A real insight from our tracking of Comintelli, exactly as it was checked. See Comintelli's live page
And what they stopped.
Real drafts these checks rejected or retracted, and why. Vendors stay unnamed so we never repeat a claim we threw out.
“A vendor cut its annual plan price.”
That was the normal annual-billing discount, not a price cut.
“A vendor is running 400 active ads.”
The ad library shows at most 200 results per page. We hit that limit; we did not count 400 ads.
“A vendor is listing exactly 50 open roles.”
50 is the job board’s page limit, not a real count. The true number is unknown.
Every draft runs through 29 checks. See them all.
350+ sources, by category
Every source is public and checkable. We store the source URL, the scrape timestamp and a verbatim extract, never a paraphrase. See how source coverage works →
The 7 source categories above name ~55 platforms. Why 350+? Google News alone surfaces more than 450 publications. Add nine review platforms, five social networks, three ad libraries and each company’s own web pages, and the total is over 476. We say 350+ to stay conservative.
Monitoring cadence
Weekly, not real-time. Ads and news are scraped daily; pricing pages, changelogs, reviews and the rest of the 350+ sources on the Sunday pipeline. Every source still lands in the following Monday's briefing, not within hours. If your team needs hour-grain signals, IndustryLens is not the right tool.
When we get it wrong
Checks reduce errors; they do not eliminate them. When a customer, a competitor or our own QA contests a published claim, we update it, retract it or stand by it, and we note the correction on the relevant report page.
You see this audit trail in the “View evidence” panel under any finding that contains a derived figure.
Frequently asked questions
Why 350+ sources, isn't that duplicative?
Source overlap is a feature, not a bug. When a review site, a job posting and a company blog all describe the same launch, the claim is easier to verify. When they contradict each other, the insight is flagged instead of shipped. Redundancy is how hallucination detection works at scale.
Do you use AI to generate insights?
Yes, an AI synthesis pass converts raw observations into insight drafts, but that pass is followed by an adversarial verification pass that checks every claim against its source. Insights that fail verification are dropped, not edited. Every insight also runs through 29 named checks, each looking for one kind of mistake, before it reaches a customer briefing.
How does IndustryLens differ from Klue or Crayon?
Klue and Crayon are strong enterprise tools that require a dedicated CI analyst to get value. IndustryLens is built for teams without that resource, one person can run CI for the entire org. We also differ on transparency: every claim ships with its validation result and a clickable source URL, so your team can verify what we deliver.
What happens if a source goes behind a login or changes its format?
Bot-block detection (Layer A, check 4) and format validation (Layer A, check 5) catch this automatically. The source is flagged as degraded, removed from the active pool for that run, and a pipeline alert is triggered. Insights that relied on that source in previous runs are not re-surfaced until the source recovers.
Corrections and disputes
See a finding you believe is wrong? Contact us with the report URL and the finding. We update it with a citation, retract it, or stand by it with the full audit trail, within five business days, logged in the report's dateModified field.
See the pipeline working on your competitors
Add your competitors and get your first verified Monday briefing, with sources and audit trails, within a week.