My pipeline is dry
The signals that mean a buyer is actually in market
- Published: Sep 11, 2026
- Updated: Sep 30, 2026
- 6 minutes read
- Make us your go-to for GTM
Gartner found buyers spend just 17% of the purchase process meeting suppliers, so the signals you can observe are a thin slice of something happening mostly out of view. Most of what gets sold as intent is activity, and the same behaviour has a dozen explanations.
Someone downloaded your pricing guide. Your tool scored them as high intent and an alert went to a rep.
They were probably a competitor, a student, or somebody who wanted the number without talking to anyone. None of those is a buyer.
That is the intent problem. Most of what gets sold as a buying signal is activity, and activity is easy to measure and mostly meaningless.
Intent tooling sells observation. Someone from a target account visited three pages. True, and not something you can act on.
The same behaviour has a dozen explanations. Research for a project six months out. A junior analyst building a market map. Someone checking what their current vendor charges. All identical on a dashboard.
Gartner’s 2024 B2B Buying Survey found buyers spend just 17% of the purchase process meeting suppliers, and 75% say they prefer a rep-free buying experience. Roughly 80% of the journey is self-directed, which means the signals you can observe are a thin slice of a process happening mostly out of view.
Searches for buyer intent data have fallen 47% in the last year, and B2B intent data is down 58%. The question has not become less important. Confidence in the tooling has.
Roughly strongest to weakest. Each one has a reason it works and a specific response.
Your contact spent internal credibility introducing you to a colleague. Nobody does that for a vendor they are idly researching. Given that Gartner puts a complex buying group at six to ten people, each new name is also a real step toward the committee actually deciding.
Find out what the new person blocks on before the next call. They have different objections from your original contact and nobody has asked them yet. Our post on selling to six people breaks down what each role in the room cares about.
How long does setup take. What happens to our existing data. Who needs to be involved. These only matter to someone picturing the thing working. Feature questions can be curiosity. Implementation questions rarely are.
Answer it properly, then offer the next step in the same breath. This is the moment a call converts, and most reps answer the question and stop there.
A security document, a one-page summary, pricing in a forwardable format. They are not asking for themselves. They are assembling material to convince someone else, which means the evaluation has left their desk.
Send it in a form they can forward without editing. You are writing for the person who was not on the call.
Sometime next year becomes before the end of the quarter. Vague timelines describe a problem they are tolerating. Dated ones usually mean a deadline someone else set, and that deadline does your selling for you.
Ask what happens on that date. A real consequence means a real deal. No consequence means the date is aspiration.
It confirms an active evaluation rather than research. Weaker than the others because it also confirms you are not alone in it.
Ask what they like about the alternative before defending against it. The answer tells you which criteria are actually being used.
It might hit hard, but these are a few intent signals from prospects that mean nothing.
None are worthless as marketing metrics. They are worthless as reasons to put a deal in your forecast, which is how a pipeline ends up full while revenue stays flat.
Whether your contact has told anyone else.
A deal moves when the evaluation stops being one person’s curiosity and becomes their project. At that point they have attached their name to it internally and the cost of abandoning it rises. With committees now averaging six to ten people, this is the moment the process actually starts.
No tool measures it. You have to ask, and the question is simple. Who else needs to see this before you decide. The answer tells you whether you have a deal or a reader.
The practical version of all this is a short list of behaviours your CRM flags to a human, rather than a numeric score.
A score tells you someone is warm. A flag tells you what to do next, and only one of those closes anything. Worth noting too that these signals are about buyers arriving, not leaving. The early warnings that a customer is on their way out look completely different, and we covered those separately.
Buyer intent signals are behaviours indicating someone is actively evaluating a purchase rather than researching generally. The reliable ones involve other people joining the process or the buyer preparing to justify the decision internally.
A new person joining the conversation is strongest, because your contact spent internal credibility to make it happen. Implementation questions and requests for forwardable material follow closely, since both indicate the evaluation has moved beyond one desk.
It depends what you do with it. Intent data shows anonymous account-level activity, which is a starting point rather than an answer. Searches for buyer intent data have fallen 47% year on year, which suggests buyers are finding the outputs harder to act on than expected.
Because downloading is a way to avoid talking to someone. Competitors, students and people comparing prices download at the same rate as buyers, and nothing in the action distinguishes them.
Ask who else needs to see it before they decide. If the answer names people and a process, the evaluation is real. If they are still the only person involved, you have interest rather than a deal.
Clients come back for more
From stuck to a working revenue system