Automation Insights

Monitoring a competitor's app the way their users see it

Priyanov 11, 20265 min read
Share
Monitoring a competitor's app the way their users see it

Most competitor monitoring starts with the App Store or Google Play listing. You check their screenshots, read their description and look at the latest release notes. But these only tell you what the company chooses to publish.

They don't tell you that the onboarding flow changed from six screens to four, that the annual subscription became cheaper in Germany, or that some users are seeing a completely different paywall. Those changes happen inside the app.

If you want to understand what a competitor is actually shipping, you need to monitor the app itself. That means opening it, navigating through important screens, recording what you see and comparing it with previous observations. Here's how to build that workflow using mobile automation and real devices.

Why the outside view lags

Three things make the app different from anything you can read on the web.

Rollouts are gradual. On iOS, a phased release goes out over 7 days: 1% of users with automatic updates on day one, 2% on day two, up to 100% on day seven. Google Play's staged rollouts work by percentage too, and users aren't notified they're in one. A new paywall can be live for a week before most people see it.

Experiments run on top of releases. Tools like Firebase A/B Testing let a team change parameters of an app that's already installed and show different variants to different users. Same version number, different screens.

Prices and offers depend on where you are. Plans, currencies, trials and promos often differ by country. A screenshot from your office tells you about one market.

Put together: there isn't one version of a competitor's app. There's the one a specific user, in a specific country, on a specific day, sees.

What's worth watching

Don't try to watch everything. Pick the few screens that tell you about their strategy:

  • Onboarding. Number of screens, what they ask for, and where sign-up sits.
  • Paywall. Plan names, prices, trial length, which plan is preselected.
  • Navigation. The tabs in the bottom bar. A new tab is usually a new bet.
  • Version. So you can line up a change with a release.

On Android, the version is one command away on a test phone. Everything else needs someone, or something, to walk through the app.

The setup

Here's the loop we'd build:

A person could do this. It's 15 minutes per country, every day, and it's exactly the kind of job that stops happening in week three. An agent on a phone does it the same way every time and writes down what it saw.

Use a real phone for this, not an emulator. Apps can check Play Integrity and treat uncertified devices differently, and a monitoring setup that sees a degraded version of the app is worse than none.

Making the data trustworthy

A few things we learned the hard way:

  • Keep prices as strings. "59,99 €" and "$59.99/yr" carry the currency and the billing period. Parse later, store exactly what was shown.
  • Expect noise. An agent can miscount a screen, and a popup can land in the wrong place. Treat a single change as "worth a look," and two days in a row as a real change.
  • One device sees one bucket. If the competitor runs an A/B test, your test account lands in one variant. To see more of them you need more accounts on more devices, and even then you're sampling, not measuring. Don't report "40% of users see the new paywall" from three phones.
  • Countries are more than an IP. Routing traffic through Germany changes the network location. The store account's country and the language setting can matter too, so keep one account and device per market if pricing is what you care about.

Stay inside the lines

This only works if you stay on the right side of the rules, and they're stricter than people assume.

  • Use accounts you're allowed to create. Sign up the way any user would, under the app's terms. Don't use someone else's login, and don't buy or share accounts.
  • Look at what any user sees. Onboarding, public screens, the paywall your own account is shown. That's the same view a customer gets.
  • Leave other people's data alone. No collecting other users' profiles, posts, messages or personal details. Under laws like the GDPR, personal data is regulated whether or not it's visible in an app.
  • Don't break things to see them. No paywall bypasses, no decompiling or poking at private APIs, no faking purchases. If the terms forbid automated access, respect that.
  • Keep the pace human. One walk-through per market per day is plenty. Hammering a competitor's backend is the fastest way to get your test accounts banned, and it's rude.

When in doubt, ask your legal team before you start, not after the first alert.

Frequently Asked Questions

Why not just read the competitor's App Store page?

Store pages are edited separately from the app and lag behind it. Phased releases, staged rollouts and in-app experiments all change what users see without touching the listing.

Can I see every A/B variant a competitor is testing?

No. Each account lands in one variant. More accounts and devices give you a better sample, but you can't see their whole experiment from the outside.

Do I need a real phone, or will an emulator do?

For apps that check device integrity, use a real phone. Otherwise you risk monitoring a version of the app that real users never see.

The takeaway

The store page shows what a competitor wants you to see. Their app shows what their users actually get, one country and one rollout bucket at a time. Pick three screens, check them daily on a real phone, and diff the results. If you want to set it up, start with one device at cloud.mobilerun.ai.

Also Read: