B2B buyers who are thinking about outsourcing sales now ask search engines and AI assistants full questions, not keywords. In the three months to mid-November 2026, 488 of the 1,000 queries that showed itsalesaas.com in Google Search were six words or longer, with an average of 13.5 words. They asked about in-house versus outsourced teams, AI agents versus people, cost, and whether cold email is legal.
This post shares that first-party data: how we collected it, the five question themes, how long queries behave compared with short ones, which queries were clearly written by machines, and what it means for sales leaders at IT and SaaS companies. Outside figures come from Google, Pew Research Center, Gartner, 6sense and the original GEO research paper, each linked where it is used.
How we collected the data
We exported the query table of our own Search Console property for the last three months (data to mid-November 2026). Search Console lists up to 1,000 queries per view, sorted by clicks and impressions, and leaves out rare queries, so this is the visible top of our search demand, not every search. We counted words in each query and grouped the queries by theme with simple keyword rules.
Two details from Google's documentation matter for reading the numbers. First, Google states that sites appearing in AI features such as AI Overviews and AI Mode are included in the overall search traffic in Search Console, inside the Performance report under the Web search type, as explained on Google Search Central's page on AI features. Second, for position Search Console reports the topmost position of your property in the results, and two appearances on one results page count as one impression, according to Google's Search Console help on the Performance data.
So a long query in our data can come from classic search, from AI Overviews or from AI Mode. Search Console does not split them, and we do not claim it can. What the data does show is what people typed, how often our pages appeared, and where.
Key findings at a glance
The short version: long queries were almost half of our visible demand, they ranked far higher than short ones, and they produced no clicks at all in the listed rows.
| Measure | Short queries (1 to 5 words) | Long queries (6+ words) |
|---|---|---|
| Number of queries (of 1,000) | 512 | 488 |
| Share of impressions | 52% | 48% (15,837 of 32,801) |
| Average position, weighted by impressions | 33.5 | 6.1 |
| Clicks in the listed rows | 26 | 0 |
| Phrased as a question | not counted | 311 of 488 |
One outlier needs a note. A single question about which agentic sales tools early stage companies can use had 9,974 impressions on its own. Even without it, long queries still made up about a quarter of all impressions. By length, 317 queries had one to three words, 195 had four or five, 217 had six to nine, 224 had ten to nineteen, and 47 had twenty words or more. 389 of the long queries were in English, 57 in Spanish, 26 in Czech and 16 in Turkish.
What do B2B buyers ask about sales outsourcing?
Five themes cover most of the long queries. A query can belong to two themes (for example cost and outsourcing), and the outlier above is left out of this table.
| Theme | Long queries | Impressions | Average position | Typical question |
|---|---|---|---|---|
| Outsource or hire in-house | 202 | 2,551 | 11.0 | when to outsource for saas startup |
| AI agents and people in sales | 185 | 2,151 | 6.6 | how should you split the sales cycle between ai agents and human reps |
| Cost and pricing models | 98 | 1,447 | 7.4 | how much does it cost to outsource sales |
| Checking AI recommendations | 24 | 270 | 6.2 | which tool checks if ChatGPT recommends my company |
| Is cold email legal | 15 | 338 | 5.4 | is cold email legal under gdpr in 2026? |
Theme 1: should we outsource sales or hire in-house?
The biggest theme is the build or buy decision. Buyers ask when a SaaS startup should outsource, whether to make a first sales hire or use outsourced sales development, and how a lead generation agency differs from an outbound agency. These are decision questions, asked before anyone contacts a vendor.
The answer depends on three things: how proven your sales motion is, how fast you need pipeline, and whether you can manage a sales hire well. Our outsourced SDR vs in-house SDR comparison puts ramp time, cost and risk side by side with a calculator, and our page on lead generation agency vs sales outsourcing explains where each service stops. If the question is which kind of partner to pick, those two pages answer it in more depth than any short reply can.
Theme 2: AI agents or human reps?
The second theme is AI in the sales team. Buyers ask whether AI sales agents will replace human reps, how teams are using AI instead of hiring more SDRs, whether an agent can handle everything a rep does after a meeting, and how to split the sales cycle between AI and people.
The pattern in these questions is practical, not philosophical: people want to know which tasks to hand over. Research, list building, first drafts and meeting notes move to software fast. Discovery calls, objection handling, negotiation and the relationships that close deals stay with people. We break this down task by task in AI sales agents vs human reps, and the economics of an AI SDR against an outsourced team are in AI SDR vs outsourced sales.
Theme 3: how much does it cost to outsource sales?
Cost questions are specific. Buyers compare pricing models (retainer, commission, cost per lead, pay per meeting), ask for monthly retainer options, look for outsourced SDR pricing per month, and ask how to compare quotes that are built differently. Several of these queries name the year, which tells you buyers expect current numbers.
This is where a public price list pays off. A buyer who asks an assistant what outsourced sales costs gets an answer built from pages that state prices. Our guide to B2B sales outsourcing cost in 2026 and the comparison of sales outsourcing pricing models both give ranges with sources, and our own plans are on the pricing page.
Theme 4: is cold email legal in Europe?
