In short
- Online sales = traffic × conversion × average order value. The three numbers multiply, so small gains compound fast.
- Conversion is the cheapest lever — you improve it without spending a single złoty on ads.
- Average order value goes up with modifiers, combos and a free-delivery threshold set just above your current basket.
- Add traffic last, once the page already sells. Otherwise you are paying for visits that never turn into orders.
Most restaurants look for growth in one place: “we need more people on the site.” That is the most expensive route. Online sales are a product of three numbers, which is exactly why modest gains in three places add up to more than 50%.
+20%
traffic
+15%
conversion
+10%
order value
+52%
total sales
The math that does all the work
Take a pizzeria with 4,000 monthly visits to its ordering page, a 3% conversion rate and an average basket of 70 PLN. That is 8,400 PLN a month. Now watch what happens when you move each number on its own — and all three together.
| Scenario | Traffic | Conversion | Basket | Monthly sales |
|---|---|---|---|---|
| Starting point | 4,000 | 3.0% | 70 PLN | 8,400 PLN |
| Traffic only (+20%) | 4,800 | 3.0% | 70 PLN | 10,080 PLN |
| Conversion only (+15%) | 4,000 | 3.45% | 70 PLN | 9,660 PLN |
| All three levers | 4,800 | 3.45% | 77 PLN | 12,761 PLN |
Month 1: conversion, or growth for free
Conversion is the share of visitors who actually place an order. In hospitality it is usually lost on the same five things.
- Steps to payment. From tapping a dish to confirming the order there should be at most three screens. Forcing account registration before checkout is the most expensive form in the industry.
- Mobile loading time. Over 70% of orders come from phones. Camera-resolution photos can push menu loading into several seconds — and half the visitors will not wait.
- Dish photos. Items without a photo sell noticeably worse than items with one. You do not need all of them — start with your 20 bestsellers.
- Visible costs. Delivery fees and minimum order value revealed only at checkout are a classic reason for abandonment. Show them as soon as the address is entered.
- Payment methods. BLIK, card and instant transfer. In Poland, missing BLIK costs real orders, especially among customers under 35.
Test it yourself: place an order from your phone as a brand-new customer, and count the taps and the seconds. If it takes more than 90 seconds, there is a double-digit percentage waiting to be recovered.
Month 2: average order value
Same customer, same acquisition cost, bigger bill. Four mechanics that work in practically every venue:
- 1
Modifiers instead of separate menu items
Extra cheese, double meat, premium sauce, larger size. Every modifier adds a few złoty of margin with no extra service work.
- 2
Combos with a concrete benefit
“Pizza + drink + sauce, 6 PLN off” beats a percentage discount because the customer sees an actual amount. A combo should raise the bill, not replace it with a cheaper version.
- 3
A free-delivery threshold just above the basket
If your average basket is 70 PLN, set the threshold at 85–90 PLN. Too high and it discourages, too low and you give margin away for nothing.
- 4
One well-aimed upsell in the cart
A drink or a dessert offered once, on the last screen. Three prompts across three screens annoy people and lower conversion.
Month 3: traffic that pays for itself
Only now does adding visits make sense. Order matters: free and repeatable sources first, paid ones after.
- 1Google Business Profile. Accurate hours, menu, an ordering link in the profile, fresh photos every month, replies to every review. For local hospitality this is usually the single strongest traffic source.
- 2Your own customer base. Collect marketing consent at checkout, then text people who have not ordered in 30 days on a Friday afternoon. Costs pennies, converts better than any ad.
- 3A QR code in the delivery bag. A customer who ordered through an aggregator today can order directly tomorrow. A card with a code and a concrete incentive (“free drink when you order on our site”) costs a few groszy.
- 4Local paid campaigns. Switch them on last and cap them to your real delivery radius. Any earlier and you are paying for traffic the site cannot yet convert.
Four numbers to check every Monday
| Metric | Where it comes from | Direction |
|---|---|---|
| Menu visits | Ordering page analytics | up |
| Conversion | Orders ÷ visits | up |
| Average basket | Revenue ÷ order count | up |
| Share of direct orders | Own orders ÷ all orders | up |
Four numbers, one screen, five minutes a week. If one of them flatlines for three weeks, you know exactly which lever to pull in the next sprint.
Restaurants rarely run out of marketing ideas. What they lack is knowing which of the three numbers is holding them back right now.
Want to see these numbers for your venue?
On a demo we will show you the dashboard with conversion, average basket and share of direct orders — using your own menu.
Book a demoFrequently asked questions
Is 50% growth realistic for every restaurant?
It is realistic for venues starting from an unpolished ordering flow — which is most of them. If your checkout is already optimised, conversion is high and the basket is tuned, you are left mainly with the traffic lever: slower growth, but cheaper to sustain.
Where do I start if I only have time for one thing?
Shorten the mobile ordering flow and add photos to your 20 bestselling items. Those are the two changes with the best effort-to-effect ratio.
How long before I see results?
Conversion and basket changes show up in the data within 2–3 weeks. Work on your Google profile and customer base compounds over 2–3 months.
About the author
Adrian Tański
CEO & Founder, Dinevo
10 years in tech, previously CTO at a delivery startup. Writes about online sales and local marketing.