Built for cooking apps

Your users browse recipes. Most never cook one. The data knows why — we don't, yet.

Cooking apps spike on day 1 and drop fast. "Try this recipe!" emails push browsing, not cooking — and the behavior that predicts who actually subscribes is usually something different from both. The 48-hour PAI analysis tests it on your cohort, not a category benchmark.

Book a 48-hour analysis → No commitment · No setup before the call
Browse-to-cook gap
~10×
recipes viewed per one cooked
Trial-to-paid avg
3–8%
subscription category typical
Behaviors we'll test
15–25
against your conversion
Time to insight
48h
from data access
◆ The honest framing

We won't tell you "cooking apps need a meal plan in week 1." Category averages don't run your business.

Every cooking-app blog says the same things — push the meal plan, drive the shopping list, ping the user at dinnertime. Maybe one of those is your milestone. Maybe none of them is. The only way to know is to test them against your subscription data. That's what the 48-hour analysis does. The output is a milestone — yours, specifically, with confidence bounds. If the analysis finds no strong signal, we tell you that too.

✕
We won't quote category benchmarks at you
Your cohort, your milestone. Whatever the typical cooking app does is irrelevant once we run the analysis on yours.
✕
We won't pretend we already know your answer
Anyone selling you a "cooking-app activation framework" before seeing your data is guessing. We'd rather earn the answer.
✕
We won't promise lift before we see results
If the analysis surfaces no strong milestone in your data, we tell you. You keep the report. We part as friends.
Why guessing fails

The activation milestone most cooking teams pick. Usually wrong.

These are the milestones cooking teams default to because they sound right. None of these has been tested against your conversion data. Without that test, you're driving users toward a target you invented.

Myth 01 · The save one
"Save your first recipe"
Saving feels like commitment. It isn't. Most savers never cook — it measures aspirational browsing, not behavior change.
Myth 02 · The browse one
"Browse 5+ recipes in week 1"
Catches the recipe-window-shoppers — they generate page views, not subscriptions. Engagement metric, not retention metric.
Myth 03 · The setup one
"Pick dietary preferences"
Onboarding completion masquerading as activation. Users who configure preferences aren't the users who pay.
For your app

We don't guess. We test these against your conversion.

For a cooking app, we typically test 15–25 candidate behaviors against your trial-to-paid or 12-week subscription outcome. The winner is whatever has real lift + coverage in your cohort. Examples we test:

Completion
First recipe cooked
marked done · within 7d
Depth
Recipes cooked in week 1
≥1 / ≥2 / ≥3 completed
Intent
Shopping list created
≥1 list with items added
Planning
Meal plan built
week ahead scheduled
Proof-of-cook
Photo / review submitted
cook actually happened
Personalization
Dietary preferences set
diet · allergens · cuisine
Pantry
Ingredients logged
≥5 items in pantry
Habit
Notification opt-in + open
dinner reminder followed
Social
Share / follow / save
social loop engaged
Return
D2 / D7 return visit
at least one return session
Voice / video
Hands-free mode used
cooking-mode launched
+ 4–14 more
Whatever's in your data
behaviors unique to your product
We test all of them in parallel against trial-to-paid (or 12-week retention). The winner is the milestone — different for every product, never a formula, never a category guess.
Methodology proof · not category transfer

Here's what the method found for one edtech app.

We don't have a cooking case study yet — and we won't pretend we do. What we can show is the same analysis on a different vertical. The method transfers. The milestone won't be the same — your cooking app's signal will be specific to your cohort.

For Typesy (edtech), one early behavior retained users at 39.5% vs a 34% baseline — a 5.5pp lift on n=24,118 trials. The winning behavior was "complete ≥1 course within 4 days." Your cooking app's milestone will not be that. It will be a different behavior in your data — and we don't know which one until we run the analysis.

