A report is a set of claims with evidence attached. A story is a set of claims. Proof Review is the step in CommunityOS that keeps the two from mixing: no completed action counts — for reporting or for rewards — until it is verified.
Most community marketing reporting fails not because the numbers are small but because the numbers are unfalsifiable. Impressions, reach, "engagements" — computed over audiences that are majority-inauthentic in the worst cases, aggregated by platforms with their own incentives, impossible for the person receiving the report to check. Somewhere between the campaign and the slide deck, marketing turned into storytelling. Proof Review is the mechanism for turning it back.
What is Proof Review, mechanically?
The last gate in the Loop. An activation mission goes out to a scored community member — write a thread, publish a write-up, engage a launch. When the member reports it complete, the completion enters a pending state with its evidence: the URL, a screenshot, the engagement captured at review time. An operator reviews the evidence against the ask. One tap to approve — the action enters the verified ledger and the reward releases. Reject sends it back with a reason; nothing counts, nothing pays.
Two design choices matter here. Verification happens per action, not per campaign — there is no "the campaign went well." And review is human, on purpose: an operator looks at each proof, because the entire value of the ledger depends on nothing entering it that a human would not endorse. The same principle runs through outreach — reviewed, personal, sent by a human; nothing fires automatically.
What actually gets verified?
- Existence — the post or artifact is real and live, at a URL, matching the mission.
- Authorship — it came from the scored community member the mission targeted, a real account that passed Bot-Kill.
- Substance — it satisfies the ask. A thread is a thread, not a repost with a comment.
- Reception — engagement is captured at review time, recorded as observed, timestamped fact.
Equally important is what does not get in: impressions estimates, projected reach, anything the platform asserts but the operator cannot inspect. The ledger would rather be smaller and true.
Why does verification change reporting?
Because it changes what a stakeholder can do with the document. A verified-only report supports interrogation: every line is an action, by a named real member, with evidence one click away. When a client's CFO asks "how do I know this happened?" the answer is the ledger itself, not a methodology appendix. This is the reporting stack agencies run on CommunityOS — community work expressed as a defensible P&L line instead of a vibes slide.
Unverified reporting asks the reader to trust the reporter. Verified reporting asks the reader to check.
There is a compounding effect, too. Snapshots track what changed between scans — new Champions, archetype transitions, community growth that survived filtering — so each campaign's verified actions accumulate into a longitudinal record. Quarter two's report does not restate quarter one's claims; it builds on quarter one's evidence.
Why not just automate the verification?
Full automation reintroduces the exact vulnerability the system exists to remove. Anything checked purely by machine gets gamed at machine speed — quest platforms proved this at scale, paying out on millions of "completed tasks" that were clicks from farms (the comparison is drawn out on the Zealy page). The economics work because activation targets a filtered, ranked population in the hundreds — the Act Now queue, 298 people in the first production scan, not 78,181 — so human review is not a bottleneck; it is affordable rigor at exactly the scale where rigor pays.
What does this mean for how you buy community tools?
One question separates the categories: when this tool tells me something happened, can I see the evidence? Listening platforms summarize what was said. Quest platforms count what was clicked. A verified activation ledger records what real, scored people demonstrably did — and that is the only one of the three a finance team will underwrite. The pricing that goes with it is public, on the pricing page.
Quick answers
What is Proof Review?
Proof Review is the verification step in CommunityOS. Every completed community mission — a post, a thread, a piece of content — is checked against evidence (URL, screenshot, captured engagement) before it counts toward reporting or releases a reward.
How do agencies prove community ROI to clients?
By reporting only verified actions: who was activated, what they actually did, with evidence attached to every line. A verified-only ledger survives CFO scrutiny in a way impressions-based reporting cannot.
What happens if a submitted proof is rejected?
Rejection sends the mission back to the member with a reason; the reward is not released and the action never enters reporting. Only operator-approved actions count.