Concepts
Parity targets
How a recommended regional price is derived, from purchasing-power data through dampening, bounds, and rounding.
A parity target is the price Paywall Parity recommends for one market, for one plan. It is deterministic: the same inputs always produce the same target, and every input is stored with the result.
The short version
Take the base price. Convert it at the market's exchange rate. Multiply by an economic factor that reflects what money is worth there. Clamp it to the strategy's bounds. Round it to something a store will accept and a human will recognize as a price.
The economic factor
Three published indicators feed it, each expressed as a factor relative to the base market:
| Input | What it captures |
|---|---|
| PPP | What a unit of currency actually buys locally. The primary signal. |
| Price level | How local consumer prices compare to the base market. |
| Income | GDP per capita, as a proxy for ability to pay. |
PPP is required. The other two are optional — a market missing them still gets a target, with a correspondingly lower confidence score.
Dampening
Raw PPP is too aggressive to use directly. A market whose PPP factor is 0.25 does not mean your subscription should cost a quarter of the US price; it means the cost of a local basket of goods is a quarter. Software is not a local basket of goods, and pricing straight off raw PPP produces targets that destroy revenue without a matching gain in conversion.
So each factor is pulled towards 1 by a strategy-controlled intensity:
dampened = 1 - intensity x (1 - factor)
At intensity 0 the factor is ignored entirely (dampened = 1). At intensity 1 it is applied in full. Balanced uses a PPP intensity of 0.65, so a raw PPP factor of 0.25 becomes:
1 - 0.65 x (1 - 0.25) = 0.5125
Roughly half the base price, not a quarter.
The three dampened factors are then multiplied together to give the combined factor.
Bounds
The combined factor is clamped to the strategy's minPriceRatio and maxPriceRatio. This is what stops a market with broken input data from producing an absurd target.
A market that lands exactly on a bound is worth inspecting. It usually means the data is wrong rather than the market being genuinely extreme — a dollarized economy whose World Bank PPP compared against its local exchange rate yields a nonsense ratio, for instance. Markets that pin to a bound for that reason belong in Protected markets.
The pipeline, in order
Each stage is stored in the calculation snapshot, so any target can be taken apart afterwards.
- Converted — base price at the market's FX rate.
- Candidate — converted price multiplied by the clamped economic factor.
- Override — a market-level absolute price or ratio replaces the candidate, if one is set.
- Constrained — clamped to any floor and ceiling configured for the market.
- Rounded — snapped to the market's rounding increment.
- Final — constrained again, because rounding can push a price back outside its bounds.
For a protected market, the final answer is the current price regardless of all of the above. The calculation still runs and is still explained, so you can see what would have been recommended.
What it is not
The target is not a forecast. It does not predict revenue, conversion, or elasticity, and nothing in the calculation observes how your customers actually behaved. It is a defensible starting position derived from published economic data — which is a different and more honest claim.
Measuring what a price change actually did is planned, not built.