Get the latest news source for local headlines from Nebraska Note. Covering politics, education, crime, health and more.

Probably Genius Publishes Its Benchmark Method So Anyone Can Rerun It

Nov 6, 2026

Probably Genius has published the method behind its benchmark of 169 professional-service articles from 45 domains. The average score was 45.7 and none reached the floor of 80. The corpus rules, judge protocol and logged cost are public, so anyone can rerun it.

-- Probably Genius, the St. Petersburg AI visibility company, has published the full measurement method behind its professional-service content benchmark, including the corpus rules, the judge protocol, and the cost of the run. The company scored 169 professional-service articles drawn from 45 domains across nine niches against the Integrity Gate, the scoring instrument its own client writing must clear before it publishes. Not one article reached the floor. The full method and results are now public.

The company opened the Integrity Gate as a free tool on August 27. What was not published alongside it, in the detail needed to check the work, was how the measurement itself was run.

The rules are plain. The corpus was capped at eight articles per domain, so no single publisher could shape the result. Every article was validated live before it was scored. Each article was graded 0 to 100 across five dimensions (intent match, information gain, entity density, citation quality, and extractability), with a per-section check beneath the headline score. Three independent judge runs were taken and the median stands, under a standing rule inside the company: the judge is never the writer. The entire program, including calibration, judge selection, the 169-article corpus, and the company's own portfolio scored on the same day, cost $22.44 in judge fees, logged.

The results have not changed since publication. The 169 articles averaged 45.7, with a median of 46. Zero reached 80. The strongest article in the field scored 78 on a median of three judge runs. And 142 of the 169, roughly 84 percent, failed the per-section check regardless of their overall score. The weakest dimension across the market was citation quality.

"Publishing a number is easy, and it is worth very little on its own," said Christopher Shaw, co-founder and chief of ops at Probably Genius. "We put the method out because the only useful version of this research is the one somebody else can run. The corpus rules are there, the judge protocol is there, and so is the bill. If a firm thinks its own writing would score better than the field, the instrument is public and so is the procedure. Go and check us."

The company applies the same instrument to its own work. Its public standard is that client articles are scored 0 to 100 through the Integrity Gate, and nothing publishes under 80.

Read the 169-article benchmark study. The Integrity Gate tool produces the same score for any article.

About Probably Genius

Probably Genius LLC is a done-for-you AI visibility operator for established, expert-led businesses. Founded in late 2025 by Jacquie Baker and Christopher Shaw, the company maps what a business actually knows, structures that expertise so AI systems can verify it, and builds the record that makes the business recommendable. Probably Genius is headquartered in St. Petersburg, Florida, in the Tampa Bay area, serves clients in Austin, Texas, and works with expert-led businesses across the United States. Learn more at probablygenius.com.

Media Contact

Christopher Shaw, Co-Founder / Chief of Ops
Probably Genius LLC
[email protected]

Contact Info:
Name: Christopher Shaw
Email: Send Email
Organization: Probably Genius LLC
Phone: (737) 353-5750
Website: https://probablygenius.com

Release ID: 89202884

Should you identify any discrepancies, concerns, or inaccuracies in the content provided in this press release or require assistance with a press release takedown, we strongly urge you to notify us promptly by contacting [email protected] (it is important to note that this email is the authorized channel for such matters, sending multiple emails to multiple addresses does not necessarily help expedite your request). Our responsive team is committed to addressing your concerns within 8 hours by taking necessary actions to resolve identified issues diligently or guiding you through the necessary steps for removal. Our dedication lies in providing accurate and reliable information.