runs without you. If the knowledge is in your head, it dies when you take a day off. If it's in a wiki, anyone can do it.

Team Training: NotebookLM

We upload our training documents, brand voice guide, coffee knowledge base, and service standards into NotebookLM. New team members can ask it questions in natural language. "How do I dial in the grinder?" "What's our complaint handling process?" "What temperature should the fridge be?" It's like having a senior trainer available 24/7 who never gets tired of answering the same question.

The part most people miss: it will turn those same documents into audio and video explainers. Change a process, upload the new page, and it produces something your team can actually watch or listen to on the way to work. That beats a laminated notice nobody reads, and it means a process change lands with everyone the same way instead of depending on who was on shift when you explained it.

AI Tools (The New Layer)

Everything above is the foundation. The layer we're adding now is AI. Not future-tech. Tools you can use this week.

• Claude. This is the one we recommend, and we have tried all of them properly rather than for twenty minutes. It writes better, it holds a long detailed task without losing the thread, and it is the one that sounds least like a robot wrote it, which matters when the output is going in front of your customers. Free to start. Use it for social captions, menu descriptions, replies to Google reviews in your voice, roster analysis, recipe costing. ChatGPT is fine and has a good voice mode if you want to talk to it hands-free on the drive in. But if you are only going to learn one, learn Claude.
• NotebookLM. Free, from Google, and the fastest win in this whole list. Upload your SOPs, training docs and coffee specs and your team can ask it questions in plain English. "What's the allergen info for the banana bread?" Available 24/7, no waiting for the manager.
• Launchpad (via BOM). Our platform for wholesale partners. Ordering reminders, marketing templates, operations tools. Free with your coffee account. This is the stuff we built for ourselves that we're now opening up.

Other Platforms

• Google Business Profile. Free. Non-negotiable. This is how people find you. Update it weekly.
• Instagram. Your primary social platform. Post 3-5 times per week. Stories daily.
• Canva. For menus, social graphics, signage. The free plan is enough to start.
• A spreadsheet for inventory tracking until your volume justifies a dedicated platform like MarketMan. Do not buy stock software before you have par levels.
• Loom. Free. Record your screen or your bench, share a link. Show the weekend casual how to set the cabinet up once instead of explaining it a fourth time.

Look at that total against one thing: the hours it gives back.

Payroll that used to take three hours a week now takes twenty minutes. That's roughly ten hours a month, which at the end of an eleven hour day is the difference between the books getting done and the books getting postponed again. We know which one wins when you're tired, because we lost that argument with ourselves for years.

The stack isn't there to make you feel modern. It's there so the work that actually moves the business, the relationships, the local market, the follow-up, has somewhere to live in your week.

Your coffee is the one thing your customers consume every single day. A regular comes in every weekday, that's 260 coffees a year. 260 chances to keep them or lose them. The difference between $39/kg and $49/kg beans is 20 cents per cup. Your customer pays $6.50. That 20 cents is the gap between good and great.

What To Look For In A Roaster

• Consistency. Not how good the best cup is. How good the worst cup is. That's your real standard.
• Training support. Will they come to your site, train your staff on your machine, and help you dial in? This is worth more than a $2/kg discount.
• Delivery reliability. Running out of beans on a Friday morning is a business emergency. Your roaster should make that impossible.
• Relationship. Do they know your name? Will they pick up the phone on a Saturday? Do they actually care if you succeed?
• Volume flexibility. What is their delivery minimum, and does it work for a site that hasn't opened yet? Will they still handle 100kg a week when you grow? Ask both questions. You want a roaster who can scale with you, not one who is only interested at one end of that range.

Ask About The Roast Date, Not The Price

This is the question that separates roasters, and almost nobody asks it.

Coffee is not shelf stable in the way the packaging implies. It peaks between 7 and 21 days off roast and it should be used inside four weeks. Before about a week it is still degassing and your shots will be wild and hard to dial. Past a month it goes flat and dusty and no barista alive can rescue it.

So ask any roaster you are considering: what date was this roasted, and how often do you roast? If a bag arrives with no roast date on it, or with a best-before eighteen months out and nothing else, you have your answer. You are buying stock that has been sitting somewhere.

Then order to that. Order little and often rather than a pallet because the price per kilo looked better. Beans bought cheap and used in week six cost you more than beans bought properly, because the difference lands in the cup that decides whether somebody comes back.

• Roast date printed on every bag, not a best-before
• Store sealed, cool and dark. The dry store is perfect
• Never the fridge. It pulls in moisture and whatever else is in there
• Only open the bag you are using. Air is the enemy from the moment the seal goes

Dialling In, Which Is A Daily Job

Your grind is not a setting you find once. Humidity, temperature and the age of the beans all move it, which is why the first thing that happens every morning is a shot down the sink.

Teach your team to read the cup, because two problems cover almost everything:

On grind itself, one correction that catches nearly every new operator. Espresso is not table salt. Table salt is medium-fine, somewhere near moka pot territory. Espresso is finer than that. Think caster sugar, or fine sand that clumps slightly when you pinch it. Plunger is the other end entirely, closer to raw sugar or breadcrumbs.

Milk Is Half The Drink

In a flat white the milk is most of what is in the cup, and it gets a fraction of the attention the coffee does.

Steam to 55 to 65 degrees. That is where milk is at its sweetest, because the sugars and proteins have woken up but have not been wrecked. Above about 70 it scalds, the sweetness disappears and you are left with something flat and burnt. When a customer asks for extra hot, they are asking you to cook it. Make it anyway, and know what you are giving them.

Full cream steams best, froths easiest and tastes sweetest, because the fat carries the flavour. Skim will always drink thinner and the foam will not hold as long.

Alternative milks are not dairy with a different label. Each one behaves differently and your team needs to know how, because a split soy in front of a customer is a remake and a bad memory.

All of them want to be steamed cooler than dairy, and all of them want barista-specific blends where you can get them. Charge for them. They cost you around 50 cents a serve and absorbing that on every oat latte is a decision you are making without noticing.

Water, Because Coffee Is Mostly Water

A cup of coffee is about 98% water, and the mineral content decides how much flavour comes out of the bean.

The industry target is around 150 parts per million of dissolved solids, with 75 to 250 being workable. Too few minerals and the coffee cannot extract properly, so it drinks flat no matter what you do at the grinder. Too many and it goes heavy and chalky, and your machine scales up.

This is why the commercial answer differs from the one you would give someone at home. At home, if your tap water tastes good, it is good. In a venue you are also protecting a very expensive boiler from Perth's hardness, which is why we point people at reverse osmosis with remineralisation. Straight RO on its own strips everything and gives you that flat cup. RO plus minerals dialled back to target gives you both: no scale, and water that tastes the same in February as it does in August.

Cleaning Is A Flavour Decision

Old coffee oils go rancid. Once they have coated your group heads and your burrs, every cup carries them, and the customer does not think "that machine needs a service". They think your coffee is bitter.

• Backflush weekly with a proper detergent. Cafetto or equivalent
• Descale every 2 to 3 months, whatever your water looks like. Scale kills machines
• Clean the grinder burrs monthly. Trapped old grounds taint everything that follows
• Group heads, portafilters, baskets and steam wands daily. Wipe the wand and purge it after every single jug

What We Do For Our Cafes

When a new cafe comes to us, we have a conversation before we talk about beans. We ask about your daily target, your opening hours, your staff, your rent, your books. If the numbers don't add up, we say so. Not to lose a customer. Because we'd rather have an honest conversation now than watch another cafe close in 18 months.

