Anonymous case · Consumer food app

100K+ users for a food-tech app in 6 months

A food-tech app with a good product but stalled growth. Eighteen thousand active users, over €3 for every new user, few coming back after thirty days. In six months: 100,000+ active users, cost per new user halved, users staying twice as long. Zero magic: a growth engine that makes paid advertising, store visibility, refer-a-friend and user retention over time work together.

100K+
Active users · 6 months
−50%
Cost per new user
+100%
Users returning at 30 days
6 months
Duration of the work
Food-tech app dashboard screen with install and retention metrics - growth marketing case study

Want to receive articles like this in your inbox?

One food-tech case at a time: the move, the number that came out of it, and what you can copy on Monday.

Updated on 17 May 2026
Reading time 13 min
Matteo Coloru
Written by Matteo Coloru

The starting context

In short: a food-tech app with a good product, 18,000 active users, growth stalled for four months. The founder had budget to grow but no method to use it. The key question: "How do I go from 18K to 100K without burning cash?".

The app was born from two partners with solid technical foundations and a clear product vision: a food-tech platform focused on a precise niche, with a well-crafted user experience and the first signs that the product really met its market. The product worked. The problem was getting it known and downloaded.

In the previous twelve months, the app had reached 18,000 monthly active users through a mix of word of mouth, a few mentions in trade outlets and an in-house Facebook advertising campaign with a modest budget. Over the last four months, however, growth had flattened. Every new user cost more and more, users didn't stay, and the two partners were close to the exhaustion of those trying to do everything alone.

The brief was clear: "We have an investment round closing, we need to show the project can scale in 6 months. We need 100K active users and a cost per new customer below the category average." In other words: what was needed was a growth engine, not a campaign.

The diagnosis: 3 bottlenecks

Direct answer: three knots were blocking growth. Acquisition ran on a single channel (Meta only), store visibility was neglected (30-40% of free installs left on the table), user retention wasn't measured by entry group (impossible to tell what really worked). Without untying these three knots, growing would only have multiplied the waste.

The first month was diagnosis only. Only data for a month. Zero new campaigns, zero surface tweaks. After connecting the measurement tools (one to understand where each user came from, one to read what they did inside the app), the picture emerged clearly.

Bottleneck Symptom Estimated impact
All on Meta 92% of budget on a single channel, an audience increasingly tired of seeing the same ads Cost per new user rising 6% every month
Neglected store listing Generic listing, weak images, zero work on search keywords 30-40% of free installs left on the table
Unsegmented user retention Only the overall active-user figure, no reading by entry group Product decisions made on instinct, not on numbers

The fourth knot, which emerged in the second month, was subtler: the app had no refer-a-friend system inside the product. Satisfied users wanted to recommend it, but had to do it by hand. No nudge to word of mouth, no reward, no way to track who brought whom. A drawer full of potential growth, locked.

The growth engine in 4 pillars

What the growth engine is: a system where paid advertising, store visibility, refer-a-friend and user retention feed each other. Every user who arrives from advertising improves the store listing (more reviews), every review lowers the cost of free users, every group of users who stay produces new invitations, every invitation lowers the average cost per new customer. Four pillars in a circle, not four separate compartments.

The most frequent mistake when growing an app is treating the four channels as separate compartments. You run paid campaigns, tend the store listing, launch refer-a-friend, watch who stays. But if the circle doesn't close, each channel delivers 60% of what it could.

The growth engine we designed followed this logic:

  1. Paid advertising brings in on-target users and brings reviews that improve the store listing.
  2. The store listing makes more of those who see it download, lowers the real cost per user and brings free installs.
  3. Refer-a-friend turns the users who stay into a very low-cost acquisition channel.
  4. User retention shows which customers are worth the most, so you can tune the campaigns' creatives and audience to attract similar ones.

Each of the four pillars had its own numbers, a clear owner and a review every week. Zero "let's go by feel". Clear decisions on clear numbers.

In practice: we moved the budget from 92% Meta to 55% Meta + 25% TikTok + 12% Google app campaigns + 8% App Store advertising. But the real lever was ad production: from 4 to over 25 active creatives, rotated every week. Cost per new user fell from €3.20 to €1.60 in 90 days.

When you grow an app with advertising, what makes the difference isn't the budget, it's the speed at which you produce new ads. The platforms (Meta, TikTok, Google) reward fresh ads with a lower cost. An ad left running too many weeks tires the audience and makes the cost shoot up.