Candidate signal Retention Decision
Key action B in 14 days 37.9% 81% cov · drop
Key action A in 14 days 36.1% 82% cov · drop
Complete ≥1 course within 4 days 39.5% 79% cov · primary

n = 24,118 · 12-week window · p < 0.001 · baseline 34.0%

Whatever your cooking app's milestone is, it will surface the same way: candidate behaviors scored against subscription conversion, winner picked on lift + coverage + significance. Read the full case study →

12-week retention · Typesy · edtech
34%
Before
→
39.5%
After
Typesy Typing app · eReflect · method, not category transfer
After we find it

Every trial user gets a state. Every day.

Once the milestone is set, each user is classified daily based on where they are relative to it. The state determines which email they get — and whether anything goes at all.

NEW_USERdays 0–1
FIRST_RECIPE_SAVEDday 1–3
FIRST_RECIPE_COOKEDmilestone hit
BROWSING_NOT_COOKINGdays 3–7
HIGH_INTENTpaid early
CHURN_RISKno session 5+ days
DORMANT_3Dsilent 3 days
DORMANT_7Dsilent 7 days
TRIAL_EXPIREDpast trial end
How it works

Three steps. One milestone.

No new SDK, no schema changes, no rebuild. We read from the behavioral data you already capture — and you stay in control of every send.

01 · Connect
Amplituderecipes · saves · cooks
Mixpanelsessions · plans
Firebasepantry · shopping
Read your existing cooking events
Recipe views, saves, cooks, meal plans, shopping lists — whatever you already track. Nothing added to your stack. Read-only DB access is the entire integration.
02 · Analyze
recipe_save62%
browse_5plus71%
recipe_cook92%
meal_plan48%
shop_list55%
Find your activation milestone
We score every candidate cooking behavior against your subscription conversion. The winner becomes the milestone. 48 hours from data access to the report — your milestone, your data, your call.
03 · Drive + approve
Cooked first recipe · day 3sent
Saved but didn't cook · d5staged
Meal plan builtsent
No session 5dqueued
State-matched emails, your approval
Each user state gets its own template — drafted by us, approved once by you, then personalized at send time. Launches in monitoring mode. Nothing leaves the queue until you flip the switch.

Want the full 5-step detail? How it works →

FAQ for cooking founders

Objections we've heard from cooking teams.

We're freemium with a paid tier, not trial-based. Does this work?
Yes. We swap "trial-to-paid" for "free-to-paid" or "free-to-W12-retained" — whatever outcome you care about. The method is the same: candidate behaviors scored against the conversion you want to lift.
We lean heavily on push notifications — does email even work for us?
Yes — and we can drive push instead of email, or both. The output of Sendlyr is a state machine: the channel is configurable. If push is your dominant channel, the state-matched messages route there.
Our content is seasonal — pumpkin in October, grilling in summer. Breaks the analysis?
No. We bucket cohorts by signup month and compute matched-cohort conversion, not raw averages. Seasonality affects content engagement; it doesn't break the milestone signal.
Our users are mostly mobile-only. Can you reach them?
Yes. The state machine drives whatever channel you already have — email, push, in-app messages, SMS. Mobile-only cohorts work the same as cross-platform ones; the analysis runs on the events, not the surface.
We have multiple subscription tiers. Which do you optimize for?
Yours to pick. We optimize for the conversion you point us at — usually entry-tier paid conversion, but we can target upsell to higher tiers or annual retention. We test the milestone against whichever outcome matters most.
Our recipes are user-generated. Does the analysis care about content quality?
Not directly. We test user behaviors, not content metadata. UGC vs editorial content doesn't change the method — the milestone is about what users do, not which recipes they engage with.
What's the minimum cohort size?
Practical minimum is ~10,000 trial / free users in the last 90 days. Below that the analysis can't reach statistical significance and we'd be guessing. We'll tell you on the call if you're under-scoped.
What if the analysis finds no strong milestone?
It happens. If no behavior shows meaningful lift and coverage, we tell you. You keep the report and the methodology. We part as friends. No obligation to continue.
◆ Get started

Find your cooking app's activation milestone.

We don't know what it is — yet. Neither do you. The 48-hour analysis tests every candidate behavior against your conversion and tells both of us. No commitment. Free for qualified teams.

Book a 48-hour analysis →
30-min call No setup before 48-hour turnaround