We train your team on your machine. We deliver ourselves across Perth metro. We get you Brownes milk at partner rates, ordered direct from them. We answer the phone. No contracts. 30 days. If we're not right, we part as friends.

We also offer our wholesale partners free access to Launchpad, our platform for building AI tools into your business. Automated ordering reminders, marketing templates, roster optimisation.

Coffee is how we connect with people. The systems are what we're actually here for.

How To Choose A House Blend

Pick your house blend for the drink you sell most, not for the cup you enjoy most on a day off. In almost every Australian cafe that means a flat white, which means you want something that stays itself through milk rather than disappearing into it.

The second question is who is standing at the machine. A blend that only sings in the hands of your best barista is a liability on a Sunday when your best barista is not there. Forgiving beats interesting for a house blend, every time. Keep interesting for the guest slot.

Here is how ours split, as an example of the thinking rather than a menu:

That last point matters more than the blend you land on. Your team has to be able to say what it tastes like in one sentence, to somebody who does not care about coffee, without sounding like a wine list. Give them the words on day one.

Other Suppliers

• Milk. Negotiate volume. Brownes is standard in Perth. Ask your roaster about bundled pricing. Alternative milks (oat, almond, soy) add ~$0.50 per serve. Charge for them.
• Food. Keep it simple at launch. Toasties, pastries, cabinet items. Expand based on what sells, not what you hope will sell.
• Cups and packaging. WA law narrows this choice before your budget does, so read the next section before you place a single order. Start generic (BioPak and the other big suppliers all carry compliant ranges). Brand when volume justifies minimum orders. Branded cups are a marketing expense, not a necessity.
• Cleaning. Cafetto for espresso. Commercial sanitiser. Set up regular orders so you never run out.

The WA Plastics Ban

Western Australia has banned more than 20 single-use plastic items and a fair few of them are things a cafe orders without thinking. Plastic straws, stirrers, cutlery, plates, bowls and unlidded food containers went in July 2022. Plastic cold drink cups went in October 2022. Plastic coffee cups, plastic cup lids for hot and cold drinks, and disposable plastic food trays went in March 2024. Lidded plastic bowls, trays and containers, and the lids themselves, went in September 2024.

Plastic-lined paperboard cups and lids are only legal where the whole item is certified to the Australian composting standards (AS 4736 or AS 5810). Unlined paperboard lids are fine. Buy compliant from day one, ask your supplier for the certification and keep it on file. The Department of Water and Environmental Regulation enforces this, not your council, and the exemptions shift as later stages roll in, so check the current banned items list before you commit to a big order.

The coffee gets people in the door. The experience is what puts you in their head.

That's the mind share argument from Chapter 1, playing out one guest at a time. Nobody remembers a venue because the fit-out was nice. They remember it because somebody looked up, knew their order, and the whole thing took ninety seconds.

Map The Journey Before You Train Anyone

Most owners train staff on tasks. How to use the till, how to steam milk, where the cloths live. Then they wonder why the experience is inconsistent.

Do this first instead. Walk your own venue as a customer, on paper, before you hire anybody. From the moment they see the sign to the moment they are back on the footpath. Every single point where one of your people could touch them.

At each point, answer three things. How should this guest be treated here? What could go wrong here? And is there an opportunity here that we are currently walking past?

That last one is where the money is hiding. Most venues have four or five moments a day, per customer, where a sentence would have increased the order or brought them back sooner, and nobody has ever named those moments out loud. Once you have mapped them, your training writes itself, because you are no longer teaching tasks. You are teaching a journey.

Then ask the harder question. Does your model actually prioritise average order value and retention, or does it just prioritise getting the queue down? Those pull in different directions on a busy morning, and if you have never decided which one wins, your team will decide it for you, differently, every shift.

Train Them Like A Concierge

The best thing we ever did with a team was stop treating them as people who make coffee and start treating them like a concierge.

A concierge knows things. Where the nearest ATM is. Which bus stop, and which direction. How you get to the train, and how long it takes. Where the post office is and when it shuts. Where the toilets are in the shopping centre. Whether that restaurant across the road is any good.

None of that is on your menu, and all of it is why people remember you. Somebody new to the area, a tourist, a tradie on an unfamiliar job, a parent with a pram trying to work out where to go next. You answer one small question well and you stop being a shop. You become the place that helped.

Make it deliberate. Put the local knowledge in your wiki alongside the coffee specs, and ask about it when you train. It costs nothing and almost nobody does it.

The 18 Steps of Service

We developed 18 steps of service across our venues. Not a script. A framework of minimum standards, so every guest gets the same experience regardless of who's on shift and regardless of whether you're in the building.

Complaint Handling: The 6 R's

We developed this framework after years of handling complaints badly. Now every staff member knows the process. It's on the wall. It's in the wiki. It's non-negotiable.

1. Remove. Remove the problem item from the guest immediately. Don't let them sit with a bad coffee or wrong dish in front of them. Take it away.
2. Report. Inform the shift manager or owner. Even if you can handle it yourself, the person in charge needs to know. Every complaint is data.
3. Replace. Replace the item immediately. Correct order, fresh preparation, no waiting. The guest should have the right thing in their hands within 3 minutes.
4. Recover. Go beyond the replacement. A free pastry. Their coffee comped. A genuine apology. Recovery is where you turn a bad experience into loyalty. People remember how you fixed it more than the mistake itself.
5. Refund. If the guest wants a refund, give it without hesitation. Don't argue. Don't ask questions. A $5 refund costs less than a 1-star Google review that 500 people will read.
6. Rectify. Fix the root cause. Wrong grind? Recalibrate. Wrong order? Retrain on order confirmation. Slow service? Review the floor positions. Every complaint reveals a system gap. Close it.

Say The Thing Before They Have To Ask

Almost every bad review about waiting is not really about the wait. It is about not being told.

A guest who is told "that's about eight minutes, we've got a run on" will wait eight minutes happily. The same guest, told nothing, starts getting annoyed at four and has decided you are disorganised by six. Same wait. Completely different memory of it.

Same with running out. If the banana bread is gone, say so at the counter with a suggestion attached, not after they have ordered it. Every kitchen in the world calls this being 86, and the whole point of calling it out loud is that the person on the till stops selling it the second it goes. Build the habit: whoever finds the empty tray tells the front, immediately, every time.

Know When To Call For Help

Speed of service is a team sport, and it falls over when nobody is allowed to say the queue is beating them.

Set the trigger out loud so it is not a judgement call made by whoever is most stressed. Something like: three drinks behind, or four people deep at the till, or a docket sitting more than six minutes. When that line is crossed, somebody steps in from wherever they are, no permission needed and no discussion on the floor about it.

Written down, this takes thirty seconds to teach. Left unwritten, your best person quietly drowns twice a week and your slowest morning stays slow forever.

Reading Your Guests

Hospitality isn't about scripts. It's about reading people and giving them what they need.

Every one of these is a mind share moment. This is the work that gets skipped when you're buried, and it is the entire difference between a venue people use and a venue people choose.

The tradie at 6am: wants speed, strength, minimal conversation. Have their regular started before they reach the counter. "The usual, mate?" is the highest compliment you can give.

The mum with the pram: needs the door held open, space to park the pram, a smile, and someone who doesn't rush her. She's probably had 3 hours sleep. Be kind.

The business meeting: wants a quiet corner, good coffee, minimal interruption. Check once, then leave them alone.

The first-date couple: nervous energy. Don't hover. Make the coffee beautiful. Give them the nice cups.

