Perceptron DataWeekly checks against known answers
Every Sunday an automated watchdog runs the live actors on Apify with fixed inputs whose correct answers are known, and checks the content of the results — not only whether the run succeeded. The results are published here as they are, failures included.
| Actor | Last check | Sample against the original source |
|---|---|---|
| Research Agent | 12 Oct 2026: 10 / 10 | 93.1% (108 / 116) |
| PDF to Markdown | first check on 25 Oct 2026 | 69.5% (73 / 105) |
| Advertiser Check (Google & Bing) | first check on 25 Oct 2026 | – |
| Imprint & VAT Check | first check on 25 Oct 2026 | 82.9% (136 / 164) |
| Free Imprint & Contact Check | first check on 25 Oct 2026 | 87.4% (111 / 127) |
| Messe München Exhibitors | first check on 25 Oct 2026 | – |
| Koelnmesse Exhibitors | first check on 25 Oct 2026 | – |
| kununu Employer Data | first check on 25 Oct 2026 | – |
| Arbeitsagentur Jobs | first check on 25 Oct 2026 | – |
| App Store & Google Play Data | first check on 25 Oct 2026 | – |
Not in the content check yet: MaStR Energy Assets, Data Source Finder, EUDAMED Medical Devices, TED Tenders. They are run once a week with an empty input like Apify's own test (does it run and return data?), but their results are not checked against known answers.
What this shows: on the day of the check, the live actor gave the known answers for these inputs. It does not prove that every result is correct. The sample against the original source measures accuracy on real result rows.