OEM preload claims need the denominator: devices, preloads or first opens
Digital Turbine’s public terms distinguish fill rate on activated devices from CVR to first open—do not treat device population as opted-in ad IDs.
Digital Turbine’s public glossary defines fill rate as the share of activated devices that received a preload, and CVR as movement from preload to first open. Billing can be CPP (download at first boot) or CPI (first open). Those denominators are not “all devices in market” and not “opted-in advertising IDs.” Terms.
Name the inventory event
OEM/preload supply is delivered through carrier and manufacturer partnerships. A reach claim about millions of devices describes partnership coverage or activations, not necessarily addressable GAID/IDFA inventory under the same privacy rules as in-app auctions. The Playrix case reports high incrementality for dynamic installs in a vendor-advertiser study; it still does not convert device population into an advertising-ID denominator.
Buyer checklist
Our recommendation is to require which event is counted (activated device, preload, download, first open), which OS/geo/OEM set, and how attribution treats preloads before testing OEM inventory. If advertising-ID opt-in share is not disclosed, leave it missing.
Inspect the evidence
Download the source-population and comparability card. The extraction separates documented facts from our analysis and unknowns.
| Field | Evidence or limit |
|---|---|
| Report / publisher | Digital Turbine terms glossary + Playrix case / Digital Turbine |
| Period | Glossary dated 2020-06-03 (definitions); Playrix case undated on page |
| Population | Devices/activations in DT OEM/carrier partnerships; case example US Android dynamic installs for Playrix |
| Metric under review | Fill rate (preloads on activated devices); CVR preload→first open; CPI vs CPP billing events |
| Aggregation | Campaign/placement definitions; case-level incrementality claim for one advertiser |
| Critical limit | Device or activation population is not the same denominator as opted-in advertising IDs; glossary does not publish an ID opt-in rate for OEM inventory. |
| Comparable with | Another on-device campaign using the same event definitions (preload vs first open) |
| Not comparable with | In-app auction CPI on GAID/IDFA populations without restating the event |
Further reading in the existing archive: The benchmark report is selling you something; measurement-lab posts on MMP versus network install counts. These links provide background; this check does not independently verify their full contents.