The regular who looks off: "You alright today?" Three words that build genuine loyalty. Cafes are community spaces. Sometimes you're the only person who asks.

Perth has a hospitality staffing crisis. Everyone knows it. Nobody's solved it. Here's what actually works.

When To Hire

Start with two. You and one other. You open, they close. Or vice versa. Overlap during rush. Keep it lean until the numbers justify more.

Add a third when you can't take a day off. If you haven't had a day off in 6 weeks, you need help. Burnout kills cafes as fast as bad numbers.

Add a fourth when the queue costs you revenue. If customers are leaving because you can't serve fast enough, that's money walking out the door.

The rule: a full-time barista should pull 150+ cups per shift. If they're making 80, you've over-hired.

There is no clever answer to it. It is a genuinely narrow window and you will get it wrong both ways in your first year. What helps is seeing the cost of the roster while you're still building it, which is the whole reason the rostering section below exists.

The order matters. Replace yourself in production first (the machine), then in admin (the books), then in management (the roster). We hired another barista when what we actually needed was someone to do the ordering and the close so we could get home. Cost us a year of evenings to work that out.

How To Choose Them

Hiring the wrong person is one of the most expensive mistakes available to you, and it is one of the few you make entirely on your own.

Here is the rule we hold to, and it sounds harsh written down. Never hire someone who sits below your current lowest performer. Every hire should lift the average, not fill a gap. The moment you hire down because you are desperate for a body on a Saturday, you have told the whole team what the standard really is, and the good ones notice long before you do.

Being short-staffed for another fortnight is painful. Carrying the wrong person for eight months, then managing them out, then rehiring, costs you far more, and it costs your best people their patience.

• Make it a real process, not a chat. Same questions, same order, every candidate. You are trying to compare people, and you cannot compare two different conversations.
• Actually call the references. Almost nobody does. Ask the one question that works: "would you hire them again?" The pause before the answer tells you more than the answer.
• Hire for the attitude, train the skill. You can teach anyone to pour a flat white. You cannot teach someone to care whether the guest is having a good morning.
• Trust the first impression. First impressions are the entire product in hospitality. If they do not make one on you, in an interview, when they are trying, they will not make one on a customer at 7am on a Tuesday.
• Give them a shift on the floor. A paid trial tells you in three hours what an interview never will.

Pay Above Award

Almost every new cafe runs on casuals. So the casual rate is the number that matters, and it's the one people get wrong. A Level 2 casual is not $27 an hour. The 25% casual loading is built into the rate, which puts them at $33.85 before a single penalty applies. Offer $29 thinking you're generous and you're underpaying by nearly $5 an hour, and Fair Work will hand you the back-pay bill.

These are Restaurant Award rates. If you're inside a hotel, motel or resort you're on the Hospitality Award and the numbers are different. Rates also move every 1 July, so pull the current Fair Work pay guide for your award before each year's first pay run instead of trusting a figure you read in a guide.

$80-160 a week per person. That's the difference between award and above award. It saves you the $3-5K of recruiting, training, and lost productivity when they leave for the cafe that pays more. The maths is obvious.

What Each Person Has To Generate

Here is a number almost no operator can tell you off the top of their head, and it is the one that makes rostering a decision instead of a guess.

Work out what one hour of somebody standing on your floor has to produce in sales to be worth having. Take their loaded cost, wage plus super plus workers comp, and divide by the labour percentage you are targeting.

This is not about being ruthless with people's shifts. It is about knowing which hours are genuinely productive, which hours are an investment you are choosing to make, and which hours you have simply never looked at. Most venues find one or two hours a day that have been quietly losing money since the week they opened.

Budget For Public Holidays Before You Open

Public holidays can be some of the best trading days of your year. They can also be the days that quietly undo a month, and the difference is entirely whether you planned for them.

A casual on a public holiday is on 250% of base. That is around $67.70 an hour on the Restaurant Award, before super. Four staff across an eight-hour public holiday is roughly $2,200 in wages before you have sold a coffee.

So decide in advance, not the week before. If the day will trade, trade it properly with the right team on and a plan to make the volume work. If it will not, close, and say so on your socials early enough that nobody turns up to a locked door. What kills people is opening on a public holiday out of vague obligation, doing half a normal day, and paying penalty rates for the privilege.

Put every public holiday in your cash flow before you open. WA has eleven, and in years where Anzac Day, Christmas or Boxing Day falls on a weekend you also pick up substitute days, so 2026 actually carries thirteen. They are published years ahead. They are not a surprise, and they surprise people every single year.

Employment Hero Setup

On day one, set up Employment Hero HR and Payroll. Every new starter goes through digital onboarding: personal details, tax file declaration, super choice form, bank details, emergency contacts, Fair Work Information Statement. All digital. All stored. All compliant.

The payroll system automatically calculates award rates, penalty rates, overtime, and super. On the Restaurant Award, a full-timer gets Saturday 125%, Sunday 150%, public holidays 225%. A casual gets Saturday 150%, Sunday 150%, public holidays 250%, all measured off the base rate, not off their loaded rate. Different numbers, easy to muddle by hand, which is exactly why you let the software do it. It handles STP reporting too. It does everything that used to take hours.

Payday Super. Since 1 July 2026 super is no longer quarterly. You pay it every pay cycle, and the contribution has to be received by your employee's super fund, with enough information to allocate it to their account, within 7 business days of payday. Received, not sent. That word does the damage. If you pay through a clearing house or your payroll software, their processing time comes out of your 7 days, not on top of it, so pay super on payday and don't sit on it. Miss it and you're liable for the super guarantee charge, per pay period now, not per quarter. Employment Hero handles the mechanics, but you need to know it exists.

Use the built-in roster to build shifts with live wage cost projections. If Tuesday's roster costs $850 in wages and your projected revenue is $1,800, your wage cost is 47%.

Now, 47% is not automatically a disaster. As Chapter 3 covers, 40 to 50% is where independents actually sit in 2026, and anyone selling you 30 to 35% as the pass mark is describing a different decade. What matters is whether that Tuesday is the shape you meant it to be, and whether the rest of the week carries it.

The win is seeing the number before you publish the roster instead of after the week is gone. That's a decision you can still make. The other kind is just news.

Training

Not a half-day induction. A structured program. The goal isn't to clone yourself. It's to build a system that produces the same result whether you're standing there or not.

Everyone treats training as a cost to be minimised, then spends the year absorbing the consequences of not doing it. Slower service, remakes, mistakes on the till, the same question asked forty times, and your best person burning out because they are effectively running a school in the middle of a rush.

Start With Why

Before you teach anybody how, tell them why.

Why did you open this? Why should anybody care that it exists? There are plenty of places selling good coffee and good food, so what is yours actually for? Simon Sinek built a career on this and he is right about it: people do not commit to what you do, they commit to why you do it.

If you cannot answer that in a sentence, your team certainly cannot, and you will get exactly what you trained for. Compliance. People doing the tasks correctly and caring about none of it.

Tell them the story. How the site was found, what you were trying to build, what you are not willing to compromise on. It takes ten minutes on somebody's first day and it is the difference between staff and a team.

Get the order right. Training comes first, and the labour cost comes down after it, because the same people got faster. Not because anyone got sent home.

• Week 1: Shadow shifts. Watch, learn, ask questions. No solo work.
• Week 2: Supervised shifts. Making drinks with someone watching. Feedback after every shift.
• Week 3: Semi-independent. Can handle quiet periods solo. Supported during rush.
• Week 4: Assessment. Can they open independently? Can they handle a complaint? Do they know every menu item?
• Ongoing: Monthly check-in. What's working? What needs improvement? Any training gaps?

