Understanding When to Use Web2App and How to Do It Well — Thomas Petit, Independent App Consultant

Blinkist buys paid content placements on Taboola and Outbrain (ads next to news content) and sends them to a web page that looks like an article, matching what the reader expects after clicking, rather than its full web onboarding. Traffic from Facebook gets a completely different web experience. The guest copied these tactics in 2016.

Noom's web onboarding is about 50 screens. It works because weight loss users have high intent and can be qualified with a long run of questions; a quick-value app like Photoroom should instead show the aha moment fast.

Photoroom offers at least 10 of its app features as free one-off tools on its website with no login. Users get value immediately, remember the app and learn it has more features, so they come back to the app when they need more. A long-game acquisition channel that doesn't need instant conversion.

Send much (not all) of the Instagram and TikTok ad traffic to a web survey. By about question six the answers reveal the persona and how likely they are to pay, so the funnel reports high-value users back to the TikTok and Instagram algorithms within a short window, helping them find more people like that. Payment still happens in the iOS app, not on the web; tracking from web to install to payment stays fuzzy.

For a premium couples therapy app whose ads weren't catchy and whose topic is heavy, the team started users in a softer, fun flow (a minigame or playful quiz) that people complete easily, then moved them gradually toward the real product and qualified them. They first considered building a separate simple game app as a lead source for the main app, then built the flow on the web instead because minigames are flexible to build there. Fun ads for the quiz are cheaper to get clicks on.

Ads ask 'What's your IQ?' or 'Do you have a higher IQ than your neighbor?' and lead to a web quiz/minigame with easy questions, ending with a result like 'you're in the top 5%, let's train you to get to the 20%'. Users are then sent to the app with its minigames, where monetization happens. No payment on the web: the flow is about engagement and warming users up before the store.

An app used by many small teams and SMBs faced customers who wanted to buy for the team but couldn't through in-app purchase: they needed company invoices and to go through HR's billing process. Web billing handles group purchases and invoices in the company's name, which IAP cannot. Price mattered less than billing ($10, $50 or $100).

For subscribers billed on the web, the cancel flow stays easy (a clear confirm button), but one screen before confirmation shows a counteroffer: a refund of half of what they already paid while keeping the subscription active. Only possible when you own the transaction; not doable on the App Store.

8fit, Blinkist and Babbel, big Berlin clients of Outbrain and Taboola, tried the networks' new app product that sent clicks straight to the store. They made creatives more and more obviously about an app (screenshots, 'download the app' text) to prepare users, but store-to-install conversion stayed very low. Readers of native ads don't expect to land on a store.

Once app-install campaigns plateau, add Meta campaigns that go to a web flow, with adjusted creative (different call to action and wording). Meta stops showing install ads to people who never click them (e.g. after ~50 unclicked store ads) and only reached ~2M of a 20M eligible audience over six months at five-digit daily spend; a different conversion signal makes it show ads to different people. Keep both running; teams at scale never go 100% web or 100% app. Not worth it before you hit limits on the simpler route.

Google App campaigns bundle Search with YouTube and display and cover both OSes. Buying web inventory instead lets you separate brand, competitor and generic search keywords and show each a different selling point, control budget, pick devices, block YouTube channels and use audiences unavailable in App campaigns. One client runs YouTube ads to a web flow alongside Google App (UAC) campaigns; they attract a slightly different audience on different channels and times.

Older Android versions correlate with lower subscription rates, while recent high-end phones (Pixel, Samsung, Xiaomi at $800-1,000+) convert about like iPhones. Google App campaigns can't target by OS version, but web campaigns (e.g. YouTube) can: run e.g. Android 10+ only, or separate campaigns bidding X on recent flagship devices and Y on the rest. Meta supports this targeting natively. The landing page also sees device details, so traffic can be split there.

One client's app gets meaningful revenue from the Amazon Appstore; there are about 50M active Amazon tablet users in the US alone and many in Latin America. Many ad networks can't send to the Amazon Appstore (or the option is broken), so a web landing step lets you route those users to the right store.

A gaming company sends ad traffic to a very simple landing page before the store, which lets it do attribution differently and explain the game with full-screen graphics the store doesn't allow.

SKAdNetwork required conversion signals within 24 hours. Subscription apps were fine (most subscriptions start on day zero), but games usually monetize on day 7-30. Many game companies redesigned the day-zero experience to generate signals that qualify users' propensity to pay later (not to force day-one payment) and sent them back to Facebook.

Don't pick web or in-app billing wholesale. Web checkout loses the App Store's trust and one-tap purchase, and web processors lack the local methods stores support (MB WAY in Portugal, Bizum in Spain, carrier billing in Argentina, prepaid cards). Test by segment: send users in countries where you can't offer their preferred method to IAP despite the fee; keep users who can pay with Apple Pay on the web there. High-intent traffic (brand search, big promotions) loses less conversion on the web.