Which app categories invest most in a dedicated cold-start screen?

Cold-start screenshots concentrate in Education (23) and Health & Fitness (17) — the only two categories that clear N>=8 [1]. Everything else (Magazines & Newspapers 6, Weather 5, News 5, and lower) is an absolute count, not a rate [1]. Across the corpus the dedicated cold-start category spans just 58 canonical screens over 107 apps, so read this as a small, concentrated pattern [2].

Education (23) and Health & Fitness (17) hold the most cold-start screenshots — the only categories clearing N>=8 — Lazyweb Research, July 2026.

Lazyweb Research · n=107 · Published 2026-07-07

onboardingux-patternsmobileretentionsaas

The category picture

Raw cold-start screenshots mapped to the dedicated cold-start canonical category, by app-store category [1]:

CategoryCold-start screenshotsReliable?
Education23Yes (N>=8)
Health & Fitness17Yes (N>=8)
Magazines & Newspapers6No (N<8)
Weather5No (N<8)
News5No (N<8)
Food & Drink4No (N<8)
Productivity3No (N<8)
Entertainment3No (N<8)
Books3No (N<8)

Education and Health & Fitness together hold 40 of the 374 raw cold-start screenshots.

Why these categories

Both leading categories depend on establishing a day-one habit, which is exactly what a dedicated cold-start screen sets up: a learning app frames a first lesson, a fitness app frames a first goal or workout. Named cold-start apps in these spaces include betterme, fitbod, fiton, apple-fitness, calm, and headspace, alongside content apps like nyt and blinkist [3]. The pattern is about routine formation, not just filling emptiness.

How to apply it

If you build an education or fitness app, a purpose-built cold-start screen is peer-validated — model it on betterme or fitbod. In categories below N=8 the data can't establish a convention; a bespoke cold-start screen there is a differentiator you should justify independently. Because the whole family is small (58 canonical screens), benchmark against named examples rather than a category rate [2].

Caveats

Only Education (23) and Health & Fitness (17) clear N>=8; all other categories are absolute counts, never percentages [1]. The cold-start family (58 canonical screens, 374 raw shots, 107 apps) is the smallest in the corpus, so no percentage headline is appropriate [2]. Screenshot counts reflect captured screens, not runtime frequency [1].

The numbers

StatComputed from
Cold-start by category: Education 23, Health & Fitness 17, Magazines 6, Weather 5, News 5, Food & Drink 4, Productivity 3, Entertainment 3, Books 3cold_start_by_category full list; denominator 374 raw cold-start screenshots; only Education/H&F N>=8
58 cold-start canonical screens, 374 raw screenshots, 107 appscold_start_canonical: canonical_screens 58, raw_screenshots 374, companies 107
Cold-start apps: betterme, fitbod, fiton, apple-fitness, calm, headspace, nyt, blinkistqualitative named_examples cold_start
Methodology. Universe is 374 raw cold-start screenshots from 107 mobile apps (July 2026), split by app-store category; only categories with N>=8 are treated as reliable. The family is below n=50 overall, so no percentage headline is used.

Sources & citations

  1. [1] Lazyweb Research analysis of 374 cold-start screenshots (mobile-app screenshot corpus), July 2026. Raw cold-start screenshots by app-store category; only Education and Health & Fitness clear N>=8.
  2. [2] Lazyweb Research analysis of 23,407 canonical screens (mobile-app screenshot corpus), July 2026. Cold-start canonical category: 58 screens, 374 raw screenshots, 107 companies.
  3. [3] Lazyweb Research analysis of 809 tracked apps (mobile-app screenshot corpus), July 2026. Named cold-start apps in education and fitness categories.

Source: Lazyweb Research — proprietary analysis of real, in-market app screens. Cite as Lazyweb Research, 2026-07-07.

Related questions

Explore the underlying screens, flows, and A/B tests inside Lazyweb. More research