Upload all training materials to your Notion wiki. Use NotebookLM so staff can self-serve answers. "How do I clean the grinder?" "What's the allergen info for the banana bread?" "What's the complaint process?" Available 24/7.

Build The Orientation Once

You will onboard people constantly. In hospitality that is not a failure, it is the job. So stop treating each new starter as a one-off and build the thing once.

Make it repeatable, so it does not depend on who is on that day. Trainable, so somebody other than you can deliver it. And accessible, so a new starter can go back to it at 9pm the night before their second shift without asking anyone.

Then add to it every time you find a gap. Every question a new person asks that is not answered in there is a hole, and you will hear about each hole exactly once if you write the answer down.

Write It Down Once, Use It Forever

Create SOPs for everything. Opening, closing, dial-in, complaints, cash-up, cleaning the grinder, what to do when the machine throws an error at 6:40am.

It feels like admin you do not have time for. It is the opposite. You will spend most of your working life training people, and every process you have written down is a conversation you never have again. Six months in, when you take a day off and the place runs properly without you, that is what you will be collecting on.

Teach Them To Fix Things

The most valuable thing you can put in your team's hands is not another drink recipe. It is the ability to solve a problem without you.

Basic preventive maintenance, taught properly, saves you a fortune and a lot of ruined mornings. How to backflush and why. What a grinder sounds like before the burrs go. How to check a seal. What to do when extraction drifts. Which noise means stop and call someone, and which means adjust and carry on. Most call-outs we hear about were avoidable, and the fix was already inside the building.

Use this as your measure. If people keep bringing the same problems to you, that is not them being needy. It is a signal you have not passed the knowledge on. Every recurring question is a training gap wearing a disguise.

Brief Every Shift, Debrief Every Shift

Five minutes at the start. Five at the end. Non-negotiable, and the first thing everyone drops.

The brief sets the day. What is on, what is off, what is the focus, who is where, what is the one thing we are pushing today. The debrief closes it. What worked, what did not, any complaints, any compliments, any stock problems, one thing to fix tomorrow.

Your job in the brief is to lift them up and hand them what they need to succeed in the next eight hours. Which leads to the uncomfortable version of this:

Never Let Poor Performance Slide

This is the hardest thing on this page and the one that separates venues that get better from venues that slowly get worse.

The moment you accept a poor performer, you have set a new standard, and you set it for everybody. Your good people are watching. They see what gets tolerated, and they work out very quickly what the real bar is, which is always the lowest thing you let through.

Address it early, privately, and specifically. Not "you need to lift", but "this is what I saw, this is what it should look like, here is the support, let's look again in two weeks." Most people rise to that. The ones who do not have told you something useful.

Letting it run because you cannot face the conversation is not kindness. It costs you your best staff, who leave for somewhere that has standards.

Culture

Here is the part that decides everything else. Your team hands your guests exactly what you hand your team. Nobody has ever been made to feel welcome by somebody who does not feel it themselves. If the standards in this chapter are the technical half of the job, this is the bigger half, and it is the one you cannot delegate.

The cafes with low turnover have one thing in common: the owner gives a damn. Flexible scheduling. Meal allowance. Genuine interest in their lives outside work. Know who's studying, who's got kids, who needs school holidays off. One-size-fits-one. Treat every team member as an individual and they'll stop looking at job ads. It's not complicated. It's just caring.

Your menu is a financial document disguised as a food list. Every item on it should earn its place.

Food Cost Targets

The Dish Pricing Calculator

This is the tool we wish we'd had. Put the ingredients of one serve in, put your menu price in, and it tells you the truth.

A note on GST, because this is where people fool themselves. Your menu price includes GST, and that 10% is never yours. It belongs to the tax office and you are holding it. So every margin below is worked out on the money you actually keep, which is your price divided by 1.1. Costing against the sticker price makes every dish look about 9% healthier than it is.

Do this for every item before you print a menu, not after. The ones that do not work need a smaller portion, a cheaper input, or a higher price. There is no fourth option and no amount of hoping changes it.

Menu Engineering

Every menu item falls into one of four categories:

Stars: high profit, high popularity. Your best items. Promote them. Feature them. Never take them off.

Plowhorses: low profit, high popularity. People love them but they don't make you money. Re-engineer (smaller portion, cheaper ingredients) or increase price.

Puzzles: high profit, low popularity. Great margins but nobody orders them. Better menu positioning, staff recommendations, or a name change can help.

Dogs: low profit, low popularity. Remove them. No sentimentality. Every dog on your menu wastes prep time, ingredients, and menu real estate.

Review your menu quarterly. Pull the POS data. Which items sell? Which don't? Which have the best margins? Adjust accordingly, with no ego about the dishes you love that nobody buys.

We let a quarterly review slide into a yearly one more than once, and it always cost us the same way. A dish nobody orders still takes fridge space, prep time, a line on the board and a paragraph of training. It's never one big loss. It's a small tax you keep paying because nobody looked.

Master Ten Things Instead Of Doing Thirty Adequately

New operators build big menus out of fear. Fear that somebody will walk in, not find what they want, and leave. So everything goes on, just in case.

What a big menu actually buys you is more stock, more waste, more fridge space, more prep, longer tickets, harder training, and a kitchen that is average at everything.

Look at the places people queue for and travel across town for. Most of them do one thing exceptionally. A bakery known for one pastry. A shop known for one sandwich. Nobody makes a pilgrimage for a venue with a laminated eight-page menu.

A tight menu is faster to execute, easier to teach, cheaper to stock, less wasteful, and far more likely to produce the one dish people tell their friends about. You can always add. Taking something off after people have got used to it is much harder.

Start smaller than feels comfortable. Let demand tell you what to add, rather than your imagination telling you what people might want.

There Is A Science To The Menu Itself

The menu is not a list. It is the most-read piece of sales material you will ever produce, and most venues design it by accident.

• People do not read menus, they scan them. The top of a section and the bottom get looked at. The middle gets skimmed. Put the dishes you actually want to sell where the eye lands, not wherever they happened to get typed.
• Take the dollar signs off. A column of $ symbols reads as a price list and makes people shop on price. Write 16, not $16.00.
• Never right-align prices in a neat column. That is the single most common mistake. It invites people to run down the column and pick the cheapest thing without reading a word about the food. Put the price right after the description instead.
• Describe, do not label. "Smashed avo" is a label. Naming the bread, the feta and where it came from is a description, and description reliably sells the higher-priced version.
• Limit the choices in a section. Somewhere around six or seven items per section is where people can still decide. Beyond that they get overwhelmed and default to the thing they always order.
• Give your best margin item some room. White space around a dish, or a box, does more for its sales than putting it first.

Pricing Psychology

• Don't be the cheapest. Competing on price in hospitality is a race to zero. Your customer isn't comparing your $6.50 flat white to the one across the road. They're comparing it to making instant at home. The gap between those two experiences is massive. Price for the experience, not the ingredients.
• Charge for alternatives. Oat milk, extra shot, decaf, syrup. Each is a $0.50-1.00 add-on. These are margin boosters, not upsells.
• Keep prices clean. $6.50, not $6.47. Faster to ring up, faster to hand change, and it does not look like you ran a formula over the menu.
• But respect the round-number barrier. The gap between $9.90 and $10.00 is ten cents and it does not read like ten cents. Double figures land as a different price bracket, and that is the point where people notice and comment. $9.90 sells and nobody mentions it. $10.00 gets a remark at the counter. Same again at $20. So if a dish is landing just under a barrier, sit it just under deliberately, and if it genuinely has to go over, push it far enough over to be worth the crossing rather than nudging to $10.20.
• Review pricing every 6 months. Costs go up. Your prices should too. Absorbing every cost increase out of your own margin is a decision, even when it feels like kindness.

