Feedback

What honest feedback is actually worth

12 August 2026 · 6 min read

One developer telling you where they got stuck is worth more than a thousand silent installs. Here is why, and how to ask so you get something useful back.

The install counter tells you nothing

A thousand installs and no other information is a number you can stare at for a week without learning anything. Did people open it? Did they understand the first screen? Did they quit at the sign-up, at the permission prompt, or three screens in when the thing they came for turned out to be behind a button they never found?

The counter answers none of that. One person saying "I tapped around for a minute, could not work out how to start a new list, and gave up" answers all of it.

What one honest tester actually gives you

The first-run experience, which you can no longer see

You have opened your app a thousand times. You physically cannot experience it as a stranger any more. Someone else can, once — and that once is worth more than all your own testing.

The gap between what it does and what it looks like it does

Almost every app has a feature people never find. You will not spot it, because you know it is there.

The reason behind the drop-off

Analytics tell you people leave on screen two. A person tells you why, which is the part you need in order to fix it.

A real device you do not own

Different screen, different Android version, different font size. Half of all layout bugs only exist on hardware you have never held.

Why the two obvious sources both fail

Friends and family are kind and not your users. They already know what the app is for because you told them, so they never test the thing that matters most: whether it explains itself. And they round their reaction up to be nice.

Store reviews arrive too late and say too little. By the time someone writes one, they have already decided — and most write nothing at all. The people who quit in the first thirty seconds, the ones you most need to hear from, never leave a trace.

The feedback that changes your app arrives before the person decides how they feel about it. After that point you are getting a verdict, not information.

How to ask so you get something usable

"What do you think?" produces "looks good" every time. Ask narrower questions and you get answers you can act on:

1
Give them one job. "Open it and try to create your first entry" beats "have a look around". Now their confusion has a location.
2
Ask where they stopped, not whether they liked it. "Was there a moment you did not know what to do next?" is the single most productive question you can ask.
3
Ask what they expected to happen. The gap between what they expected and what happened is your bug list, and it is usually short.
4
Say you want the harsh version. People hold back by default. Explicit permission to be blunt changes the quality of what you get more than any other single thing.

How to take it

Badly-taken feedback dries up immediately. A few rules that cost nothing:

  • Do not explain. If you find yourself saying it works if you tap the icon in the corner, you have just found the problem — they did not find the icon.
  • Write it down before you argue with it. Ideally in one list where you can see which points repeat.
  • Two people hitting the same wall is a fact, not an opinion. Fix it even if you disagree.
  • Say thank you and say what you changed. People who see their feedback used will test the next version too.

Where to find people who will tell you

Other developers are unusually good at this. They notice interface problems because they build interfaces, they can describe a failure in terms you can act on, and they know exactly how it feels to have published something nobody has opened.

That is the trade AppSwappers runs on: you install and honestly review someone else app, they do the same for yours. Not a rating swap — points are earned for taking part, never for a good score, which is the only way the feedback stays worth reading.

If you are still at the stage of getting anyone at all through the door, start with your first fifty downloads and the places developers gather.

Frequently asked questions

How much feedback do I need before it is useful?

Far less than people expect. Five people going through your app properly will surface most of the obvious problems, because the same two or three things trip nearly everyone up. Beyond about ten you mostly hear the same points again, which is its own kind of useful signal.

Why will my friends not give me useful feedback?

Because they like you and they are not your user. They read the app the way you explained it rather than the way it presents itself, and they soften anything negative to avoid hurting your feelings. Both make their reaction unrepresentative of a stranger who has three seconds of patience.

Is negative feedback bad for my app?

Feedback that arrives in private before launch is the cheapest thing you will ever get. The same problem discovered by a stranger arrives as a one-star public review instead. What hurts is not the criticism, it is finding out too late.

How do I get feedback from people who are not friends?

Developer communities and testing exchanges are the usual routes. Other developers make unusually good testers: they notice interface problems, they can describe what went wrong in terms you can act on, and they know how it feels to be on the other side of it.

Real installs from real developers

AppSwappers is an exchange: you test someone else's app, they test yours. No bots, no payments, no promises about numbers — just developers helping each other out of the same cold start.