B2B Services / Appointment Setting · engaged July 2026, ongoing · measured July 7 to August 22, 2026
The client
GenSales books sales meetings for B2B companies using US-based callers, and has since 2002. Their buyers are owners and sales leaders at HVAC, roofing, manufacturing, insurance, and healthcare companies who now type questions like “appointment setting companies with US-based callers” and “who books sales appointments for commercial HVAC businesses” into ChatGPT and Google before they ever open a browser tab.
A 24-year-old company that AI could not see
In July 2026 we asked the AI engines the questions GenSales’ buyers ask. On 11 of the 14 prompts we tracked, GenSales did not appear once. Share of voice across the set: 0.5%. Belkins, SalesRoads, Callbox, CIENCE, and Martal Group held the rest.
It was not a track record problem. It was three fixable problems stacked up:
- Their own content was competing with itself. Several blog posts answered the same question. When two pages say the same thing, AI engines often cite neither.
- The brand was split three ways. The main site and two separate landing-page domains looked to an engine like three small companies instead of one established one. One of those sites, the roofing one, had no analytics tag at all. Leads were arriving there and nobody could measure it.
- A different company was wearing their name. An unrelated “Gen Sales” that recruits door-to-door reps owns the clean LinkedIn slug. The engines blended the two.
What we changed
We spent the first week reading everything they had already published, and then we rewrote instead of adding. Six articles rebuilt, two refreshed, four new. One question, one page, a real named author with a real LinkedIn. Schema on the main site. The missing analytics tag installed. The brand name made identical everywhere we could reach. Clutch profile verified and reviews requested. Reddit comments in the threads the engines cite for this category. First outreach to the publications already ranking for “best appointment setting companies.” All of it running at the same time, because AI answers assemble from many sources at once.
What happened, in the tracking we set up on day one
We tagged the original 14 prompts before any work started, so this is the same list, measured the same way.
- 75%. “Companies that book sales appointments for commercial HVAC businesses” now names GenSales in three of every four AI answers.
- 0% to 48% on “top-rated US appointment setters for commercial insurance brokers.”
- 0% to 36% on “outsourced B2B SDR services for the healthcare industry.”
- 0% to 20% on “best lead generation services for commercial roofers.”
- From 3 of 14 prompts to 8 of 14. Share of voice across the set went from 0.5% to 4.1% against the five largest names in the category.
- AI engines now pull GenSales content as a source about 15 to 17% of the time on tracked prompts. In early July it was about 1%.
When the commercial HVAC number came up on the August 12 call, Erin Leemon, GenSales’ Operations Director, said it for us: “We’re happy about that.” A week later:
“We’re already seeing a difference with the little bit that we’ve done.”
Erin Leemon, Operations Director, GenSales
What we are not saying
We are not telling you this made the phone ring, because we cannot prove that yet and neither can GenSales. Their inbound roofing leads are up since July; roofing was always their best inbound channel. ChatGPT sessions in their Google Analytics went from 6 to 15 across comparable 30-day windows, which is real and small. Visibility moves first. We publish the first step because it is the one you can verify.
Why it worked
GenSales did not need more content. They needed the content they had to stop competing with itself, a name AI engines could match to a single company, and third-party surfaces that said the same thing their site said. Fix those and a 24-year track record finally counts. That is what the free AI visibility audit measures first for any business: where the engines cannot identify you, cannot verify you, or have nothing of yours worth quoting.