We redesigned ad production with a system that, in 12 weeks, took the active library from 4 to over 25 creatives:

  • Creator videos: 8 small creators to shoot authentic videos, 2 new ads every week.
  • Quick variants: small changes to the ads that were working (a different opening image, the first 3 seconds, the call to action).
  • Orderly paired tests: every new ad tried for 7 days in a dedicated group, a clear decision at the end of the period.
  • Ads by area: creatives tailored to areas with relevant differences in language or habits.

On the channel side, diversification was gradual but decisive:

Channel Initial budget share Budget share at 6 months Cost per final user
Advertising on Meta92%55%1.80 €
Advertising on TikTok0%25%1.20 €
Google app campaigns5%12%2.10 €
Advertising on the App Store3%8%1.40 €

TikTok turned out to be the pleasant surprise. The lowest cost per user of all, but above all a younger audience with a 7-day return 18% higher than Meta's. The figure changed the strategy: it isn't enough to look at how much it costs to bring in a user; what counts is how much they are worth over time depending on the channel they come from.

Store visibility: the underrated channel

The real numbers: in 90 days of orderly work on the store listing, the app went from position 28 to 6 for the category's main search keyword. Free installs tripled (from 1,200 to 3,800 a month). The real cost per user fell by 22% thanks to the growing weight of free installs in the total.

Store visibility is the most underrated channel in app marketing. Here's an image: it's like having a shop on the main street and leaving the window dirty. Everyone who walks past carries straight on.

The work on the store listing touched six levers, all measured and tweaked every two weeks:

  1. Search keywords: over 40 relevant keywords, ranked by how many people search for them and how hard it is to climb, split into 3 priorities.
  2. Title and subtitle: rewritten with the main keyword and a clear promise. Zero marketing phrases, words a user would really search for.
  3. Images redone: new screens with a line above each, focused on the benefit to the user, not on the technical feature.
  4. Preview video: 30 seconds showing the app really being used, not an inflated trailer.
  5. Curated reviews: an in-app message asking the most satisfied users for a review, a public reply to every negative review within 24 hours.
  6. More downloads from the listing: monthly test on images and text to get more of the listing's visitors to download.
The store listing works day and night without asking you for another euro. Neglecting it means paying for traffic you could have had for free. Matteo Coloru

The leverage effect was strong: by the end of the work, 38% of monthly installs came for free (against 8% at the start). In money terms, every free install is an install you didn't pay for. At 100K, we're talking about roughly €60,000 a month of advertising saved.

Refer-a-friend: word of mouth you can calculate

Concrete result: refer-a-friend, launched in month 3, brought 18% of new installs over the following three months. Each user who invited brought on average 2.3 new users. From this channel, the cost per new customer was €0.70, against the overall average of €1.60. Almost half.

Refer-a-friend is the most powerful lever for a consumer app, but also the easiest to get wrong. The typical mistake is copying the competitor without understanding the psychological and practical spring that makes it work.

We designed a double-reward system (for both the person inviting and the person accepting) with four levers:

  • Double reward: €5 to the person inviting when the friend uses the app for the first time, €5 to the friend as a welcome. Amount set on margin, not on the fear of missing the chance.
  • Easy sharing: personal link created automatically, to send in two taps on WhatsApp, Telegram or SMS.
  • Right moment: the invitation appears after a positive moment in the app (for example after a successful action, never cold).
  • Reliable tracking: the link recognises who brought whom on iPhone and Android, even when the app is downloaded starting from scratch.

Operational lesson

Refer-a-friend only works if people already like the product and come back to it. Launching it earlier is a shortcut that burns the rewards budget without creating solid growth. First you make users stay, then you switch on refer-a-friend.

Users who stay: the real metric

What we changed: from the overall active-user figure alone to a weekly check of how many return after 1, 7, 30 and 90 days, split by acquisition channel and user type. The 30-day return went from 14% to 28% in six months, doubling the value of each user.

User retention multiplies every euro invested in acquisition. An app where 14% remains at 30 days has to bring in twice as many users as one at 28% to get the same real usage. Yet most young apps don't measure who comes back group by group. They look at the overall figure and fool themselves that all is well.

We introduced a retention check on three levels:

  1. By week of arrival: users grouped by the week they downloaded, return measured at 1, 7, 30 and 90 days.
  2. By channel: same scheme, but split by source of arrival (Meta, TikTok, store listing, refer-a-friend).
  3. By behaviour: users grouped by the key action taken in the first 3 days (for example finishing the first steps, using the main feature).

The most important discovery came from the third level: those who finished the first steps within the first 24 hours came back at 30 days almost three times as often as those who didn't. Immediate product change: first steps cut from 9 to 4, completion up from 42% to 71%, average 30-day return up 9 points in 6 weeks.

