Guidelines
How Paywall Parity calculates a regional price, what evidence sits behind each recommendation, and which capabilities are not built yet.
Every page carries one of those labels, and a page marked planned describes something that does not exist yet. The alternative — discovering the gap after building a workflow around it — is worse for everyone. Machine-readable copies of every page are at /guidelines/raw.
What the product does, how to get a first result, and what it will never do behind your back.
The closed pricing loop, what each stage is responsible for, and which stages are live today.
Connect Google Play, import a catalogue, and read a full regional matrix. Roughly fifteen minutes, no store writes.
Every rule that stands between a calculation and a real price change, and why each one exists.
The vocabulary every screen uses: baselines, parity targets, strategies, rounding, confidence, drift.
What Google's own converted price is, why it is shown beside your price, and every reason the column can be blank.
How a recommended regional price is derived, from purchasing-power data through dampening, bounds, and rounding.
The three strategy presets, the exact intensities and bounds behind each, and how to choose between them.
How the confidence score is computed, the thresholds that gate a recommendation, and when the system abstains entirely.
The four drift severity bands, what triggers each, and why an exchange-rate movement can never authorize a store write.
Markets whose price is held regardless of what the strategy computes, and how to tell a real recommendation from a data artifact.
The operating loop, one task at a time, from connecting a store to measuring what a change did.
The eight steps from a service-account key to an imported catalogue, and what to check when one of them fails.
How the inbox is ordered, what makes an item approvable, and what approval does and does not do.
How a deployment draft is built and validated, what publishing does, and the release gate that keeps it off by default.
Exact formulas, inputs, statuses, and error codes. Look things up here rather than inferring them.
What each provider supplies, what it is allowed to change, and where its limits are.
The Android Publisher endpoints used, the scope and permissions required, and the limits of what the provider can do.
Apple support is planned and not implemented. What exists, what has to change, and what would have to be proven.
Outcome measurement is planned and not built. What it will require, and why no uplift claim is made today.
Permissions, credential custody, retention, audit records, and how failures are reported.