What three platforms actually cost a kitchen
Every delivery marketplace gives a merchant its own device and its own merchant portal. Each one is complete and sensible on its own terms. The trouble is that a restaurant is not three restaurants. One kitchen, one set of ingredients, one payroll, one KRA obligation — but four places where a sale can be recorded, and only one of them is yours.
The cost shows up in three specific places. The close-of-day figure becomes an addition exercise done by a tired person at 11pm, which means it is sometimes wrong and nobody can tell when. The stock count drifts, because the counter draws down inventory and the tablets do not. And the question an owner most needs answered — is this platform worth its commission? — cannot be answered at all, because the revenue it generated and the stock it consumed live in different systems.
None of that is a technology problem in the interesting sense. It is a record-keeping problem, and the fix is to decide that there is one place where a sale is recorded, regardless of which door the customer came through.
Want to see it running in your own business first?
Sign UpHow a kitchen consolidates three platforms
The order of operations matters here, because the temptation is to start with the integration and work backwards. Start with the record instead.
- 1
Decide which system holds the truth
One system has to be the place where a sale is a sale. For most kitchens that is the POS, because it is already the only system that sees counter trade, cash, M-Pesa and card. The marketplaces then become channels that feed it, rather than four parallel books.
- 2
Get the platform orders into it as sales, tagged by channel
An Uber Eats order and a walk-in order are both the sale of food. Recording them as the same kind of object — with a channel tag that says which one it was — is what makes a single daily figure possible without losing the split.
- 3
Point every channel at one stock position
This is the step that fixes the stock count. If the counter and all three platforms draw from the same inventory, the count reflects everything sold. If they do not, the count reflects counter trade and the difference is guesswork.
- 4
Put the ticket where the kitchen can see it
A kitchen that watches three screens misses orders on at least one of them during a rush. Bringing the tickets onto one kitchen display, in the order they arrived, is a preparation fix rather than an accounting one, but it is usually the change staff notice first.
- 5
Reconcile each payout against the sales record, weekly
Each marketplace pays out net of commission on its own schedule. Once your own system holds what each order was worth gross, comparing a payout to it becomes a five-minute check instead of an investigation. Do it weekly, while you can still remember the week.
- 6
Only then compare the platforms
With the sales recorded and the stock drawn down from one position, you can finally see revenue per platform against the food it consumed. That is the number that decides whether a commission rate is acceptable, and it is the whole reason for doing the preceding five steps.
Where multi-platform kitchens usually go wrong
Treating the tablet totals as the books
A marketplace portal reports what that marketplace did. It cannot report what the business did, and it has no reason to. Using three portals as a management account means never seeing the business whole.
Expecting the POS to manage the menu
Your Uber Eats listing, its pricing and its promotions live in Uber Eats. A POS that records the resulting sale is not the same thing as a POS that publishes your menu, and any vendor blurring the two is worth a direct question. Veira does not push menus or prices out to a delivery platform.
Assuming consolidation means fewer devices on day one
The platforms still own the customer-facing app and the rider. The device may well stay. What changes is that the order it receives also exists in your own records, which is the part that was missing.
Comparing platforms on revenue alone
The platform with the highest revenue is often not the most profitable one, because commission rates differ and so does the product mix each one sells. Revenue per platform is half an answer; revenue against food cost and commission is the answer.
Leaving reconciliation until month end
A payout that does not match the sales record is a question about one specific week. Asked four weeks later, it is an archaeology project, and most owners quietly stop asking.
A worked example: a Kilimani kitchen on three platforms
Take a mid-sized Kilimani restaurant doing roughly KES 1.8m a month, split about 55% counter and 45% delivery across Uber Eats, Glovo and Bolt Food. Before consolidating, the close-of-day routine was a till Z-report plus three tablet screenshots, added by hand into a WhatsApp message to the owner.
Two things were invisible in that routine. The first was a recurring KES 40,000-odd monthly gap between what the tablets showed and what hit the bank, which turned out to be a mix of commission, promotional contributions the restaurant had agreed to months earlier, and a handful of refunded orders nobody had recorded. The second was that one platform was selling a disproportionate share of a low-margin combo, so its headline revenue looked healthy while its contribution did not.
Neither of those was discovered by better maths on the same inputs. They were discovered by having the orders recorded as sales in one place, with the stock drawn down from one position, so the gross value of what each platform sold could be set against both the payout received and the food it consumed. The fix to the combo was a menu change made in each platform portal — which is where menus are managed — informed by a number the restaurant did not previously have.
Trading without eTIMS-compliant tax invoices risks KRA penalties, blocked VAT input claims for your customers, and receipts a business buyer cannot expense.
Veira signs every sale to KRA eTIMS automatically, so each receipt is compliant the moment it prints, with no separate device to reconcile.
How Veira handles a kitchen on several platforms
Veira integrates with Uber Eats, Glovo and Bolt Food, and the shape of all three is the same: the order is recorded as a sale in Veira rather than living only on the platform device, it draws stock from the same position the counter draws from, and it appears in the daily report tagged to its channel so you can see the platforms apart and together.
What that gives a kitchen running all three is one close-of-day figure, a stock count that accounts for delivery volume, and a per-channel revenue breakdown to set against each commission rate. Orders can be shown on the Veira kitchen display alongside counter tickets, and because each sale is a sale in your own system, it issues a compliant KRA eTIMS invoice like any other.
What Veira does not do is worth being equally clear about. It does not manage your listings, pricing or promotions on any platform — those stay in the platform portals. It does not assign or track riders. And it does not reconcile a payout for you: it holds the sales record that makes reconciling one accurate and quick. If a vendor tells you otherwise about their own product, ask to see it done on live data before you sign.
Frequently asked questions
Can one POS really handle Uber Eats, Glovo and Bolt Food together?
Will I be able to stop using the delivery tablets?
Can I see each platform separately as well as combined?
Does this tell me which delivery platform is most profitable?
How do delivery orders affect my eTIMS obligations?
What happens to delivery orders when the internet drops?
Do I need a separate plan or add-on for delivery?
A kitchen on three delivery platforms does not need a fourth system. It needs the three it has to report into one place, so the day closes once and the commission question can finally be answered with a number. If that is the position you are in, book a demo and bring last week's payout statements — that is the conversation worth having.
Need help, or want to talk it through first? Chat with us on WhatsApp