Migrating 30 live games to a new MMP: Tripledot's Adjust move, and what to ask before you switch

The event

Adjust announced on 4 March that it has migrated more than thirty live Tripledot Studios titles (Solitaire and Woodoku among them) onto its measurement platform, moving billions of install records in the process. Adjust's own description leans on an automated device-import process that preserves original attribution sources and install timestamps; the company says the switch produced no impact on live campaigns and no data gaps.

Those are Adjust's claims. They come from the vendor describing its own migration of its own new customer, so read them as an account of a successful sale rather than an independent audit. Nothing here is implausible; MMP migrations of this kind happen and can go cleanly. But "no data gaps" is a marketing sentence until a studio's own reporting confirms it across a full attribution window, and Tripledot hasn't published its own account of the switch.

What actually breaks in an MMP migration to Adjust

Studios that have been through this exercise report trouble in a narrower set of places than the "big bang" framing suggests. Historical cohort continuity comes first. If lifetime value curves or retention cohorts or predictive models rest on install date and source, any re-mapping during migration needs to preserve both exactly, or every historical cohort report becomes incomparable to anything reported after cutover. Then the postback chain: every network integration, every custom postback, every internal warehouse pipeline reading from the old MMP's API needs re-pointing and testing before cutover rather than during it. Fraud and SDK signature parity is the third trouble spot, because a new MMP's fraud filters rarely match the old one's thresholds exactly, and a studio that doesn't benchmark spend by source before and after the switch can mistake a filter change for a real shift in traffic quality.

A fourth area gets overlooked more often than the first three: internal reporting and dashboards. Any business intelligence layer or finance report or creative-performance dashboard built on the old MMP's export schema or API field names will silently break, or worse, silently miscalculate if the new MMP structures the same data differently. Campaign and creative ID mapping is a common failure point here. If creative-level identifiers don't carry over cleanly, a studio loses the ability to tie historical creative performance to the concepts and variants that produced it, which matters directly to any team running the kind of iterative creative testing this desk covers regularly.

Questions to ask before signing

Before committing a live portfolio to a migration, get the incoming vendor's written commitment on cohort continuity: will lifetime attribution windows and cohort definitions match exactly, not approximately, across the cutover date. Ask for a parallel-run period in which both MMPs receive live postbacks at the same time, long enough to reconcile spend and revenue by source before anyone switches the old system off, rather than trusting a single cutover weekend. Ask how SKAdNetwork and AdAttributionKit conversion value mappings carry over; they're configured per app and per network, which makes them easy to miss in a bulk migration. And ask what happens to raw historical data. Is it fully portable in a format your own data team can validate, or does it stay locked inside the new platform's reporting layer?

None of this is a reason to avoid switching MMPs. It's a reason to treat "we migrated thirty games with no impact" as a claim to test against your own numbers in the first reporting cycle after cutover, not a guarantee to build a budget review around.

Related archive reading

These articles provide related context and remain subject to their stated review status.

• The MMP Migration Checklist to Run Before You Switch
• The MMP RFP Checklist to Run Before You Switch Vendors