Digital menus and direct ordering: how a Mexican restaurant stops giving away its margin
Delivery apps charge commissions that eat much of the profit per dish. A direct channel doesn't replace them — it balances them. Here's what to build, in what order, and what to expect from each piece.
Delivery platforms solved a real problem for restaurants: immediate demand with no marketing spend. The cost of that convenience is commissions that in México typically land in the 20% to 30% range per order, plus the fact that the customer isn't yours — they belong to the app.
The conclusion isn't 'leave the apps'. For most restaurants that would be suicidal. The conclusion is that an order arriving through your own channel leaves several times more margin than the same order through the app, and that building that channel today costs a fraction of what it did five years ago.
Start with the menu, not with ordering
The classic mistake is starting with the shopping cart. The digital menu is the piece that pays back fastest and carries the least friction:
- A PDF menu is the worst possible format: heavy, unreadable on mobile, and Google can't index the dishes.
- A real HTML menu gets indexed: when someone searches 'pozole in Playa del Carmen', your dishes can show up.
- Updating prices stops being a redesign. In high season that matters.
- The table QR code points to a living page, not to a file you uploaded eight months ago.
Your own ordering channel
Once the menu exists and gets indexed, direct ordering is a short step. The minimum viable version is simpler than it looks:
- A cart on the menu you already have — no separate catalogue to maintain twice.
- Online payment or cash on delivery. Starting with cash on delivery only is a valid choice and removes the biggest technical friction point.
- Orders arrive to the kitchen over WhatsApp, not to a dashboard nobody will open mid-rush. This detail decides whether the system gets used or abandoned.
- Automatic confirmation to the customer with an estimated time.
That confirmation and follow-up flow is exactly the pattern we describe in process automation — a restaurant is just one particular use case.
Reservations: only if you have a waitlist
A reservation system is useful when demand exceeds tables. If that's not your situation, it adds complexity and returns nothing. A simple form with WhatsApp confirmation covers most independent restaurants.
Without local SEO, none of this gets seen
A restaurant lives on searches with immediate intent and a short radius: 'breakfast near me', 'seafood open now'. That's won in the Local Pack, not in traditional organic results. Your Google listing, your reviews and consistent name, address and phone matter more than any on-site trick.
The order we recommend
| Phase | What you build | Why now |
|---|---|---|
| 1 | Indexable digital menu + up-to-date Google listing | This is what makes you appear |
| 2 | WhatsApp with after-hours auto-replies | Recovers the orders you lose at night today |
| 3 | Direct ordering with cash on delivery | Margin without technical friction |
| 4 | Online payment | Only when volume justifies it |
| 5 | Reservations | Only if there's a waitlist |
If you'd rather see it assembled than build it piece by piece, this is how we deliver it in restaurant websites, and the search side is detailed in local SEO and Google Maps.