The reporting entry must sit next to the offending content
Aliases: in-context reporting · report entry placement · contextual report
What it is
Where the reporting entry lives decides both the rate and the quality of reports: an entry next to the offending content (the long-press or overflow menu on the item itself) lets users report with full context; a report buried in a help centre or deep settings becomes a treasure hunt requiring extra motivation — genuine reports leak away, and what remains in the queue is something else.
Why it happens
Proximity's value is context binding: a report launched beside the content automatically carries the item ID, author, timestamp, and UI state, making triage sharper and review faster; a report from the help centre demands the user reconstruct "where I saw what," and every loss in that description converts directly into review cost. Proximity also captures timing — reporting motivation peaks at the moment of seeing the problem and decays with any delay. The complete entry structure has three layers: content-level (the item's own menu), account-level (the profile), and appeal-level (disputing an outcome) — missing any layer locks out one class of users. Prominence trades against abuse: too visible invites malicious reporting, too hidden suppresses real reporting; the baseline is "findable, but not on the main path."
Where it stops holding
Proximate does not mean classification-free: a one-tap report with no taxonomy drowns valid signals in noise — "entry near, taxonomy sharp" is a combined requirement. Multimodal content needs different proximate forms: video needs time-point anchoring and live streams need timestamped snapshot capture, otherwise "the moment being reported" cannot be fixed as evidence. Across devices the form may differ (long-press on mobile, hover on desktop), but the three entry layers must exist with equal function on every surface.
Applying it
- Put the report entry on the first level of the content menu, auto-carrying content context, never asking users to re-describe "where they saw it."
- Use users' language for categories (spam, fraud, harassment, infringement, involving minors), coarse categories first then fine ones; category wording goes through copy review.
- Verification: report-funnel analysis — exposure-to-submission conversion, taxonomy completion, and the share who abandon with "couldn't find the entry"; measure the last one as before/after on entry redesigns.