FlowsBook a callSign inCreate free accountSign up
Workflows by industry

How delivery apps recruit restaurants in a new city before launch day

To recruit restaurants for a delivery app in a new city, map the supply first, pick launch districts, then work a ranked list of independents with field visits.

Part of the guide Workflow automation

By the xMap team · · 5 min read

The fastest way to recruit restaurants for a delivery app in a new city is to treat it as a ranked list, not a door-to-door sweep. Map every restaurant in the districts you will launch in, keep the well-rated independents that customers already like, check which cuisines each district is missing, remove the ones you have already signed, and send the merchant team out with a daily route of the best ten to fifteen. Launch with fewer, better restaurants in a few districts rather than a thin spread across the whole city.

This article is for city launch managers, merchant acquisition teams and founders of delivery and quick commerce apps. It covers how to choose where to recruit first, what makes a restaurant worth a visit, a made-up launch in Lima, the mistakes that slow a launch, and an AI employee that keeps the list fresh while the team signs.

Choose the launch districts before the restaurants

Customers judge an app in the first week by whether they find something they want, close enough to arrive warm. That depends on density of choice in their own district, not on how many restaurants the app has across the whole city. So the first decision is which three to six districts to open, and only then which restaurants inside them.

Good launch districts usually have many restaurants close together, a mix of cuisines, and roads a rider can cross quickly. A district where the restaurants are spread along one long road is harder to serve well than one where they cluster in a few blocks. Count restaurants per district, look at how they are spread on a map, and favour the compact ones.

Big cities have more restaurants than any team can visit. xMap's places database lists 2,125 restaurants within 10 km of the centre of Toronto and 1,640 within 10 km of the centre of Sydney, November 2026. Of the 290 nearest Toronto's centre, 249 are independent; of the 303 nearest Sydney's, 293 are. Those are the restaurants a new app is most likely to sign, because large chains often negotiate deals centrally.

What makes a restaurant worth a visit

  • Loved already. A rating of 4.5 or more with many reviews means people choose it on purpose. These are the names that make customers download an app.
  • Fills a gap. A district with six burger places and no good vegetarian kitchen needs the vegetarian kitchen more than a seventh burger.
  • Kitchen that travels. Some food survives a twenty-minute ride and some does not. A quick AI read of the menu and category flags the obvious problems.
  • Just opened. A restaurant in its first months is looking for customers and has fewer habits to change.
  • Not already yours. Remove every restaurant you have signed or that told you no in the last few months.

A made-up launch in Lima

All the figures here are invented. A delivery app plans to open in Lima in eight weeks with four launch districts and a merchant team of three. It wants 60 restaurants signed by launch, at least 12 per district, with no district missing pizza, chicken, ceviche or a vegetarian option.

In the first week, the flow searches the four districts and returns several hundred restaurants. It keeps independents rated 4.3 or more with a phone or a website, reads each one to tag its cuisine, and marks the cuisines each district lacks. One district turns out to have strong ceviche and chicken but only one pizza place, so pizza restaurants there get a higher score. The team gets a route of twelve restaurants per person each morning, ordered so nobody crosses the city twice.

By week five, two districts have reached their target and the third is close. The fourth is behind because its best restaurants are already busy and slow to answer. The launch manager decides to delay that district by two weeks rather than fill it with weaker names. The daily list for the remaining three weeks focuses there, and newly opened places in all four districts keep arriving each Monday.

Mistakes that slow a city launch

  • Recruiting city-wide from day one. Riders and restaurants spread too thin, and the first orders arrive cold.
  • Chasing the highest-rated names only. A district also needs everyday food at everyday prices.
  • Leaving the gap check until the end. If a district has no breakfast option, you want to know in week one, not week seven.
  • Visiting during service. Lunch and dinner are the worst times to talk to an owner. Plan visits for mid-morning and mid-afternoon.
  • Assuming a restaurant without a delivery mention on its website does not deliver. It may already be on another app. Treat it as a hint and ask.

An AI employee for merchant acquisition

In xMap Flows the recruitment list is an AI employee that runs every weekday morning until launch and weekly after it. It is a good example of what workflow automation does for a launch team: the boring sorting happens before anyone arrives at work.

  1. Schedule: weekdays at 7 in the morning, Lima time.
  2. Area of interest: the launch districts drawn on the map.
  3. Nearby places: restaurant types, independents only, rated 4 or more, with a phone or website.
  4. Match with my table: keep only restaurants not yet in your merchant table.
  5. xMap LLM: tag the cuisine, note whether the food travels, and score each one against the district's gaps.
  6. Sort and limit: the best scores first, a sensible number per rep.
  7. Split by territory and Route planner: a compact patch and a visiting order for each rep.
  8. Email: each rep's route, with the reason to visit each restaurant.

What lands in each rep's inbox is a numbered route: restaurant, district, cuisine, rating, the gap it fills and a directions link for each stop. The launch manager gets the same list with a line per district saying how many restaurants remain unsigned and which cuisines are still missing.

To start from a ready flow, try Well-loved restaurants that likely do not deliver yet, New restaurants to sign up the week they open, Cuisines missing in each district and Field sales routes to sign up new restaurants.

Questions

How many restaurants should a delivery app sign before launch?

There is no single number. Set a target per launch district, enough that a customer opening the app finds choice in the main cuisines, and launch districts as they reach it rather than the whole city at once.

Can the flow tell which restaurants already deliver with another app?

Not reliably. It can read a restaurant's website for mentions of delivery, which is a useful hint, but the only way to know is to ask the owner.

Should we approach chains or independents first?

Usually independents for the field team, because they decide locally. Chains are often handled by a separate partnerships team through their head office.

What happens after launch?

Switch the flow to weekly and keep Only new on, so the team hears about each restaurant that opens in the districts, the week it appears.

Templates that do this

Read next

More on workflows by industry · Every article · Templates

Put it to work

An AI employee in xMap Flows does this on live data, on its own timetable. Start from a template or describe it in plain words.