Probabilistic plus deterministic rows: do not add them as one truth
An export that mixes deterministic matches and probabilistic or modelled rows can look complete while double-counting the same player. A reproducible illustrative example with editable inputs and explicit limits.
An export that mixes deterministic matches and probabilistic or modelled rows can look complete while double-counting the same player. This lab keeps the methods in separate columns.
Define the calculation before using it
Enter deterministic attributed conversions, probabilistic or modelled conversions, and the count already present in both methods after an authorised join. Report each method and the overlap; do not present the sum as incremental conversions.
Work through the synthetic example
400 deterministic and 180 probabilistic rows with 40 already in both are 540 distinct rows, not 580. If a forecast used 580, it treated a modelling method as extra people.
| Output | Worked-example result |
|---|---|
| Naive sum | 580.0000 |
| Distinct rows | 540.0000 |
| Overlap share of naive % | 6.8966 |
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 both 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
Attribution products distinguish matching methods. Public documentation does not authorise adding every method’s totals. The overlap here is a teaching input. See AppsFlyer attribution model, especially “Attribution model”.
Vendor modelling methods differ and are often undocumented in public help pages. This does not validate any vendor’s probabilistic matcher.
Background: Reading an MMP dashboard without fooling yourself and AppsFlyer, Adjust, Singular: a buyer's comparison. These existing articles provide context; the present calculation does not verify every archived claim.
No actual campaign outcome or recommended industry default is claimed.
