Acquire users who pay

We predict what each new user will be worth within hours of install and send that value to Meta, Google and TikTok. Their bidding then spends your budget on finding more of the users who pay.

Book a callSee what we build

MetaGoogleTikTokNew usersevents from theirfirst hoursValue modelpredicts what eachuser will spendHarm markersflagged for yourplayer-protection teamAd platformspredicted value isthe conversion signal

Ad platforms find the users you ask for

Most campaigns optimise for an install or a first deposit. Those events arrive fast, so the platforms learn from them. They say very little about who will still be spending in a year.

Budget drifts toward cheap users who leave in the first week, while your best users cost the same to acquire as everyone else. The difference shows up in revenue months later, long after the spend has gone.

What we build

A model that scores every new user on predicted lifetime value from their first hours of behaviour, wired into the conversion API of each ad platform you buy from.

Event pipeline
Install, session, purchase and deposit events from your app or platform, joined to the campaign data from your attribution provider.
Lifetime value model
Trained on your own history of who paid and who left. It scores each user within hours of install and updates the score as more behaviour arrives.
Conversion feeds
Predicted value goes back to each ad platform in the form its value-based bidding reads: server-side or attribution-partner events, and SKAdNetwork conversion values on iOS.
Monitoring
Predicted value is checked against real revenue as it comes in, by channel and by campaign. Drift gets caught before it costs budget.
Player-protection markers
For iGaming companies, the same behavioural data flags patterns linked to harm, such as rising deposit frequency or chasing losses, and routes them to your player-protection team.

How it pays

  • More value from the same budget

    Bidding shifts toward people who look like your best payers, so each cohort you buy is worth more.

  • An early read on every campaign

    You see what a new campaign's users are worth within days. Weak campaigns get cut while their spend is still small.

  • Creative judged on value

    Each creative and audience is scored on the value of the users it brings in, and budget follows the winners.

  • Growth your licence can live with

    Your player-protection team gets earlier signals from the data the model already reads.

How we start

  1. Find the value

    We study how revenue builds over a user's life in your data and where your spend goes today. From that we work out what value-based bidding is worth to you.

  2. Build and prove it

    We train the model and connect it to one platform first. A holdout of campaigns keeps optimising the old way, so the uplift is measured against a control.

  3. Run it and compound

    We retrain as your audience shifts and roll the feed out to every channel. Every month we report what the model earned against the holdout.

Who it suits

Games, apps and iGaming companies that buy users through paid social and search and keep event data on what those users do after install. A few months of history is enough to start training.

Find the users worth paying for.

Bring a campaign you want to grow. On the first call we'll show you what your own data can predict about it.

Book a callhello@millracedata.com

More of what we build

  • Price with demand

    Prices that move with live demand and stock, for rooms, tickets, products and in-game offers. You keep the margin a fixed price list gives away.

  • Recommend what sells

    Recommendation engines for everything you sell, from products to in-game offers, tuned to the number that matters to you, whether that's basket size or retention.

  • Keep play fair

    Cheat and fraud detection built on the event data you already collect. Bad actors get caught early, and the players and revenue they would drive away stay with you.

  • AI that knows your business

    Agents and knowledge bases grounded in your own data and documents. Every team's AI tools answer with what the company already knows, and they get sharper as people log what they find.

  • The platform under it all

    A data platform that keeps every model and report fed with fresh data you can trust, on a cloud bill that makes sense.