Meta says 1,200 installs, our MMP says 900. How big a gap is normal?
Same campaign, same week. A 300-install gap feels like a lot. At what point is a mismatch worth digging into instead of just accepting it?
Answer from the Linkrunner team
A gap of 5 to 10% between an MMP and an ad platform is normal. Yours is 300 out of 1,200, or 25%, and a gap above 20% almost always points to a technical cause rather than measurement variance, so it is worth digging in. Work through it in order. Compare a 7-day cohort from a week ago rather than this week's data, to rule out processing delays. Align attribution windows, since a 7-day MMP window against a 28-day Meta click window can create a 15 to 20% gap on its own. Check both dashboards use the same timezone. Then compare your MMP's organic installs with genuinely new users in your analytics; if organic runs more than 20% higher, paid installs are leaking into organic. On Android, that leak is often Meta installs the Play referrer never saw: https://linkrunner.io/blog/meta-install-referrer
More on this: linkrunner.io/blog/attribution-discrepancy-troubleshooting-the-complete-diagnostic-guide