> Site index: https://usehardal.com/llms.txt
> Every content route also serves markdown at <url>.md

# Web ad, app install: who gets the credit?

Meta, Google and TikTok can each claim the same app install. Here is what an MMP does, and how to read the gap.

Source: https://usehardal.com/blog/web-ad-app-install-who-gets-the-credit
Published: 2026-11-19T22:00:00
Updated: 2026-11-19T22:00:00
Author: Hardal
Category: Marketing Measurement

---

Meta, Google and TikTok each claim the same app install. Which one is right? In a way, all three can be, at least inside their own dashboards.

Picture three friends, each saying they introduced you to your partner. Someone has to keep the guest list. Without that list, you end up thanking everyone and learning nothing about who actually brought the person in.

## What an MMP does, and what it does not

That guest list is the role of a mobile measurement partner, or MMP. In its own reports, the MMP decides which campaign gets credit for an install. It also connects the app to a large set of ad platforms and other partners.

It does not stop those platforms from claiming the same install in their own dashboards. Meta, Google and TikTok are self-attributing networks. They can still report credit by their own rules, even when an MMP is in place. If you compare the two sides and expect a perfect match, you will always be disappointed.

## What changed on iOS

Since iOS 14.5 in April 2021, an app needs App Tracking Transparency consent before it can use the IDFA, the identifier that used to stitch a person across apps. Without that consent, iOS install data comes through Apple's SKAdNetwork, and increasingly through its successor, AdAttributionKit. The data is aggregated, and it arrives with a delay of a day or more. You still get a signal. You do not get a person-level trail.

Android has not removed its advertising ID. The two systems are not the same, and it is easy to talk about "mobile privacy" as if they were.

The MMP and an ad platform will never match exactly. Attribution windows differ. View-through rules differ. Self-attribution differs. Some gap is normal. A large gap is the warning sign, especially if one side is consistently much higher and nobody can explain why.

## In-app revenue belongs on the server

Installs are only the start. In-app revenue is more reliable when it is sent from the server: store-validated purchases, renewals and refunds, and, for games, ad revenue, all through the MMP's server-to-server API. Device-only events are easier to lose, and easier to spoof.

The web and the app often tell two stories about the same customer. A person sees a web ad, installs the app, and buys there. If those systems do not join, the web campaign looks weak and the app looks like it created a customer from nowhere.

Growth teams can start with three questions. Who is the source of truth for installs? Are in-app purchases, renewals and refunds sent from the server, not only from the device? Do our web and app numbers tell one story?

An install two platforms both claim is not two installs.

## Today's small task

Pick one campaign and compare the installs your MMP reports with what the ad platform reports for the same week. Write down the gap. Remember that some difference is expected. What you are looking for is a gap that is hard to explain.

If you want to join web and app data with first-party, server-side signals, start at [hardal.ai](https://hardal.ai).

Hardal is the first-party data and agentic analytics platform. We help enterprises send accurate signals to their ad platforms and grow revenue, using server-side data from their websites, mobile apps and AI agents.
