How Do I Find Out Whether ChatGPT and Other AI Assistants Recommend My MSP?

By Anne Mitchell, Founder and CEO, ChannelSpring. Anne has spent 25 years in the technology and telecom channel, including leadership roles at a Fortune 100 distributor, and runs monthly AI visibility research for MSPs.

You find out by asking the assistants the questions your buyers ask, in your buyers' words, on a fixed schedule, and recording which companies they name and which pages they cite. A single search on a single afternoon tells you very little, because AI answers change from day to day. ChannelSpring runs the same buyer prompts across ChatGPT, Gemini, Perplexity and Claude every month for this reason, and reads the pattern over time.

Picture the operations director at a 120-person manufacturer in your metro area. She asks ChatGPT which managed IT providers nearby can support a growing team and pass a cyber insurance review. The assistant names three companies. If yours is one of them on most days, you are on her shortlist before she ever visits your website. If it appears on one day out of ten, you need to know that too.

AI assistants now shape the vendor shortlist

Buyers use these tools to research vendors today. According to Gartner (2026), 45 percent of B2B buyers surveyed used generative AI during a recent purchase, primarily to gather information on vendors and products. The same survey of 645 buyers found that 69 percent prefer to validate what AI tells them with a sales rep, which means the assistant often decides who gets that validation call.

A single AI search gives an unreliable answer

AI recommendations move constantly, so visibility has to be measured as a frequency. A study by Detailed (2026) checked about 1,300 commercial prompts every day for 28 days and collected more than 70,000 responses. Two random days produced the identical list of brands only 0.3 percent of the time in ChatGPT and 1.1 percent of the time in Google AI Overviews.

The same study found stability at the top. For a typical prompt, the most-named brand appeared on 82 percent of days in ChatGPT and 89 percent in AI Overviews. The practical goal for an MSP is to become one of those consistently named companies for the questions its buyers ask, and the only way to see progress toward that goal is repeated measurement.

Which prompts should an MSP track?

Track the questions your buyers type, written from their role and situation. A useful set holds 15 to 25 prompts across three groups:

• Category prompts, such as "Which managed IT providers in Denver work with 100-person law firms?"
• Problem prompts, such as "How do I get IT support that meets my cyber insurance requirements?"
• Comparison prompts, such as "Is co-managed IT or fully managed IT better for a company with one internal IT person?"

To find the questions, ask the assistants directly. ChannelSpring's AEO research is ChannelSpring's monthly buyer-question research that asks ChatGPT, Gemini, Perplexity and Claude for the 20 questions a defined buyer has been typing, then ranks each question by how many of the four assistants returned it. In ChannelSpring's October 2026 research on MSP owners deciding how to build a marketing function, six questions came back from all four assistants, and one of them was the question this article answers.

How to run an AI visibility check

Follow the same steps every time so the results stay comparable:

1. Use a clean session for each assistant, such as a temporary or incognito chat, so your own account history stays out of the answer.
2. Turn on web search where the assistant offers it, so answers reflect current web content.
3. Enter each prompt with identical wording every time.
4. Record every company named, the order it appears in, and every web page cited.
5. Repeat on a fixed schedule: monthly for the full set, and weekly for your five most important prompts.
6. Calculate the share of runs that name your company, and track the same figure for your three closest competitors.

A shared spreadsheet with one row per run is enough to start. The trend line across several months is the number to watch.

Free reports from Google and Microsoft

Two free reports show when your pages appear in AI answers. The Google Search Console Generative AI performance report shows impressions for links to your site in AI Overviews and AI Mode, and Google states it rolled out to all websites worldwide as of August 31, 2026. Smaller sites may see little data at first.

Microsoft's Bing Webmaster Tools AI Performance report shows where your content is cited in Copilot and Bing AI answers. According to the Microsoft Bing Blog (2026), a June 2026 update added Citation Share, which shows your site's percentage of all citations for a given query.

Here is how the three methods compare:

• Manual prompt tracking shows which companies each assistant names for your buyers' questions and which pages it cites. It measures presence, so run it often.
• Google Search Console's Generative AI report shows impressions for your pages in AI Overviews and AI Mode. It leaves out whether the answer named your company.
• Bing Webmaster Tools' AI Performance report shows citations of your content in Copilot and Bing AI answers, including Citation Share. It leaves out ChatGPT, Gemini, Perplexity and Claude.

What to do with the results

Each pattern points to a specific next step:

• Your company is named often. Keep the cited pages current and accurate, and refresh their statistics and examples each quarter.
• Your pages are cited, but your company is unnamed. Put your company name beside the claims on those pages, so the name travels with the passage the assistant extracts.
• Competitors are named and you are absent. Open the pages the assistants cite for them. Note the case studies, reviews, directory profiles and specific service pages they publish, then build your own versions. What does it take to get cited by AI search tools? covers the content that earns citations.
• Nobody in your market is named consistently. The question is open, and a clear, specific answer page can claim it.

For the buyer's side of this process, read What happens when a buyer asks ChatGPT who to call? Once you know which questions matter, the answers belong in your blog, your FAQ page and your service pages.

Where to start this month

Write ten prompts this week, run them in all four assistants, and record the results. Repeat the same ten next month. ChannelSpring builds and runs this research every month for growth-stage MSPs and turns the results into the content that earns citations. Contact ChannelSpring to see where your MSP shows up today.

About the author. Anne Mitchell is the Founder and CEO of ChannelSpring, a fractional CMO practice for MSPs, MSSPs, VARs and IT consultants. She has 25 years of experience in the technology and telecom channel, including leadership roles at a Fortune 100 distributor and work with IBM, Verizon, Oracle and VMware. Connect with Anne on LinkedIn.