The Soft Launch: Opening a Café Quietly on Purpose
2 October 2026 · MidaOne
The fit-out finished late, the signage is up, and the pressure to open properly is enormous — you are paying rent on an empty room. So most first-time owners do the grand opening: an announcement, a full menu, free coffee, a photographer. And they find out in front of two hundred strangers that their slowest drink takes four minutes, that two people cannot pass each other behind that counter, and that the card terminal sits on the wrong side of the till. Every one of those is cheap to discover in front of thirty forgiving people and expensive to discover in front of reviewers.
A soft launch is a test, not a quiet opening
The distinction matters, because most soft launches fail at exactly this point. Opening quietly and hoping it goes well is not a soft launch — it is just a bad opening with less traffic. A soft launch is a period with a written list of things you do not yet know, and a way of finding out each one. If you cannot say what you are testing this week, you are not soft-launching. You are trading badly on purpose.
Two weeks is the usual shape, and there is a practical reason for it: it is long enough to see both weekdays and both weekend days twice, which is the minimum needed to tell a pattern from an incident. A single busy Saturday tells you nothing. The same jam in the same place on two Saturdays is a fact you can act on.
What has to be in place before you serve anybody
A soft launch is trading. Serving a paying customer, or giving away a drink to the public, is operating a food business, and the fact that you called it a trial does not change what you need to have in hand. Owners sometimes assume a quiet fortnight is a grey area before the paperwork lands. It is not.
Food safety is regulated at emirate level, and the authorities, portal names and sequence differ between them, so confirm your own position with your own municipality rather than with a checklist. The shape is broadly consistent, though: the trade licence comes first, and it does not by itself permit you to handle food. In Dubai, a food establishment needs a permit for food-related activities from the municipality's food safety department, and the premises are inspected before that permit is issued — so the inspection sits before you trade, not after. Opening a café in Dubai covers the sequence, and what a food safety inspection looks at covers the visit itself.
Get written confirmation of where you stand before you set a date. The staff side needs to be real too — people working your soft launch are working, with whatever your contracts and visas require already in place. The soft launch is the moment to test your throughput, not your licensing.
What two weeks should actually be testing
Pick four or five things, write them on the back of the till, and give each one a way of being answered. These are the ones that repay the effort for nearly every café:
| What you are testing | How you find out | What you change if it fails |
|---|---|---|
| Can you serve the rush? | Time the queue and the slowest drink through two mornings | The layout, the prep done before open, or one piece of equipment |
| Does the menu earn its length? | Look at what actually sold, by item, after a week | Cut the bottom of the list before regulars adopt it |
| Can the team work together? | Watch one full rush without stepping in | Who stands where, and who owns the till |
| Does the till match reality? | Close and count every single day from day one | Button layout, modifiers, and the closing routine itself |
| Is the space right? | Sit in it at the hour you expect to be busiest | Seating, sound, and the queue's path to the door |
The throughput one is the hardest to do honestly, because the instinct in a jam is to step behind the counter and fix it yourself. Resist it for one rush and take notes instead — the bottleneck is usually one step, not general slowness, and finding your café's real peak-hour bottleneck is a method rather than a guess. The menu one is almost as valuable: a fortnight of real sales by item is a better guide to what to cut than any amount of pre-opening conviction, which is where menu engineering starts to be useful.
Who to invite, and in what order
Sequence your audience by how much they will forgive, starting with the most forgiving. Each wave tells you something the previous one could not.
- Your own people, and their families. Half a day, full service, no money. This is a rehearsal — the point is to find the obvious breakages with nobody's goodwill at stake.
- The neighbours. The shops either side, the offices upstairs, the building's security and maintenance team. They are your weekday morning trade for the next five years and they will tell you the truth because they expect to come back.
- Friends of staff, and a few regulars from wherever your barista worked before. They will order properly, pay properly and complain properly — which is the first honest test of the till and the prices.
- Walk-ins, on shortened hours. Door open, no announcement, closing early. This is the first time the queue builds without you knowing who is in it.
- Full hours, still no announcement. If this week holds, you are ready to actually open.
