Last updated: June 2026

Most MSPs checking their AI search visibility run one prompt in ChatGPT, see their company name, and call it a day. That single test is worth almost nothing.

SparkToro and Gumshoe.ai ran 2,961 prompts across ChatGPT, Claude, and Google AI with 600 volunteers in January 2026. The same prompt returned the same brand list less than 1% of the time. Same list in the same order? Less than 0.1%. One screenshot proves nothing.

But here’s what the study also found: while individual responses are essentially random, frequency of appearance across many runs is stable. You can’t track your “position.” You can track whether you’re in the consideration set at all. That’s a different measurement problem, and it requires a different framework.

This post gives you the exact framework. Specific prompts to run. GA4 setup you can finish in 15 minutes. Tools that are actually worth the money at MSP scale. And a scorecard you can hand to your partners or board without a 30-minute explainer attached.

Why Can’t You Just Check Your AI Ranking Like You Check Google?

AI doesn’t rank pages. It synthesizes answers from multiple sources, and the output changes with every query, making traditional position tracking meaningless.

Google gives you a stable number. Position 4 on Tuesday, probably position 4 on Wednesday. You can screenshot it. You can build a report around it. AI broke that contract entirely.

Language models are probabilistic. They generate answers by predicting the most likely next word in a sequence, drawing from training data, retrieved web content, and conversation context. The same question asked twice produces two different answers with two different brand lists. Superlines found that 93% of AI search sessions end without a website click, which means most of your AI visibility isn’t producing traffic you can measure in analytics at all.

That doesn’t mean measurement is impossible. It means you need different metrics. Position is out. Frequency is in.

What Should an MSP Actually Track for AI Visibility?

Track four things: citation frequency, entity accuracy, competitive citation share, and AI-referred conversion rate. Everything else is noise at MSP scale.

These four KPIs map directly to the questions your leadership actually cares about. Are we showing up? Are we described correctly? Who’s showing up instead? And is any of this producing business?

Notice what’s not on this list. “AI ranking position.” You’ll see tools that claim to show you where you rank in ChatGPT. The SparkToro data makes that metric borderline meaningless for individual queries. Frequency across many queries is the signal. Individual position is noise.

One more KPI worth watching, even though it’s indirect: branded search volume. Multiple practitioners have observed that when AI platforms cite a brand frequently, branded search volume in Google Search Console rises 60-90 days later. If your branded searches are climbing and you’re doing AI visibility work, that’s a correlation worth tracking.

How Do You Run a Monthly AI Visibility Audit?

Run 15-20 prompts across ChatGPT, Perplexity, Google AI Overviews, Claude, Gemini, and Copilot. Score each on mention, accuracy, and competitor presence. Track changes month over month.

The whole audit takes about 30 minutes once you’ve set up the prompt list. First time through, budget an hour. After that, it’s mechanical.

The Prompts to Run (MSP-Specific)

Split your prompts into three categories. Ten each is the sweet spot for a monthly audit, per Passionfruit‘s research on audit methodology. Above 30, the marginal insight drops and the time cost climbs. Here’s what works for MSPs.

Entity recognition (does AI know you exist?)

• “What is [Your MSP Name]?” and “Tell me about [Your MSP Name].”
• “Who provides managed IT services in [your city]?”
• “What MSPs serve [your industry vertical] in [your region]?”

Visibility and consideration (are you on the shortlist?)

• “Best managed IT services providers in [your city]”
• “Compare MSPs in [your city] for a [50/100/200]-person company”
• “I need cybersecurity monitoring for my mid-sized business in [your region]. Who should I call?”
• “What should I look for when choosing an MSP for my [industry] company?”

Recommendation and preference (does AI pick you?)

• “If I could only call one MSP in [your city], which should it be?”
• “What’s the best MSP for a company migrating from on-prem to Azure in [your region]?”
• “[Your MSP] vs [Competitor Name], which is better for [use case]?”

What to Document for Each Response

For every prompt, on every platform, write down five things: whether you appeared at all (yes/no), the exact description the platform gave you (copy-paste it verbatim), which competitors appeared alongside you, whether any information was wrong, and whether a link to your site was included.

Save the verbatim descriptions. Over time, the drift in those sentences is your “AI brand drift” indicator. If Perplexity called you a “full-stack managed IT provider” last month and now calls you a “cybersecurity vendor,” something in the content ecosystem shifted. Catch it early.

How Often to Run It

Monthly for most MSPs. If you’re actively publishing content and building AI visibility, bump the entity recognition prompts to biweekly. The consideration and recommendation prompts need at least 30 days between runs for meaningful pattern changes.

How Do You Set Up GA4 to Track AI-Referred Traffic?

Create a custom channel group in GA4 using regex filters for AI platform domains. The native AI Assistant channel launched May 13, 2026, but it only captures ChatGPT, Gemini, and Claude. Perplexity and Copilot still fall through to Referral or Direct.

Bigger problem: between 35% and 70% of AI referral sessions arrive without referrer data. Those land in Direct traffic, invisible. Your AI traffic number is always an undercount.

Here’s the 15-minute fix.

