Reading Activation Reports
Start with the question
Section titled “Start with the question”Reports are most useful when tied to a decision:
- Did recipients interact with the physical product?
- Did the experience hold attention?
- Did people take the intended action?
- Which placements or program variants performed differently?
- Did verified objects behave as expected?
Metric families
Section titled “Metric families”| Family | Examples | What it does not prove alone |
|---|---|---|
| Reach | Taps, sessions, experience views | Attention or conversion |
| Engagement | Engaged time, scroll, media milestones | Purchase intent |
| Action | CTA clicks, form progression, claims | Confirmed revenue |
| Outcome | Qualified lead, confirmed payment, fulfillment | Long-term value |
| Trust | Verified, invalid, or replayed secure interactions | Customer identity unless the workflow establishes it |
| Quality | Internal tests, bots, suspicious bursts, dormant objects | Fraud without investigation |
Filtered and total views
Section titled “Filtered and total views”Reports may default to a filtered scope that removes known internal testing or identified noise from headline metrics. Raw records remain available to appropriately authorized operators; exclusions do not rewrite history.
Geography
Section titled “Geography”Network-derived geography is approximate. A region or city label is not GPS and should not be used to infer a person’s precise location.
Small samples
Section titled “Small samples”For small physical runs, counts and individual customer feedback often matter more than percentages. Avoid strong comparisons when sample sizes, distribution, timing, or placement differ materially.
For a deeper technical account of raw events, quality scopes, policy resolution, and authoritative outcomes, see Data and Control Model.