I looked at AppsFlyer properly for the first time about six months ago. Read the docs, watched the intro video, got on a demo call. The sales rep was nice. The product looked impressive.
Then I asked about pricing. "It depends on your install volume, but typically starts around $500 per month." For a company doing £500 in monthly revenue, that's not a tool. That's a liability.
What enterprise attribution tools actually cost
AppsFlyer and Branch are the two dominant players in mobile attribution. They're genuinely good products for what they do. They're also built for companies running six-figure ad campaigns who need sophisticated postback systems, fraud detection, incrementality measurement, and 47 different dashboard views.
AppsFlyer's pricing isn't publicly listed, but the floor for a real plan with useful features is around £700-£1,000 per month. Branch is similar. Adjust, which is owned by AppLovin now, tends to be in the same range once you're past their free tier's hard caps.
The free tiers exist but are severely limited. AppsFlyer's free plan caps you at 10,000 non-organic installs per month and removes most of the attribution features that actually matter. Branch's free plan is similarly constrained. Both are really "get started" tiers meant to onboard you before upselling.
For most indie developers, even hitting those free tier limits is aspirational. But the features you actually need — granular link attribution, purchase tracking, webhook integrations — are paywalled behind plans that cost as much as a London rent payment.
What attribution actually tells you
Attribution answers one question: where did this user come from?
More specifically, when someone makes a purchase in your app, attribution tells you which marketing action led to that purchase. Was it the Instagram post? The Reddit thread? The Product Hunt launch? The YouTube video someone made about your app six months ago?
Without attribution, you're guessing. You might have 200 downloads this week and no idea whether they came from something you did or a random spike in organic search. You can't improve what you don't understand.
The data attribution gives you isn't complicated, but it's valuable. You learn which channels bring paying users versus just browsers. You learn your rough cost-per-acquisition if you're spending money on promotion. You learn which content types convert, which helps you decide where to spend your limited time.
For an indie developer working evenings after the kids go to bed, time is the real currency. Attribution tells you where your time is paying off.
The cost mismatch
Here's the core problem. Enterprise attribution tools charge enterprise prices because they're selling to enterprise customers who have enterprise budgets and enterprise problems.
A company spending £100,000 per month on user acquisition can justify spending £1,000/month on the tools to track that spend. That's 1% overhead. Totally reasonable.
A solo developer making £800/month from their app cannot justify spending £700/month on attribution tools. That's 87% overhead. It wipes out the revenue.
The pricing model makes sense for the enterprise market. It just doesn't make sense for everyone else.
And "everyone else" is a lot of developers. The App Store has something like 1.8 million apps. The vast majority of those are made by solo developers or tiny teams. Most of them are not printing money. They need to be careful with costs.
What smaller attribution tools offer
Below the enterprise tier, there are a few options worth knowing about.
Some tools offer genuinely free tiers with useful attribution. Firebase Dynamic Links was the obvious one for years until Google killed it in 2025. That left a gap that various tools have tried to fill.
Link shorteners with basic click tracking (Bitly, etc.) give you tap counts but no purchase attribution. Useful for seeing which content gets clicks. Not useful for knowing which content drives revenue.
RevenueCat itself has some attribution fields but relies on you setting the appUserID correctly and doesn't do the link-tap-to-purchase matching itself. You need something to feed that attribution data in.
LinkOwl sits at the other end of this spectrum. Free to use, with purchase attribution that works through a webhook integration with RevenueCat or Stripe. You pay 5p per attributed purchase, nothing per month. If your app makes £100 this month and 80% of those purchases came through tracked links, you pay £4. That's a reasonable cost structure for a pre-revenue or early-revenue app.
Why the 5p-per-sale model changes the equation
The monthly subscription model doesn't work for indie developers because indie apps have variable, often low revenue months. Paying £500/month when you make £300 that month is painful. Paying £5 when you make £300 is fine.
The per-sale model also aligns incentives differently. You're paying for attribution when attribution is working. If a month goes badly and you make nothing, you pay nothing for the tracking tools either. The cost scales with the outcome rather than against it.
This is the model that makes attribution accessible for developers who aren't yet profitable enough to afford enterprise tools but need the data to become profitable.
A few things attribution won't tell you
It's worth being clear about what attribution data can't do.
It won't tell you why someone bought. That takes user research, surveys, talking to customers. Attribution shows you the path, not the motivation.
It won't fix a bad product. If your conversion rate from install to purchase is 0.3%, better attribution data will just help you understand the problem more clearly. It won't solve it.
And with iOS privacy changes since ATT, fingerprint-based attribution has become unreliable. Any tool relying primarily on fingerprinting is giving you estimated data. Link-based attribution is more reliable because it doesn't depend on device identifiers at all — the link tap itself is the tracking mechanism.
If you're an indie developer who's been posting content, running promotions, or sending email campaigns without tracking which of those activities actually lead to revenue, it's worth setting up attribution. The data is genuinely useful. And you don't need to spend £1,000 a month to get it.