Go to Admin, then Data Display, then Channel Groups in your GA4 property. Create a new channel group (don’t edit the default). Add a rule for AI traffic using this regex on Session Source:

chatgpt.com|chat.openai.com|perplexity.ai|claude.ai|gemini.google.com|copilot.microsoft.com

Name the channel “AI Search.” Place it above the Referral rule in the priority list. GA4 evaluates rules top-down, so if AI Search sits below Referral, it never fires.

Why bother if the numbers are small? Because the visitors who do arrive from AI platforms convert at wildly higher rates than organic search. Opollo’s 2026 benchmark of 312 IT and tech service firms found AI visitors converted at 14.2% compared to 2.8% for organic. Seer Interactive reported Perplexity traffic converting at 10.5% versus 1.76% for Google organic. Small channel. Disproportionate value.

Which AI Visibility Tools Are Worth Considering?

This is a fast-moving category. No single tool is perfect, and most are priced for enterprise companies with six-figure marketing budgets. Here’s what’s available as of mid-2026, without endorsing any one over another.

Honest take for most MSPs: a free manual audit is probably the right starting point. You don’t need a $200/month tool to run 15 prompts in ChatGPT and write down what comes back. Start manual. If AI traffic shows up in GA4 and you’re actively publishing content, then consider whether a paid tool saves enough time to justify the cost.

One important caveat about accuracy: independent testing has found significant discrepancies between what some tools report and what’s actually happening in AI responses. Ahrefs Brand Radar, for example, reported 3 ChatGPT mentions for one brand when the actual count was 123. These tools are useful for trends and direction, not precision. Treat the numbers as signals, not gospel.

What Do Realistic AI Visibility Benchmarks Look Like at 30, 60, and 90 Days?

At 30 days, expect a baseline score and possibly first citations on one or two platforms. At 60, citation frequency should start climbing. At 90, look for sustained mention rates and competitive share gains.

Important context: these benchmarks assume you’re actively doing the work. Publishing content. Building entity signals. Adding schema. Fixing your robots.txt. If you’re just measuring without acting, the numbers won’t move.

Profound’s data shows the median time to first AI citation is 6.81 days, with 90% of cited pages earning their first citation within 37 days. If a page has been live for 6 weeks and hasn’t been cited at all, the problem is usually technical (robots.txt blocking AI crawlers, missing schema) rather than content quality.

Foglift‘s Q1 2026 study of 4,217 brands found that companies implementing structured data, FAQ content, authoritative backlinks, AI crawler access, and consistent entity references typically saw a 15-25 point improvement within 90 days. For MSPs, those signals are table stakes. If you haven’t fixed them, that’s your starting point before worrying about citation frequency.

Who should invest in AI visibility tracking right now? MSPs spending $2,000+/month on content and SEO. If you’re publishing regularly and don’t know whether AI platforms are picking up what you’re putting down, you’re flying blind. The audit takes 30 minutes a month. There’s no reason to skip it.

Who can wait? MSPs with fewer than 10 pages on their website and no active content program. Fix the foundation first. You can’t measure visibility you haven’t built yet. Come back to this after your site has real service pages, location-specific content, and at least 3 months of published posts.

How Do You Report AI Visibility to Leadership?

Build a one-page monthly scorecard with four sections: where you appear, how you’re described, who appears instead, and what business impact you’re seeing from AI-referred traffic.

Don’t overcomplicate this. Your partners or board don’t need to understand retrieval-augmented generation or probabilistic outputs. They need to know four things. Semrush’s reporting framework puts it well. Are we visible? Is the description accurate? Is our position improving relative to competitors? And what business results are improving as a result?

The AI Visibility Scorecard (monthly, one page)

Presence. “We appeared in X% of relevant AI responses this month, up from Y% last month. We were mentioned across [3/4/5/6] of the six platforms tested.”

Accuracy. “AI describes us as [exact description]. This is [accurate / partially accurate / inaccurate]. Last month we fixed [specific corrections] and accuracy improved from X% to Y%.”

Competition. “Our top competitor appeared in X% of responses versus our Y%. They were cited most frequently on [platform]. They are described as [their description], positioning them as [what].”

Business Impact. “GA4 recorded [X] AI-referred sessions this month. Those sessions converted at [X]%. Branded search volume is [up/flat/down] Y% since we started AI visibility work.”

That’s it. One page. Takes 20 minutes to compile if you’ve already run the monthly audit and your GA4 is set up. Skip the jargon. Skip the platform-by-platform breakdowns unless leadership asks for them. Lead with the business impact section when things are going well. Lead with the competition section when you need budget approval for more content investment.

Start Measuring Before You Need the Numbers

AI visibility measurement isn’t optional anymore. Not because every MSP needs an expensive tracking tool tomorrow. But because the MSPs who start building baseline data now will have 6 months of trend data when their competitors are still running one-off ChatGPT screenshots and calling it a strategy.

The practical version: run the 15-prompt audit this month. Set up the GA4 channel group this week (15 minutes). Build the scorecard template once and reuse it monthly. That’s enough to get started.

If you’re already doing MSP marketing strategy work and want to understand how AI visibility fits into the bigger picture, C4 Solutions builds AI-visible marketing systems for MSPs. If you want an outside perspective on where you stand, start with a free growth assessment.

Inquire About Your Area