The Price Rise You Keep Putting Off

This is the one that keeps operators awake, so let's be straight about it.

Everyone in this industry has had that day. One comment at the counter and you spend the drive home rethinking the whole board.

But the fear and the reality point in opposite directions.

Operators who finally do it almost always report the same thing: nobody said a word, and the takings went up. Your regulars are not carrying your price list around in their head. They're carrying how the place makes them feel. Which is the mind share argument again, arriving at the till: if you're the venue they think of, a small rise is a rounding error. If you're interchangeable, any price is too high.

You don't need a marketing degree. You need 50 regulars.

This is the chapter that would have been optional in 2015 and is the whole game now. When there were more customers than good venues, they found you. There aren't, and they won't. Every venue within two kilometres is competing for the same reduced number of times a week that people have decided to go out.

So the work moves. Less about being open, more about being remembered.

And it is work. This is the hard stuff, the bit that gets skipped because you're flat out running a service. Nurturing the customers you already have. Getting out into the local market. Exposing yourself, which mostly means walking into places where nobody asked you to come and introducing yourself anyway. Building relationships that don't pay you back this week.

Nobody rosters that. Which is exactly why doing it is an advantage.

The Numbers That Matter

50 regulars at $6.50 average spend, 5 days a week, is $84,500/year. Just from regulars. Before walk-ins, weekends, food, and retail. Get 100 regulars and you've nearly got a viable business from repeat trade alone.

Not all regulars are equal. A daily long black at $5.50 is worth $1,430/year. A twice-weekly brunch couple spending $45 a visit is worth $4,680. Know who your highest-value regulars are and treat them accordingly. Your top 20% of customers probably generate 60% of your revenue.

Build The Database From Day One

This is the thing we would go back and do differently, and it costs almost nothing.

From your very first week, capture email addresses. We use Klaviyo. Any of them will do. The tool matters far less than starting early, because a list is the one asset in your business that compounds.

These are people who have already walked in, already paid you, already liked it enough to hand over an address. Sending them something is free and it works. Compare that to paying to advertise at strangers who have never heard of you, which is what most venues do instead, usually at the exact moment cash is tightest.

Give them a reason. A free coffee on their birthday, first look at a new menu, a bag of beans in a draw. Then actually use it. A list you never email is just a spreadsheet.

Know Exactly Who You Are Building For

Do the demographic research on your area and write down one core guest. Not "everyone". One person, specific enough to picture.

Then walk your venue as them and ask what they need. This is where the small stuff lives, and the small stuff is what people tell their friends about.

If your area is full of parents with young kids, that is a real design brief:

• Is there a baby change table, and is it somewhere a person can actually get to with a pram and a toddler?
• Can a pram fit through the door and park without blocking the queue?
• Is there anywhere a kid can be a kid for four minutes without a parent apologising?
• Is there a babycino on the board, and does your team offer it without being asked?

We used to keep sanitary items in the bathrooms, free, no fuss. It cost us almost nothing and it was mentioned to us more times than any piece of equipment we ever bought. Nobody else was doing it.

Different core guest, different list. Tradies want somewhere to stand, fast service and a place to put a muddy boot. Office workers want to order ahead. Retirees want somewhere to sit at 10am for an hour and be greeted by name. Pick who you are for, then build for them properly rather than being vaguely acceptable to everybody.

Tell The Story While It Is Happening

Here is the one we genuinely regret.

You only get to build this business for the first time once. The empty shell. The floor going down. The first machine landing. The first shot you ever pull. The nervous soft open. All of it is happening whether you film it or not.

Most venues post pictures of coffee and pictures of food. Everybody posts pictures of coffee and food, and it distinguishes you from nobody. What almost nobody does is bring people along for the build, and that is the content people actually care about, because it has a person in it and something at stake.

By the time you open, you want a group of people who feel like they already have a stake in the place. They watched it happen. They will walk in on day one and bring someone.

Pre-Launch (90 Days Out)

• Instagram active and posting 3 months before opening. Document the build.
• Google Business Profile set up with photos, hours, and menu before day one.
• Join every local Facebook community group in your area.
• Partner with 3-5 neighbouring businesses. Drop in sample coffees with your opening date.
• Print 500 flyers with opening date and "free coffee" offer for letterbox drop.
• Tell every human you know. Personal networks are your first 20 customers.
• Use AI to write your first 30 days of social content in one sitting. Give Claude or ChatGPT your brand voice, your menu, and your opening date. Approve, schedule, done.

Launch Week

Free coffee for 3 days. It's one of the cheaper lines in your whole startup budget and it does two jobs at once. It gets your coffee into the hands of people who would otherwise take months to try you, and it gives your team a live rehearsal at real volume before the tills matter.

Don't just pour it. Ask every single person their name and where they work. That's the list you'll be glad you started.

Increasing Average Order Value

• Suggestive sell on every order (Step 4 of the 18 steps). 20% success rate on $5 upsell = $52K/year.
• Retail at POS. Coffee bags, keep cups, bottled drinks, pastries within arm's reach.
• Meal deals. Coffee + toast for $12 instead of $5 + $8 separate. You give $1 discount, they spend $7 more.
• Loyalty program. Stamp card or digital. Every 10th coffee free costs you $1.70 and keeps a $5,000 lifetime value customer. Make it a digital program that captures who they are and what they buy, not a paper card that tells you nothing. A stamp card buys loyalty. A database buys loyalty and tells you who your best customers are.
• Seasonal specials. Limited-time items create urgency. "Turmeric latte, this month only."

The First 100 Days

A customer's first handful of visits decides whether they become a regular or forget you exist.

We lost more first-timers than we'd like to admit, and it wasn't the coffee. We treated them like transactions and assumed the product would carry it. It doesn't. Nobody comes back for a good flat white on its own, because there are four of those within walking distance.

Visit 1: Get it right. Speed, quality, a smile. The bar is low because most cafes are average. Be noticeably better.

Visit 2-3: Recognition. "Good to see you again." Even if you're guessing. Nobody gets offended by being remembered.

Visit 4-5: Start learning the order. "Flat white, right?" This is where loyalty starts. They're testing whether you're consistent.

Visit 6-10: They're a regular now. They have a spot. They come at the same time. Protect this. If their coffee is ever wrong, fix it before they mention it.

Retention

Remember names. Write them down if you have to. "Morning, Dave" hits different to "what can I get you?"

Google reviews. Get a review block on the counter, one of the little NFC and QR stands people tap with their phone while they wait for their coffee. The whole game is removing the gap between somebody feeling good about you and somebody actually writing it down, because that feeling has a shelf life of about a minute. Ask every happy regular. 200 five-star reviews outranks a competitor with 20, and it is the single cheapest thing you can do for how many new people find you.

Instagram engagement. Repost customer stories. Reply to every comment. Feature regulars (with permission).

Surprise and delight. The regular who mentions their kid's birthday gets a free babycino with a candle in it. The tradie crew who come every day get a free round on Friday. The new mum gets her coffee brought to the table without asking. These moments cost you almost nothing and they're the reason your name comes up when someone asks where to meet.

