Alan Finney – Notes on Verified AI and Maintenance Software

A follow-up to How B2B Software “Awards” Get Sold — And Why I’m Not Buying

---

Three months ago I wrote about someone trying to sell me an award for $3,000, and I made a general claim I couldn’t fully back up at the time: that B2B software credibility is bought, that buyers can’t tell, and that the whole thing amounts to a decision-fatigue tax.

I’ve since found something specific. It isn’t an opinion, it’s a transaction, and it’s on the public record.

The thing you probably missed

On January 29, 2026, G2 announced it was acquiring Capterra, Software Advice, and GetApp from Gartner. The deal closed on February 5, 2026, for approximately $110 million, according to Gartner’s own SEC filing.

The combined business holds roughly six million reviews and reaches a stated 200 million-plus annual software buyers.

Read that again slowly, because the implication takes a second to land.

G2. Capterra. Software Advice. GetApp. One company.

Now go look at a vendor’s website

Pick any B2B software vendor. Scroll to the footer, or the “recognition” strip halfway down the homepage. You will very often find something like this:

• 🏆 G2 Leader — Spring 2026
• 🏆 Capterra Shortlist
• 🏆 GetApp Category Leader

Three badges. Three logos. Three apparently separate organizations, independently arriving at the same favorable conclusion.

They are three badges from one company.

I want to be careful here, because this is the part where it would be easy to overreach. I am not saying those badges were bought. I’m not saying the underlying reviews are fake. I’m saying something narrower and, I think, harder to argue with:

The appearance of independent corroboration is gone, and nobody told the buyer.

If you were doing your job properly as a software buyer — cross-referencing multiple sources, looking for consensus across platforms before shortlisting — you were, as of February, reading the same house twice. Possibly three times. The diligence still felt like diligence. It just stopped being triangulation.

This isn’t a scandal. That’s what makes it interesting.

Consolidation is normal. Companies buy other companies. Gartner sold a volume business it presumably didn’t want; G2 bought scale and datasets. Nobody did anything wrong, and I’d have made the same call in either chair.

The problem isn’t the transaction. The problem is that a signal changed meaning and nothing downstream updated.

Every badge minted before February meant one thing. Every badge minted after means something slightly different. The badges look identical. The vendor pages weren’t rewritten. The buyer’s mental model — “well, it’s on G2 and Capterra” — quietly stopped being true, and there was no mechanism anywhere in the system to tell them.

That’s the actual disease I was circling in May. Not corruption. Signal decay that nobody is responsible for announcing.

Exhibit two, from my own inbox

Two weeks ago I received a monthly performance report from Software Finder about my company’s free listing. Here it is, essentially in full:

“Your profile has been getting good traction this month — with steady visitor activity and new leads! Don’t miss the opportunity to unlock even more value. Upgrade to Premium and gain access to: Leads — view and connect with interested buyers…”

Notice what isn’t there.

Not a single number. Not one. No view count. No lead count. No conversion figure. “Good traction.” “Steady activity.” “New leads.”

In May I listed seven tells of a pay-to-play pitch. Tell number seven was “no verifiable distribution claims.” I wrote that about a trade magazine. Three months later a completely different operator, in a completely different format, reproduced it exactly.

Which tells me it isn’t a quirk of one bad actor. It’s structural. The vagueness is doing the commercial work: if the email said “four views and zero leads this month,” nobody upgrades. The quantity has to stay hidden, because revealing it is what you’re being charged for.

There is now an actual standard, and almost nobody mentions it

Here’s something that changed while this industry kept doing what it does. The FTC’s Consumer Reviews and Testimonials Rule (16 CFR Part 465) has been in force since October 21, 2024. It addresses fake reviews, review suppression, and manipulation. Separately, the updated Endorsement Guides require that incentivized reviews disclose their material connection “clearly and conspicuously.”

Some platforms take this seriously — G2 and TrustRadius have labeled incentivized reviews for years, and credit where it’s due.

But notice that we now have something we didn’t have during the last twenty years of this argument: a published, federal, checkable standard. Which means “is this platform trustworthy?” stops being a matter of taste and becomes a question with evidence attached.

That’s a much better fight than the one I was having in May.

What I’d actually tell a buyer

Not “don’t use review sites.” I use them. They’re genuinely useful, and the reviews on them are mostly real people describing real software.

Three things instead:

Count your actual sources, not your logos. If three of your five signals now trace to one owner, you have fewer independent data points than you think. That’s not a reason to discard them — it’s a reason to weight them once, not three times.

Ask what the number is. When a platform tells you your listing has “good traction,” ask for the figure. The answer, and whether you get one, is itself the finding.

Check the methodology page. Not whether you agree with it — whether it exists, and whether it says what’s paid and what isn’t. Absence is data.

A note on my own conflicts, since I’m asking for yours

I do sales for a CMMS vendor, and I hold stock options in it. That means I have a direct financial interest in one corner of this market, and you should weigh anything I say about maintenance software accordingly. I’ll say so again anywhere it’s relevant, rather than making you go looking for it.

That’s the whole ask, really. Not neutrality — nobody in this industry is neutral, including the platforms. Just tell people where you’re standing when you point.

Same as May: I don’t think every paid placement is a scam, there are good actors and bad actors and a lot in between, and there is no clean revolution here — only a slow drift toward formats that reward depth over decoration.

But the drift is real. And it’s easier to see now that four of the referees turned out to be wearing the same shirt.

---

Facts in this post: G2’s acquisition of Capterra, Software Advice and GetApp was announced January 29, 2026 and closed February 5, 2026 at approximately $110 million per Gartner’s SEC filing. The Software Finder email is quoted verbatim from my inbox, dated July 29, 2026. 16 CFR Part 465 took effect October 21, 2024. Corrections welcome — I’d rather be right than consistent.