Results at 6 months

Verified final numbers: from 18,000 to 108,500 active users a month (+503%), cost per new user from €3.20 to €1.60 (−50%), 30-day return from 14% to 28% (+100%), share of free installs from 8% to 38%, customer recommendation index from 32 to 51. Investment round closed in month 7, with the growth numbers confirmed.
Metric Month 0 Month 6 Change
Monthly active users18,000108,500+503%
Cost per new user3.20 €1.60 €−50%
Users returning at 30 days14%28%+100%
Free installs share8%38%+375%
Customer recommendation index3251+19 points
Paid channel distribution92% Meta4 balanced channelsspread

The most underrated figure is the +19 points on the recommendation index, meaning how many customers would recommend the app. It isn't a vanity number: such a marked rise while pushing on acquisition is rare. The reason is clear: the work on first steps, on keeping users and on refer-a-friend improved the experience people actually had, not just the volumes.

Repeatable lessons for anyone growing an app

5 operating principles: don't grow before the product meets its market, spread the paid channels before they saturate, the store listing is a channel and not a detail, those who stay are measured group by group and not in bulk, refer-a-friend multiplies but doesn't replace advertising.

1. Don't grow before the product meets its market

If less than 10% of users come back at 30 days, pushing on acquisition is financial suicide. First you work on the product and the first steps, then you open the advertising tap.

2. Spread the channels before they saturate

Don't wait for the cost on Meta to become unsustainable before opening TikTok or Google app campaigns. Distribution is built when things are going well, not when they're going badly.

3. The store listing is a channel, not a detail

Treating store visibility as "something to do when we have time" means paying two or three times as much for installs you could have had for free. It needs dedicated people, work every week, its own numbers.

4. Those who stay are measured group by group, not in bulk

The overall active-user figure hides the truth. Only looking at who comes back for each channel and each behaviour tells you what works. Without this data, every decision is a gamble.

5. Refer-a-friend multiplies

Refer-a-friend only works if users stay. Launching it on an app where people don't come back burns the rewards budget. First you fix user retention, then you switch on refer-a-friend.

Useful external resource: to dig deeper into industry averages on user retention and cost per new user, AppsFlyer publishes every year the Performance Index and the Mobile App Marketing Report, with data split by category and by region.

Frequently asked questions on how to grow a food-tech app

How much should you invest to acquire a user for a food-tech app?

The cost per new user for a food-tech app in Italy ranges between €1.80 and €4.50 depending on the segment (home delivery, recipes, nutrition, reviews). In the case described we started from €3.20 and brought the average cost down to €1.60 in six months thanks to fresher creatives, store visibility and new channels on top of paid advertising alone.

What is store visibility and why is it critical for a food-tech app?

Store visibility (ASO, App Store Optimisation) is the work on the app listing on the App Store and Google Play that makes it findable by those searching for it. For food-tech it is worth 30-40% of free installs. The levers are: title, subtitle, search keywords, images, preview video, reviews, the percentage of people who download after viewing the listing. Neglecting it means paying 2-3 times as much in advertising.

What is cohort retention and how is it measured?

It measures how many users who arrived in the same period keep using the app after a given number of days. The standard windows are 1, 7, 30 and 90 days. For food-tech, a 30-day return above 20% is considered good. In the case described we took it from 14% to 28% in six months.

How much budget do you need to launch an effective refer-a-friend programme?

Refer-a-friend starts with almost no product costs (building the mechanism and tracking) but needs a budget for rewards. For a food-tech app, a reward between €3 and €10 for the person inviting and €3 and €10 for the person accepting typically lowers the cost per new customer by 40-60% compared with paid advertising alone.

How long does it take to bring an app to 100K users?

It depends on the starting point and on how well the product really meets its market. In the case described, starting from 18,000 users, we reached 100K in 6 months with an integrated growth engine. If the product hasn't yet found its market, growing is impossible: you burn money on acquisition while users don't stay.

Meta or TikTok to grow a food-tech app?

Both, in a balanced mix. Meta remains the main channel for volume (a more mature audience, precise targeting), but TikTok brings a lower cost per user and a younger audience that comes back well. In the case described, the final mix was 55% Meta + 25% TikTok + 12% Google + 8% App Store advertising.

Case study · Want a similar result?

Thirty minutes, free

Did you find something useful on this page and want to understand how to apply it to your food brand? Write to me on WhatsApp. You tell me where you are, and you walk away with something useful. I reply myself, in person, within the day.

Message me on WhatsApp
Matteo Coloru available on WhatsApp - Manager card