Who AI Cites for
Developer-Tools GTM
We checked 30 of the queries that define the developer-tools go-to-market category. Google now shows an AI Overview on 90% of them, and the source it cites most is not a blog or a vendor. It is YouTube, by a wide margin.
By Daria Dovzhikova · Updated June 2026
The finding
- 27 of 30 category queries (90%) returned a Google AI Overview. The category is effectively AI-Overview-saturated.
- YouTube is the most-cited source: 17 of 27 Overviews, more than LinkedIn (6) and Reddit (4) combined.
- Third-party and social sources dominate; company blogs are cited least.
- Vendor pages get cited when they own a clean definition for a specific term, not through volume.
Most-cited sources in category AI Overviews
Across the 27 queries that returned an AI Overview, these domains appeared most often in the Overview's reference set:
Source: live Google SERP via DataForSEO, US, top-20 depth, 30 queries, collected 2026-06-29. Counts are appearances across 27 AI Overviews. gtm-labs.co appeared zero times, which is exactly the gap this research exists to close.
Three takeaways for developer-tools GTM
- Video is the most underrated GEO channel here. YouTube is cited more than the next three sources combined. A short, specific explainer on a category term is more likely to be pulled into an AI Overview than another text post, and almost no devtools marketers are making it.
- Named practitioners punch above their domain authority. Individual practitioner blogs and developer publications appear in the citation set despite minimal traditional SEO authority. AI Overviews reward extractable, first-hand, named expertise, the exact lane a focused operator can win.
- Vendor hubs win on term ownership, not volume. The vendor pages that get cited each own a clean definition for one term. Pick the term, own the definition, structure it for extraction.
This is the practical core of AI search optimization: be the extractable, first-hand source on a term you can own, and be present on the surfaces AI actually cites. It is the same discipline behind the developer-tools GTM machine and the agentic GTM stack.
September 2026 update: the answer engines themselves
The benchmark above is Google AI Overviews only. Since July I have also run the same 32 developer-tools GTM prompts, eight per platform, directly against ChatGPT, Perplexity, Gemini, and Claude each month, checking whether this site is cited. Four cycles in, the pattern is consistent enough to publish.
- Perplexity cites, and it cites on buying questions. Three of eight prompts in September, all of them who-do-I-hire phrasings ("who offers fractional GTM engineering for developer tools", "best fractional product marketing consultants"). It has cited this site every cycle since July.
- Concept prompts get no citation anywhere. Go-to-market for an AI agent product, marketing an MCP server, growth loops for developer tools: zero citations across four platforms, two cycles running, even though the pages exist. The Google AI Overviews for those terms have empty reference lists. With nothing third-party to corroborate, the engines answer from training and cite no one.
- ChatGPT went from one citation to zero and stayed there. Two cycles at zero of eight. ChatGPT's browsing runs on Bing's index, which is a different thing to be present in than Google's; that is the first thing I am checking.
Two findings from other people's data line up with this. SE Ranking analysed 2.5 million pages in August 2026 and reported that what AI cites "isn't llms.txt"; the file is table stakes, not a citation lever. And Tim Soulo of Ahrefs made the point that there is no such thing as prompt search volume, so the metric that matters is citation share on the prompts your buyers actually type, measured directly, which is what the monthly probe does and what an AI-visibility experiment card in the Growth Pilot is measured on.
Method and limitations
We ran 30 developer-tools and go-to-market queries (developer marketing, gtm engineer, product-led growth, go-to-market for ai agents, agentic gtm, how to market an mcp server, fractional product marketing, and more) through a live Google SERP pull via DataForSEO, US location, top-20 depth, on 2026-06-29. For each query we recorded AI Overview presence and aggregated the domains in each Overview's reference set.
This is Google AI Overview data only, US, and a single point in time. It does not directly measure ChatGPT, Perplexity, Gemini, or Claude citations, and AI Overview composition shifts with phrasing and over time. The value is directional and comparative: it shows which source types the category's AI answers are built from. We re-run it quarterly to track the trend.
Benchmark reuse terms: permission required
Version 1 · October 2, 2026
These terms apply only to GTM Labs' original compilation and presentation of the 30-query Google AI Overview citation benchmark collected on 2026-06-29 and published on this page. Reproducing, redistributing, or adapting that protected material requires prior written permission from GTM Labs, for commercial or noncommercial reuse. No general reuse license is granted by publishing the benchmark.
These terms do not restrict ordinary linking, citation, or uses permitted by applicable law. They do not claim exclusive rights over underlying facts or third-party material, which remains subject to its own rights and terms.
To request permission, email dovzhikova@gmail.com with the material you want to reuse and the intended use. Wait for written permission before proceeding with reuse that requires it; sending a request does not grant permission.
FAQ
How often do AI Overviews appear for developer-tools GTM queries?
In this benchmark, 27 of 30 category queries (90%) returned a Google AI Overview. For practical purposes the developer-tools and go-to-market category is fully AI-Overview-saturated: if you are not in the answer, you are losing the click before the organic results load.
Which source does AI cite most for GTM topics?
YouTube, by a wide margin. It appeared in 17 of the 27 Overviews, more than LinkedIn (6) and Reddit (4) combined. For a category most B2B marketers treat as text-and-blog, the AI answer layer is disproportionately built from video and social.
Does AI cite company blogs or third-party sources more for GTM?
Third-party and social sources dominate. YouTube, LinkedIn, and Reddit lead the citation set. Vendor and practitioner pages (Salesforce, Highspot, Amplitude, Go Fractional, and individual practitioner blogs) appear mainly when they own a clean definition page for a specific term, not through publishing volume.
How was this measured?
A live Google SERP pull via DataForSEO across 30 developer-tools and go-to-market queries (US, top-20 depth, collected 2026-06-29). For each query we recorded whether an AI Overview appeared and aggregated the domains in each Overview's reference set. It is Google AI Overview data only and a single point in time; ChatGPT, Perplexity, Gemini, and Claude citations require separate panels.
Related guides
Ready when you are.
Discovery calls are 30 minutes. First one's on me.