SKAN null-rate: a bid-adjustment worksheet that does not invent a market average
If a share of SKAN postbacks arrives without a usable conversion value, a bid that assumes every install reports revenue will overstate expected value. A reproducible illustrative example with editable inputs and explicit limits.
If a share of SKAN postbacks arrives without a usable conversion value, a bid that assumes every install reports revenue will overstate expected value. This sheet applies the reader’s own observed null rate.
Define the calculation before using it
Enter intended bid, observed null-rate fraction, and the residual value assumed for null postbacks (often zero). Compute an expected SKAN-informed bid. Do not paste a vendor null-rate chart as the input.
Work through the synthetic example
A 1.20 bid with a 0.35 null rate and zero residual becomes 0.78 expected value per install under this linear assumption. That is not Apple’s recommended bid.
| Output | Worked-example result |
|---|---|
| Expected value per install | 0.7800 |
| Haircut versus intended bid | 0.4200 |
Use the artifact and preserve its assumptions
Open the editable calculator to change the inputs and inspect the sensitivity view. The CSV records synthetic inputs and expected outputs; the JSON fixture keeps the equations available for reproduction. These calculations have been checked against the stated example. No measured campaign data is included.
The sensitivity rows vary only nullRate by 20% below and above the entered value. They are scenarios, not confidence limits or a forecast distribution. A row outside the model’s constraints is labelled rather than turned into a plausible-looking result.
Evidence and limits
Apple documents that conversion values are subject to privacy thresholds and may be unavailable. It does not publish a market null rate. See SKAdNetwork, especially “Postbacks / conversion values”.
Null rates are account- and geo-specific. Coarse conversion and crowd anonymity are ignored here.
Background: SKAN Conversion Value Mapping, Revisited for 2025 and SKAN 4 and AdAttributionKit, a year in: what buyers actually get back. These existing articles provide context; the present calculation does not verify every archived claim.
No actual campaign outcome or recommended industry default is claimed.