A smaller but very consistent theme is law. Buyers ask whether cold email is legal under GDPR, how the rules work in a given country such as Poland or France, and whether using a B2B company database for cold outreach is allowed. These queries had the best average position of any theme in our data, 5.4.
The short answer is that the rules differ by country, because the ePrivacy rules on electronic marketing are national laws, while GDPR governs the personal data itself. Our country-by-country guide, is cold email legal in Europe, is the most visited page on the site for exactly this reason. For your own campaigns, check the details with your own counsel.
Not every long query is a person
Some long queries were clearly written by software, and it is worth saying so. We saw queries that started with a list number (a question prefixed with "49."), queries packed with search operators that exclude a dozen social sites, a 325-word prompt that described a sales persona before asking its question, and the same question repeated in Czech, Spanish and Turkish with small changes.
These look like AI visibility tracking tools and research agents running sets of prompts, not buyers typing. That changes how you should read impressions: a query with high impressions and zero clicks may be a monitoring tool, not market demand. It also confirms that machines now read and compare your pages on behalf of people. The questions they run are still useful, because tracking tools are usually set up to mirror real buyer questions.
Why these queries rarely turn into clicks
Our long queries ranked around position 6 and still produced no clicks in the listed rows. Outside research points the same way. In a study of 68,879 Google searches by 900 US adults, Pew Research Center found that users who saw an AI summary clicked a traditional result link in 8% of visits, against 15% for those who did not, and clicked a link inside the summary in just 1% of visits.
Google itself said at I/O 2025 that people are asking more of their questions, "including more complex, longer and multimodal questions", in its post AI in Search: going beyond information to intelligence. Long questions get long answers on the results page, and many buyers stop there.
Buyers are also using AI for vendor research specifically. In a Gartner survey of 645 B2B buyers, buyers used an average of seven information sources in a recent purchase and 45% said they used GenAI, mainly to gather information on vendors and products, while 69% prefer to validate AI-generated insights with sales reps (Gartner press release, 20 May 2026).
What this means for sales teams at IT and SaaS companies
If a buyer's questions are answered before they visit you, the shortlist forms without you. The 6sense 2025 B2B Buyer Experience Report, based on nearly 4,000 buyers, found that buyers first contacted sellers about 61% of the way through their journey, and that in 95% of purchases the winning vendor was already on the day-one shortlist. Five changes follow from that:
- Answer the real questions in public. Take the questions your prospects ask on calls and in AI tools, and publish direct answers of 40 to 60 words at the top of each section, followed by detail.
- Publish numbers. Prices, ranges, timelines and terms are what cost questions need. Pages without numbers are hard for an assistant to quote.
- Cite sources and add data. The original GEO study by Aggarwal and colleagues (arXiv, accepted to KDD 2024) reports that optimisation methods such as adding citations and statistics can boost visibility in generative engine responses by up to 40%.
- Prepare reps for informed buyers. Buyers arrive with an AI answer and want it checked. Train the team to confirm, correct and add context, not to repeat the basics.
- Measure the right thing. Track impressions and positions for question-style queries in Search Console, check a fixed set of prompts in the assistants every month, and do not treat zero clicks as zero value.
Frequently asked questions
What counts as an AI-style query in Search Console?
There is no official label. In this analysis we treated queries of six words or more as long, question-style queries. Most of them read like prompts typed into an assistant: full sentences, often ending with a question mark. Search Console counts AI Overviews and AI Mode inside the Web search type, so it cannot tell you which feature showed your page.
Why do long queries rank well but get no clicks?
Long, specific questions have less competition, so a focused page can rank in the top ten. But they are also the questions where Google tends to show a generated answer, and Pew Research Center found that users click far less when an AI summary appears. Some long queries also come from monitoring tools, which never click.
How can a B2B company check what AI assistants say about it?
Write down 10 to 20 questions your buyers ask, run them in ChatGPT, Gemini, Perplexity and Google AI Mode, and record whether you are mentioned and which sources are cited. Repeat each prompt a few times, because answers vary. Our guide on whether ChatGPT recommends your company walks through the process step by step.
Do buyers still talk to sales reps if AI answers their questions?
Yes, but later and better informed. Gartner found that 69% of B2B buyers prefer to validate AI-generated insights with sales reps, and 6sense found buyers first contact sellers around 61% of the way through their journey. The rep's job shifts from explaining basics to confirming and correcting what the buyer already believes.
How we use this data at IT SalesaaS
These questions shape what we publish and how we sell. The cost, legal and outsourcing questions above are answered on our blog with sources, and our pricing page lists the plans in plain numbers. When a client needs pipeline, we run outsourced lead generation and IT sales outsourcing for software and SaaS companies, and teams that also want to show up in AI answers can add search and AI visibility through Full GTM. If you want the background on the method, read what generative engine optimization is.
Sources and further reading: Google Search Central, AI features and your website; Google Search Console Help, Performance report: about the data; Google, AI in Search: going beyond information to intelligence (May 2025); Pew Research Center, Google users are less likely to click on links when an AI summary appears (July 2025); Gartner, survey of B2B buyers on AI-generated insights (May 2026); 6sense, The B2B Buyer Experience Report for 2025; Aggarwal et al., GEO: Generative Engine Optimization (KDD 2024).