Get Out Of The Shop

Everything above happens inside your four walls. The part almost nobody does happens outside them.

Walk your block. The gym, the physio, the real estate office, the primary school, the mechanic, the office building with 40 people in it who currently drive somewhere else. Introduce yourself as the owner. Leave a bag of beans and a card. Ask what would make it easy for their team to come to you.

Do the same at the sports club, the school P&C, the local market. Take the coffee cart to the community event and don't charge for it.

None of it converts on the day. All of it compounds. Six months of that and you're not a cafe on a road, you're the local, and the local is the one people think of when they've only got two coffees out in them this week.

Two questions worth being able to answer. Are most of your customers coming back, and do you know who they are? Most venues cannot answer either one. If you can, you are already running a different business to the one next door.

Systems aren't sexy. They're the difference between a cafe that runs and one that survives on adrenaline.

A Day In The Life

The Mornings Are Yours. The Afternoons Are Borrowed.

Everything good in a cafe happens in the morning. That is when the trade is, when your team is sharp, and when a problem can still be solved before it costs you anything.

Afternoons are borrowed time. You are tired, the trade has gone, and you are running on what is left. So put the things that matter into the morning and stop pretending you will get to them at 3pm.

The clearest example is your systems. Check the POS, the card reader, the internet and the printer first thing, while support lines are open and while you still have options. There is nothing worse than discovering at 6:05am that the terminal will not talk to the bank, and finding out that priority support opens at 9. Two minutes at the start of the day, before the first customer, is not paranoia. It is the difference between a small problem and a lost morning.

Build Your Own Checklists

The lists below are ours. Yours will be different, because your building is different. So rather than copying them, understand what each one is actually for, then write yours.

An opening checklist exists to make sure the venue is ready to trade and safe to trade, and that nothing broken makes it to a customer. A closing checklist exists to make sure tomorrow starts clean, and that nothing perishable, unsafe or unlocked gets left behind.

Test every item against that. If you cannot say what a line is protecting you from, it does not belong on the list, and every pointless line makes the whole thing more likely to get ticked without being done.

Opening Checklist

• Machine on and warming (20-30 min before open)
• Grinder purge (run 5g through to clear stale grinds)
• Pull 2-3 test shots. Dial in. Adjust dose/grind until extraction is 25-30 seconds for target yield
• Check fridge temperatures (all must be below 5°C). Record in temp log
• Stock milk (minimum 20L on hand for morning rush)
• Fill coffee hoppers
• Set up POS. Count float. Open register
• Food prep: cabinet stocked, ovens on, prep containers filled
• Clean all surfaces. Floors swept. Tables wiped
• Chairs out. Signage out (A-frame). Lights on. Music on
• Check online orders, delivery status, any overnight emails
• Systems check while support is open. POS talks to the card reader, internet up, printer and bump screen connected, online ordering live. Two minutes now, or a very bad hour later
• Walk the building. Front to back, customer's eyes. Write down what needs fixing and pick one to do today

Closing Checklist

• Backflush espresso machine with Cafetto
• Clean group heads, portafilters, drip trays
• Empty and clean knock box
• Clean and sanitise milk pitchers, thermometers
• Wipe down all benches and surfaces with sanitiser
• Empty, clean, and restock display fridge/cabinet
• Sweep and mop all floors
• Empty all bins. Replace liners
• Restock cups, lids, napkins, stirrers, sugar
• Record fridge temps in closing log
• Cash up register. Reconcile with POS
• Complete daily wastage record
• Lock all doors. Set alarm if applicable
• Set up for tomorrow, do not just clean up after today. Restock before you wipe down. Fill the hoppers. Get the prep containers ready. Position the jugs. The person opening tomorrow at 5:30am is doing it in the dark on their own, and every job you leave them is a job that happens while customers are waiting
• Confirm everything synced before you walk out. POS totals reconciled, sales pushed to accounting, orders actually submitted rather than sitting in a basket. Finding out at 7am that last night's order never sent is a whole day of improvising
• 5-minute debrief: what went well, what went wrong, one thing to fix tomorrow. This is the most valuable 5 minutes of the day and it's the first thing we used to drop, because at 4:30pm you've got nothing left. Making it the last item on a laminated list is what fixed it for us. It stopped being a decision.

That is the whole philosophy of a close. Not a deep clean every night, which nobody sustains, but a venue that is genuinely ready to open. Deep cleaning has its own schedule. Readiness is nightly.

Block Your Week

The reason admin feels overwhelming is almost never the volume. It is that it all sits in one undifferentiated pile marked "later", and you carry the whole pile around every day.

Fix it by giving each job a day. Orders go on Monday. Socials on Tuesday. Admin and invoices on Wednesday. Numbers on Friday. Same day, every week, until nobody has to decide.

Two things happen. The work stops being a decision and becomes a habit, which is the only way it survives a busy week. And on Thursday you are not carrying the ordering, because ordering is Monday's problem and Monday will come round again.

Your days will be different from ours. The blocking is the point, not the schedule.

Weekly Systems

• Friday: Weekly P&L review. Revenue, costs, profit vs target.
• Friday: Stocktake. Count beans, milk, key food items. Place weekly order.
• Monday: Review Google reviews. Respond to every one.
• Monday: Post roster for the following week (via Employment Hero).
• Monthly: Deep clean. Behind fridges, inside ovens, grinder calibration.
• Monthly: Machine service check. Group seals, showerscreens, steam wand tips.
• Quarterly: Menu review. POS data, food costs, margin analysis.

Par Levels And Stocktakes

A par level is simply the amount of something you want on hand at the start of each week. Once it is written down, ordering stops being a judgement call made by whoever happens to be standing there, and becomes a job anybody can do correctly.

Set a par for every line you carry. Name who places the order and on which day. Then hold to it, because the value is entirely in the repetition.

Stocktake on the same day each week. It is nobody's favourite hour and it is the hour that tells you whether your margins are real. Without it you are guessing at your cost of goods, and a guess in that line makes every other number in your P&L a guess as well.

Waste, theft, over-portioning and creeping supplier prices all show up in a stocktake and nowhere else. You will not enjoy the first few. Do them anyway.

These checklists aren't for you. They're for the person who replaces you.

Laminate them. Put them on the wall. You set the standard once, your team executes it daily, and you spot-check weekly. That's three decisions a week instead of three hundred.

If you're personally running through the closing checklist every night six months in, you haven't built a business. You've bought a job, and it's a job that ends when your body does.

When Things Go Wrong

Machine breaks: have a backup plan. Pour overs, batch brew. Honest sign on the door. "Machine being serviced, free filter coffee today." Customers respect honesty.

Staff calls in sick: you're the backup. Always. Until you have 4+ staff.

Run out of milk at 8am: backup supplier. Know where the nearest Coles is. Keep 6L long-life in storage.

Customer complaint: the 6 R's. Remove, Report, Replace, Recover, Refund, Rectify. No exceptions.

Google review needs a response: paste it into an AI tool with your brand voice guide and let it draft a reply. Edit, post. 2 minutes instead of 20 minutes of staring at your phone wondering what to say.

Don't think about growth until you've nailed the basics. Then think about it every single day.

Grow The Business, Not The Job

There is a version of growth that makes your life worse, and it is the common one.

Revenue goes up. Hours go up faster. You add a site, or a menu, or a service, and six months later you are working more, earning about the same, and now you have two of every problem instead of one. That is not growth. That is a bigger job with your name on it.

So before any of the options below, go back to the question from the very start of this guide. Are you building something to keep, or something to sell? Growth pulls hard in different directions depending on the answer.