Two warnings. Do not make the whole thing free: a free coffee tells you nothing about whether people will pay your prices, and it teaches the neighbourhood a price that is not yours. A proper discount is fine; free for two weeks is a focus group, not a test. And keep the announcement genuinely off until the end — once the opening post goes out, you are no longer testing, you are performing, and your Google Business Profile will start collecting reviews of a café that is not finished yet.
The numbers to write down from day one
The soft launch is the only period where you have no history at all, which makes every figure you capture disproportionately valuable later. Four of them are worth the discipline:
- Covers and sales by hour. Not a daily total — by hour. Your staffing for the next year comes out of this, and no amount of later averaging recovers it.
- Sales by item, including the ones that sold once. The item nobody ordered in two weeks is carrying stock, prep time and menu space for nothing.
- Everything you gave away. Comps, remakes, staff drinks and the ones that went in the bin. Ring them through, even in week one — especially in week one, because your first cost picture is otherwise fiction. Your policy on this should be decided before you open: see staff drinks and comps.
- The daily close, every day, from the first day. Count, record, explain the difference. A café that starts counting in month three never finds out what normal looks like.
The till discipline behind all four is the thing to install now rather than later. Every sale rung up, every giveaway recorded, every shift closed and counted — it is far easier to start with that habit than to introduce it to a team that has spent a month without it, which is why training staff on the POS belongs in the rehearsal week rather than after opening.
When to stop soft-launching
Decide the exit conditions before you start, or the soft launch will either end too early, because you ran out of nerve about the rent, or drift on for two months because it never feels ready. Write down three or four conditions, and hold yourself to them.
Reasonable ones for most cafés: the peak you expect is served without a queue you would walk away from; every member of staff has worked a full rush on the till on their own; two consecutive daily closes have come out with a difference you can explain; and you know which items you are cutting. If those are true, announce it. If they are not, fix the specific one that is failing — not the whole opening — and give it another week. From the morning you open properly, the clock starts on the first stock count and the first tax dates, and a new café's first 90 days picks up where this leaves off.
Where MidaOne fits
A soft launch is the cheapest time to get the till right, and a 14-day free trial with no card required happens to be the same length as the fortnight you are testing in. MidaOne runs on the phones, tablets and laptops you already own, so you are not waiting on hardware to start taking test orders — just a standard 80mm thermal printer if you want receipts. The reports that matter in these two weeks are the ones you will look at daily: sales by hour, sales by item, sales by employee, waste by item, and shift closes showing expected against counted with the difference. Stock updates live as you sell, so your first stocktake has something to be compared against, and VAT is applied per sale at 5% from the first transaction rather than reconstructed afterwards. It stays at a flat AED 200 a month, so the system does not become a bigger decision than the menu.
Test your till in the fortnight it matters most. Free for 14 days, no card required.
Start your free trialFrequently asked questions
How long should a café soft launch last?
Around two weeks for most cafés, because it covers two of every weekday and two weekends — the minimum needed to tell a repeating pattern from a one-off bad shift. Set exit conditions before you start rather than a fixed date, so the decision to open properly is based on what you observed.
Do I need my licences in place for a soft launch?
Yes. Serving the public is trading, whatever you call it, and food safety requirements are set at emirate level with the premises typically inspected before a food permit is issued. Confirm your own position in writing with your municipality before you set a date, because the sequence and the authority names differ between emirates.
Should a café soft launch be free or discounted?
Discounted, not free. A free drink tells you nothing about whether people will pay your prices, and it anchors the neighbourhood to a number that is not yours. Charge, discount honestly, and watch what sells at a price you can actually sustain.
What should I measure during a soft launch?
Sales by hour rather than daily totals, sales by item including the ones that barely sold, everything you gave away or remade, and a counted daily close with the difference explained. These are the only two weeks where you have no history, so anything you fail to record is gone for good.
When is a café ready to open properly?
When you can serve your expected peak without a queue you would personally walk away from, every member of staff has worked a rush on the till alone, consecutive closes balance to a difference you can explain, and you know which menu items you are cutting. Announce after that, not before.
The reason to open quietly is not caution. It is that the first two weeks are the only free information you will ever get about your own café, and a grand opening spends them on applause instead. Make a list of what you do not know, open the door to the neighbours, write down what happens, and let the announcement wait for a room that is ready for it.