- Market
- Saudi Arabia
- Written for
- Commercial Director, Head of Digital
- Decision
- Decide whether the brand has earned an app, and build one worth installing
Information security for software delivery

ISO/IEC 27001
ISO/IEC 27001 certified
The decision
When the install is earned and when the browser still wins
The question is not whether an app can be built — it always can — but whether the habit exists that makes a customer keep it. The comparison is between what each channel asks of the customer and what it gives the brand back.
| Option | What it asks of the customer | Speed of a repeat order | Reach of a new offer | What it costs the brand to keep | Best for |
|---|---|---|---|---|---|
| Mobile web ordering | One tap on a link — no download, no account until checkout. | Fast enough for occasional guests; logins and addresses fade between visits. | Anywhere a link travels — menus, packaging, messages, social profiles. | One codebase, no store releases, changes live in minutes. | Every brand at the start, and many brands permanently. |
| A branded app | A download and home-screen space, granted once and revocable. | The fastest possible — saved identity, saved order, one confirmation. | Push notifications to people who opted in — strong, and easily worn out. | Two platforms, store review cycles, and a running upkeep budget. | Brands whose regulars already reorder often enough to want the shortcut. |
| Web first, app when earned | Nothing at first; the install is invited once the habit shows. | The web channel proves the reorder pattern the app will serve. | Links recruit widely; push is reserved for the committed. | The app budget is spent on evidence rather than on hope. | Most growing operators — the sequence that lets the data decide. |
The flow
Hungry people count taps, and the app is judged in them
Nobody browses a restaurant app the way they browse a shop. The customer arrives hungry, usually knowing what they want, often reordering what they had last time. The design brief is therefore the opposite of engagement: get out of the way. The last order sits on the opening screen ready to repeat, addresses and payment are remembered, and the distance from opening the app to a confirmed order is counted in taps and defended in every release. Every screen added between hunger and confirmation is a place to lose the order to a delivery aggregator whose icon sits one swipe away.
Arabic is where that discipline shows or fails. An interface converted from English reads like a translation at exactly the moment a hungry customer has no patience for one: dish names that sound like captions, numerals that switch style mid-screen, layouts mirrored by machinery rather than designed for the direction they run in. We author the Arabic surface first — the language most of the market orders in — and treat the English as the parallel text, which is the honest weighting for this market rather than the inverted one most ordering software ships with.
Loyalty belongs inside that same flow, not in a tab beside it. Points that appear on the payment screen as they are earned, a reward that offers itself at checkout when it applies — these cost the customer nothing to use and are the actual reason the app deserves its place on the phone. A loyalty programme bolted on later, with its own screens and its own login, asks the hungriest user in retail to do paperwork, and they decline.
After launch
The app keeps spending money after the launch party
The build is the visible cost. These are the drivers of the running one, and a brand should see them priced before the first line of code — because an app a brand cannot afford to maintain is worse than no app at all.
| Variable | Effect on price | Why |
|---|---|---|
| Operating systems and store rules | A floor of upkeep that exists whether or not you change anything | The platforms move on a schedule you do not control: new OS versions, new device shapes, new store review policies. An app left unmaintained does not stand still; it decays towards rejection and crashes. |
| Menu and offer velocity | Decides how much of the app must be editable without a release | A menu that changes weekly cannot wait for store review. The build must put menus, prices, offers and imagery behind server-side content the operation edits itself, or every ramadan menu becomes an engineering ticket. |
| Push notification discipline | Sets how long the channel survives its own marketing | Push is the app's scarcest asset and the easiest to burn. The running cost here is editorial: someone deciding what is worth a notification, because the customer who mutes you has taken back the channel forever. |
| Payments, loyalty and the plumbing beneath | Integration upkeep that scales with the systems the app touches | Wallet providers update their SDKs, the POS changes an interface, the loyalty ledger needs its month-end reconciliation. Each integration the app depends on is a small ongoing commitment, priced honestly at the start. |
Our role
The first release we build and the habit the brand sustains
- 01
The case, tested first
Before designing a screen we examine the reorder behaviour in your direct channel, and if it does not yet justify an install we say so — the app budget waits for the evidence, not the other way round.
- 02
The ordering core
The tap-counted flow, remembered orders and addresses, local payment methods and loyalty earning inside the transaction — built on the same menu model that feeds your other channels, so nothing drifts.
- 03
The operable app
Menus, offers and imagery editable by the operation without a store release, and a maintenance scope agreed in writing, so the running cost is a line item rather than a surprise.
What the app promises regulars — the offers, the tone, the moments worth a notification — is the brand's to run; we build the flow that keeps the promise short and the upkeep that keeps the app alive.
Common questions
What buyers ask
How do we know our restaurant brand has earned an app?+
Look at the direct channel you already run. If a meaningful group of customers reorders through your web ordering — same people, repeating, without being chased — the shortcut an app offers has an audience, and the install invitation will land on habit rather than on hope. If direct reordering is thin, the app will not create the habit; it will sit uninstalled while the budget that built it could have improved the web channel that recruits customers in the first place. The sequence that works is web first, app when the data shows the habit — and we will show you the reading before we propose the build.
Can loyalty be bolted onto the app later, or does it belong in the first release?+
The mechanics can be phased; the architecture cannot. If the first release is built with a customer identity, an order history and a payment step designed to carry earning, then points, tiers and rewards can arrive in a later release without surgery. What fails is the true bolt-on: a separate loyalty product with its own screens and its own login stitched beside the ordering flow, which asks a hungry customer to do administration and gets declined. Our practice is to build the loyalty spine into release one even when the programme itself launches later — the cost is small at that stage and prohibitive afterwards.
What does a restaurant app cost to run after launch?+
It scales with four drivers rather than a flat figure: the platform upkeep the app stores impose on everyone, the velocity of your menu and offers, the number of integrations the app depends on, and the editorial effort of using push responsibly. A brand that changes its menu weekly and connects the app to POS, payments and loyalty carries a genuinely different running cost from one with a stable menu and a simple flow — which is why we scope the upkeep in writing before the build, not after. The one running cost that is always a mistake is zero: an unmaintained app decays in public, one OS release at a time.
In this cluster
Next step
Test the case for your app
A working session that examines your reorder behaviour and either scopes a first release or says plainly that the browser still wins.
Test the case for your app
What happens next
- 1We reply within one business day.
- 2A 30-minute scoping call with a senior consultant.
- 3A written proposal with scope, team, and timeline.
Or reach us directly