• Building to keep? Then growth should be measured in the profit per hour you personally work, not in turnover. A quieter venue that pays you well and runs without you beats a busy one that owns your weekends.
• Building to sell? Then growth means making yourself removable. A buyer is not paying for your talent. They are paying for a business that keeps producing after you have gone, which is documented systems, a team that stays, and books somebody else can read.

Both answers point at the same first move, which is why it is worth saying plainly. Before you add anything, take yourself out of it. Chapter 10 has the order: replace yourself in production first, then in admin, then in management. Most owners do it backwards, hire another pair of hands on the machine, and wonder why nothing changed.

When You're Ready

• Profitable for 6 consecutive months. A trend, not a fluke.
• You can take a week off and the cafe runs without you. If it can't, you don't have a business. You have a job you bought.
• Customers are being turned away during peak. Demand beyond your capacity.
• Staff want more hours. Because they want to be there, not because they need money.
• Your systems are documented. Wiki complete. Checklists laminated. Anyone can open and close.

Growth Paths

Extend hours. Cheapest lever. Same rent, same equipment, more revenue.

Increase average order value. Suggestive selling, retail at POS, meal deals, loyalty programs. See Chapter 12.

Expand the menu. Test based on demand, not hope. If 10 customers ask for smoothies, add smoothies. If zero ask for acai bowls, don't.

Catering and events. High margin, uses existing equipment. Corporate morning teas, market stalls, local events.

Retail bags. If regulars ask for beans to take home, stock your house blend on the counter.

Second location. Don't do this until location one runs completely without you. We've seen too many people open a second site and watch both suffer.

Wholesale and supply. The one people forget. If your kitchen makes something genuinely good, other venues will buy it, and you are already paying for the space, the equipment and the staff. Sandwiches into an office. Slice into the cafe with no kitchen. It uses your quiet hours instead of demanding new ones, which is exactly the opposite of a second site.

Build with AI. We offer our wholesale partners free access to Launchpad, our platform for building AI tools into your business. Automated ordering, roster optimisation, marketing templates. Tools that give you time back to focus on what matters.

Delegation is the whole of it, taken seriously earlier than feels comfortable. If you burn out you are done, and if you get ill you are done.

Knowing When Not To

Nobody writes this part down, so here it is.

Not growing is a legitimate answer. A single site, running properly, with a team that stays and an owner who takes a holiday, is a very good business and a better life than most people in this industry have. There is no rule that says the next move has to be bigger.

Say no to growth when the current site still needs you daily, when the last three months were not profitable, when you are chasing it because a competitor opened rather than because your own numbers pulled you there, or when you genuinely cannot name which line on your P&L it improves.

That last one catches the most people. If you cannot say, before you start, exactly which number this is meant to move and by how much, you are not growing. You are just getting busier and hoping.

The Cafes That Win

After supplying 200+ businesses, the ones that make it past year two share these traits: they know their numbers weekly, they're relentlessly consistent, they have a genuine reason people choose them (not a gimmick), they adapt faster than their competitors, and they ask for help.

From their roaster. From their accountant. From their peers. The operators who try to figure everything out alone usually figure out too late that a $500 conversation would have saved them $50,000.

Trial and error is expensive. Associating yourself with people who've already made the mistakes is one of the best value investments in business.

For years our ideas lived in our heads. Tasks lived on scraps of paper. Recipes were in a binder nobody updated. The menu review from three months earlier was in a text thread we couldn't find.

Information lived everywhere and nowhere.

We're the sort of people who have 30 ideas in the shower and put out suggestions and thoughts constantly. That's a good trait right up until none of it is written down, at which point it's just noise you feel guilty about.

What fixed it was a system. Not a to-do app. A brain. A single place where everything the business knows gets captured, organised and used.

What Goes In Your Second Brain

• Standard operating procedures. How to open. How to close. How to dial in. How to handle a complaint. How to onboard a new staff member. Every single process written once, used forever.
• Recipes and prep guides. Every dish, every coffee spec, every batch recipe. With weights, temps, times. Not "some milk" but "180ml at 62 degrees."
• Supplier contacts and accounts. Who you order from, account numbers, rep names, order days, minimum orders, backup suppliers.
• Staff training materials. Your service steps. Your standards. Your expectations. New hires should be able to read this and know exactly what's expected on day one.
• Meeting notes and decisions. Every team meeting, every supplier conversation, every landlord discussion. Write it down. What was agreed, who's doing what, by when.
• Ideas and improvements. That thing you thought of at 2am. The customer suggestion that was actually good. The layout change you want to try. Capture it or lose it.
• Financial targets and reviews. Weekly P&L notes, quarterly goals, annual targets. Not just the numbers. The context. Why was March slow. What did you change in April.

The Platform

We use Notion. It's free to start, it handles everything from a simple checklist to a full wiki, and it works on your phone at 5am while the machine warms up. Other options exist. Google Docs is fine. Obsidian if you're technical. The tool matters less than the habit.

What matters is structure. We recommend three core areas:

• Wiki: all your SOPs, recipes, supplier info, training docs. Permanent knowledge. Things that are true until you change them.
• Projects: things you're working on. Menu redesign. Fit-out improvements. Marketing campaigns. Each one has tasks, deadlines, and notes.
• Inbox: where ideas land before they have a home. Quick capture. Review weekly. Move to a project or archive.

Capture, Then Decide

Two systems do almost all the work here, and they are different jobs. People try to make one tool do both and end up doing neither.

A capture system catches things the second they occur to you. The idea at 5am while the machine warms up. The thing a customer said. The supplier who needs chasing. It has to be fast enough to use one-handed while you are doing something else, or you will not use it and the thought is gone. Voice memo, notes app, whatever is already in your pocket. The bar is low on purpose.

A project system is where captured things go to become work. What is being worked on, what is the next action, who owns it, when is it due. This is the one that needs a weekly look, and without it a capture system just becomes a very long list of things you feel guilty about.

The weekly review joins them. Once a week, empty the capture into the projects, decide what is actually happening this week, and bin the rest without guilt. Twenty minutes. It is the single habit that makes every other system in this guide keep working.

If you want a place to start, Getting Things Done by David Allen is where the capture-and-review idea comes from, and Building a Second Brain by Tiago Forte is the modern version of it. Neither is about hospitality. Both work.

Training With Your Wiki

Once your SOPs are documented, use NotebookLM or a similar tool to turn them into training content. Upload your wiki, your recipes, your service standards. Then let new staff ask questions and get answers straight from your own materials. Not generic hospitality advice. Your standards. Your way of doing things. Your voice.

We've started doing this with our own team. Upload the manual, and anyone can ask "how do we handle a return?" or "what's the pour weight for a flat white?" and get the exact answer from our documentation. No waiting for the manager. No guessing. No "I think it's roughly..."

Through Launchpad, we help our wholesale partners set this up for their own businesses. Upload your SOPs, your recipes, your service standards. Your team gets a 24/7 training assistant that speaks your voice. Free for anyone who buys coffee from us.

We used to find out how the month went when the accountant called.

By then it was already six weeks old. You can't fix a March that finished in March. We'd sit there being told a number we could do absolutely nothing about, and the honest reason it happened wasn't ignorance. It was that the books came at the end of an eleven hour day, and at the end of an eleven hour day they lose to almost anything.

What changed it was making the numbers take three minutes instead of an evening.

Not knowing what you are making is almost never a knowledge problem. It is a time problem wearing a knowledge problem's clothes, which is why the answer in this chapter is a three minute ritual and not a spreadsheet course.

