Services
Six service lines, with the scope written down.
These are retained engagements and fixed-scope audits, not packages. Most clients start with the measurement audit and add media operation once the baseline is one everybody trusts. Scope is agreed in writing before anything starts, and every line below says what you actually receive.
01
User acquisition
One portfolio, one definition of a paying user, run across the stores and the networks that actually move volume at scale.
In scope
- Apple Ads across all four App Store placements — Today tab, Search tab, search results and product pages — with brand, category, competitor and discovery campaigns separated so bids and negatives can be controlled independently.
- Google App campaigns for installs, for engagement and for pre-registration, with asset groups built against where the ads actually serve: Search and search partners, Google Play, YouTube, the Display Network and Discover.
- Meta and TikTok app campaigns, including the SKAN-era campaign structures and the event mapping each platform needs to bid on anything past install.
- SDK and programmatic networks — Unity, AppLovin, Moloco, Liftoff and similar — with placement-level blocklists and a validation policy applied before spend scales.
- Bid and budget operation on a weekly cadence, with the decision log kept so you can see why a change was made and not only that it was.
What you get
- A channel plan with a stated role per channel, not a list of every network that will take your money
- Campaign structures documented so your team can operate them without us
- A weekly decision log and a monthly portfolio review
02
App Store Optimization & store listings
The store page is the only part of the funnel where every channel converges, and the two stores do not work the same way.
Capability detailIn scope
- Keyword research and metadata written to the real field limits — on the App Store, a 30-character name, a 30-character subtitle and a 100-character keyword field per localisation; on Google Play, a 30-character title, an 80-character short description and a 4,000-character description per language, with no separate keyword field at all.
- Creative sets for both stores built to spec: screenshots, app previews, feature graphic and icon, at the dimensions and formats each store enforces.
- Custom Product Pages on iOS — up to 70 per app — mapped to paid campaigns, keyword themes and lifecycle moments rather than made once and forgotten.
- Google Play custom store listings — up to 50 — targeted by country, by user or buyer state, or by search keyword.
- Store-side experiments: Product Page Optimization on iOS and store listing experiments on Play, sized so they can reach a conclusion inside the time each platform allows.
What you get
- A metadata set per localisation, with the reasoning for each field
- A tested, versioned creative set per store
- An experiment calendar with the traffic each test needs to conclude
03
Creative strategy & production
Install creative is a research programme that happens to produce ads. The output is a conclusion, not a folder of files.
In scope
- Concept development from the product itself — mechanics, onboarding, the moment of value — rather than from whatever the category is currently copying.
- Production of static, video and playable-adjacent variants at the aspect ratios each placement actually serves.
- A test design that isolates one variable at a time, with the structural discipline to keep learning attributable when the platform is doing the allocating.
- Localisation and cultural adaptation for the markets in the plan, not machine translation of the US cut.
- A creative library with results attached, so a concept retired six months ago is not reinvented from scratch.
What you get
- A monthly production slate, agreed in advance
- A test readout per concept with the decision made explicit
- A tagged creative library you keep
04
Measurement & attribution engineering
The service line the rest depends on. Implementation, migration and audit across the major MMPs and the store-side frameworks underneath them.
Capability detailIn scope
- MMP implementation and migration on AppsFlyer, Adjust, Branch, Singular, Kochava, Tenjin or Airbridge, including SDK integration review and server-to-server event mapping.
- SKAdNetwork and AdAttributionKit configuration: conversion-value schema design, window and lock strategy, source identifier structure.
- Android attribution via the Google Play Install Referrer, including the fields worth keeping and the timing assumptions that break if you read them twice.
- Deep linking and deferred deep linking, owned-channel links, and the routing rules that decide where a tapped link actually lands.
- Cost and ad-revenue ingestion, raw-data exports, and the warehouse model that makes a cohort reproducible.
What you get
- An implementation specification engineering can build from
- A conversion-value schema document with the trade-offs stated
- A reconciliation between MMP, store console and warehouse
05
Incrementality, retention & LTV
Attribution tells you which touch to credit. Incrementality tells you whether the spend caused anything.
In scope
- Geo and audience holdout design, including how long a test has to run before its confidence interval means anything at your volume.
- Cohorted payback and retention curves by source, placement and creative, built on raw data rather than platform-reported aggregates.
- Predicted LTV models used where they are honest — directional early signal — and not where they are not.
- A blended view that reconciles with finance: one spend number, one revenue number, and a documented bridge between the platform totals and the general ledger.
- Media mix review for the channels too large or too privacy-constrained to attribute cleanly.
What you get
- A holdout test plan with power calculations
- A cohort dashboard specification
- A written reconciliation of platform, MMP and finance numbers
06
Fraud & traffic validation
Every MMP sells fraud detection. Almost nobody tunes it, and a default rejection policy is a negotiation you lose twice.
Capability detailIn scope
- Click-to-install time distribution analysis by partner and sub-publisher — legitimate installs are not seconds behind the click.
- Click spam and click injection rules, and the evidence pack needed to claw back on them.
- SDK spoofing checks, including signature verification and replay detection on the integration itself.
- Device farm, emulator and residential-proxy pattern review on new partners before spend scales.
- A written validation and rejection policy your networks will actually accept, plus the dispute process when they do not.
What you get
- A partner-level validation report
- Tuned MMP fraud rules with documented thresholds
- A rejection and clawback policy, and the disputes run on your behalf
Pricing
There is no rate card, and that is deliberate.
A fixed-scope audit of one app on one store is not the same purchase as a retained programme across two platforms and six markets. We scope against your account and quote in writing, in USD, before any work starts — including what is in scope, what is not, the term, and the notice period. No media rebates, no undisclosed markups, no arbitrage on your buy.