Your Dollar Per Hour Worked

We ran venues for years without ever working this one out. It's the number that tells you the truth about your business, and it takes about ninety seconds.

This number doesn't lie. It accounts for every early morning, every Saturday, every public holiday you worked instead of claiming penalty rates. It's the number that tells you whether you own a business or bought yourself a bad job.

Track it monthly. The goal is to increase it. Not by working more hours. By working smarter hours, building systems, training your team, and increasing revenue per hour of operation.

The Five Numbers To Check Every Day

These five numbers take 3 minutes to check at close. Do it every single day. After a month, you'll see patterns. After three months, you'll make decisions based on data instead of gut feel.

Weekly Numbers Ritual

Friday afternoon. 20 minutes. Non-negotiable.

• Revenue vs target. Where are you tracking for the month?
• Food cost percentage. Pull your food purchases for the week, divide by food revenue. Should be 28-35% depending on your menu.
• Coffee cost per cup. Beans used (kg) x cost per kg, divided by cups sold. Should be $0.80-$1.20. If it's $1.50 you're over-dosing or wasting.
• Labour percentage for the week. Total wages including super and workers comp, divided by total revenue. Expect 40-50%. The target is not a number off a website, it's your own number trending the right way.
• Profit for the week. Revenue minus cost of goods minus labour minus overheads. Is this number positive? Is it growing?

The Budget That Works

Simple targets that keep you honest. Pin this to the wall next to the POS.

Add the top of every range and you're underwater. That isn't a typo, it's the whole game. The lines fight each other, and your actual job as an owner is deciding which one you hold tight this month.

And yes, the labour line is higher than most guides tell you. Chapter 3 explains why. Anyone still quoting 30 to 35% is describing a different decade, and holding yourself to it just means failing an imaginary test every Friday until you stop looking.

We have been that operator. Not because we didn't care about the numbers, but because caring about them came after a full day of service, and by then we were done.

If any line blows out, this page tells you which one within a week instead of within a quarter. That's the only real difference between the operators still trading in year three and the ones who found out too late.

This chapter is last because it's the toolkit, not because it's optional. The argument for it was made on page one.

The bottleneck is time. It was always time. Everything here exists to buy some of it back so you can put your hours into the things that actually convert to money, and into a life that isn't just the shop.

You don't need to be technical. You don't need to know what a large language model is. You need to know that right now there are tools that will save you hours a week on jobs you're currently doing badly, not because you're careless but because you're doing them at 9pm after eleven hours on your feet.

Most "best AI tools" lists are written by people who signed up for a free trial and clicked around for 20 minutes. This isn't that. We've tested everything below inside our own operation and we run over 70 AI agents across our business. What follows is the stack we'd give any cafe owner starting today.

Your AI Assistant: Pick One, Use It Daily

You need one primary AI assistant. Not five. One that you use every day until it's as natural as checking your phone.

Claude or ChatGPT. Both free to start. Both good enough. Pick whichever one clicks with you. We use Claude because the writing is sharper and it handles long, detailed tasks better. ChatGPT has a great voice mode if you want to talk to it hands-free while driving to the shop at 5am. Either one will handle your social captions, email drafts, menu descriptions, Google review replies, roster analysis, and recipe costing. The point is to use it every single day until you can't imagine running your business without it.

Your Training Assistant

NotebookLM (free, from Google). This one changes how you train staff. Upload your SOPs, your recipes, your service standards, your training docs. Then anyone on your team can ask it questions in plain English. "What's the allergen info for the banana bread?" "How do I clean the grinder?" "What's the complaint process?" It answers from your actual documents. Not generic advice. Your standards. Your way of doing things.

It also turns your documents into audio conversations. Upload your training manual and it generates a podcast-style summary your new starters can listen to before their first shift. That alone is worth the setup time.

Content and Social

Canva. Free to start. Menus, social graphics, signage, A-frame designs. Templates for everything a cafe needs. If you're paying someone to make your Instagram tiles, stop. Canva does it in 5 minutes.

AI image generation (built into Google Gemini, free). Need a social media image but don't have a photographer? Describe what you want and AI generates it. Coffee flat lays, lifestyle shots, seasonal promos. Not perfect for everything, but more than good enough for Instagram stories and daily posts.

Save Time on the Boring Stuff

Voice-to-text (built into your phone, or apps like Wispr Flow). Talk instead of type. Dictate your stocktake while walking the storeroom. Dictate emails while the machine warms up. Dictate your end-of-day notes into your second brain. Your hands are busy all day. Your voice isn't.

Loom (free). Record your screen, share a link. Show your weekend casual how to set up the display cabinet in a 3-minute video instead of explaining it for the fourth time. Show your accountant the POS issue instead of writing a paragraph about it. AI auto-generates a summary for whoever watches.

Zapier (free to start). Connects your apps together without code. New Google review? Get a notification on your phone. POS daily total? Auto-logged to a spreadsheet. Form submission on your website? Auto-email to you. If you're moving data between platforms by hand, Zapier probably does it for you.

What This Actually Looks Like

Here's a real week for a cafe owner using this stack:

• Monday morning: Ask your AI to write 5 Instagram captions for the week based on your menu specials. Approve, schedule. 10 minutes.
• Monday night: Paste your POS weekly summary into your AI. "Compare this week to last week. What stands out?" 30-second analysis.
• Wednesday: Get a 1-star Google review. Paste it in with "write a warm, professional reply." Edit, post. 2 minutes instead of 20.
• Thursday: New starter asks NotebookLM "how do we handle a customer complaint?" Gets your 6 R's framework back immediately. Without calling anyone.
• Friday: Dictate your stocktake into your phone while walking the storeroom. Voice-to-text turns it into a typed order.
• Saturday: Loom a 3-minute video showing your weekend casual how to close. They watch it before every shift.

That's 5-8 hours saved per week. No coding. No technical skills. Just using the right tools for the right jobs.

Where To Start

Don't try to set up everything at once. Here's the order:

1. Get an AI assistant. Claude or ChatGPT. Free to start. Use it every day for a month. Social captions, email replies, menu descriptions. Get comfortable with the conversation.
2. Set up NotebookLM. Upload your SOPs, your recipes, your training docs. Point your staff at it. This takes an hour and saves hundreds.
3. Start using voice-to-text. Dictate instead of type. Emails, notes, orders. Your phone already has it built in.
4. Connect one Zapier automation. Start small. New Google review sends you a notification. Or POS daily total auto-logs to a spreadsheet. One connection. Then add more.
5. Build from there. Canva for social graphics. Loom for training videos. AI images for content. Each one takes 10 minutes to learn and saves hours per week.

What We're Building For Our Partners

We've spent the last two years building AI tools for our own operation. We run over 70 AI agents that handle everything from competitor monitoring to customer retention to supply chain forecasting.

We built them because we had the same problem every operator has. Not enough hours. Too much admin. Good ideas that never got executed because the day ran us instead of the other way around.

Coffee is how we connect with people. This is where our actual focus goes: the systems and the automations, and giving operators their numbers immediately instead of six weeks late.

Now we're making that available to our wholesale partners through Launchpad. It's free for anyone who buys coffee from us. Not a generic tool. Custom-built for hospitality. Ordering reminders, marketing templates, operations dashboards, staff training AI. The systems that give you your time back.

We're giving away the ideas in this guide because ideas without execution are worthless. But the execution, the groundwork, the actual building of these systems into your business, that's where we come in. We've already done the hard part. You just need to plug in.