High Intent Leads and Where Buyer Intent Signals Come From

A high intent lead is a person or account whose recent action shows they are buying now. Typical actions are a demo request, a look at your pricing, a direct comparison of you against a named alternative, or a reply asking how you work. The action needs a date, and you need to know who took it. If either is missing, you have a guess.

So high intent is about timing. A company can fit your market perfectly and show no buying intent this quarter. Next month the same company might show plenty. Intent describes a moment, and moments expire.

Buyer intent signals come from four places:

1. Your own channels (first-party).
2. Another platform's audience, shared with you (second-party).
3. Research tracked across publisher networks (third-party).
4. What the buyer tells you directly (declared).

Most of what gets sold as "intent" is the third kind. It is measured at company level.

Disclosure: my team works mostly through cold email and LinkedIn outreach, and when I quote G2, Bombora or HubSpot below, I am quoting their own vendor documentation.

One of our 30-day campaigns, run on 5 LinkedIn profiles plus email, ended in 2 signed contracts. The full funnel is in the last section, and it shows why intent alone cannot fill a pipeline. I treat intent as a tool and volume as the engine.

What makes a lead high intent, and how is that different from fit or qualified?

Three separate questions get mixed together under the word "lead". Pulling them apart fixes most bad prioritization.

• Fit asks whether you should sell to this company at all.
• Intent asks whether someone there is acting on the problem right now.
• Qualified asks whether a real conversation has confirmed the need, the person with authority and the timing.

Keep all three as separate fields in your CRM.

A lead can be high fit, show zero intent and still be unqualified. That is still a good account to contact. A lead can also show high intent and be a terrible fit, like a student researching your category for a thesis. That second lead is the one that eats a sales week.

What does high purchase intent look like? The buyer has stopped learning about a problem and started choosing how to solve it.

• Low intent looks like education. Reading a how-to guide, downloading a broad industry report, following a company page.
• High intent looks like evaluation. Checking what something costs, lining you up against a named alternative, asking for a demo or a trial, asking who else uses you.

Why do buyers show intent long before they talk to a salesperson?

Most of the buying happens without you in the room. Nobody buys from an email. They scrutinize you first and purchase later.

Gartner surveyed 646 B2B buyers between August and September 2025, and 67% of them said they prefer a buying experience with no rep in the loop (Gartner).

Forrester's State of Business Buying 2026 says a typical buying decision now involves 13 internal participants and 9 external influencers, and more than 60% of business buyers use a trial (Forrester).

Both findings come from press releases. Neither breaks results down by company size, and the Forrester release gives no sample size. Read them as direction for mid-size and large purchases. They give no precise rule for a 20-person firm.

For anyone hunting high intent leads, that changes how you read every signal.

Intent leaves traces before anyone replies. Research, review-site visits and comparisons happen while you are still invisible.

A trial is late-stage intent. Someone who trials a product already has a shortlist.

One person's click speaks for one person. In a buying group of that size, a single pricing-page visit is weak evidence about what the whole group will do. The sale is happening online, before anyone talks to you.

Where do buyer intent signals actually come from?

Each of the four sources answers a slightly different question. Sorting accounts by intent just means grouping them by which source flagged them and how recent the signal is.

First-party data comes from your own website, emails, LinkedIn conversations, demo forms and CRM. You own it.

HubSpot's vendor documentation keeps two scores apart (HubSpot knowledge base, vendor documentation):

• An engagement score, built from actions like site visits and clicks.
• A fit score, built from details like job title, company size and revenue.

It combines them into labels from A1 to C3. A means the highest fit, and 1 means the highest engagement. This is the same intent-versus-fit split as the table above, built into a tool.

Second-party data is another platform's own audience, shared with you. G2's vendor documentation says its signals fire when a buyer does any of these on its sites (G2 documentation, vendor documentation):

• Views your product profile.
• Opens the pricing page on your G2 profile.
• Views a comparison page that includes your product.
• Views an alternatives page for your product.

The data covers G2, Capterra, Software Advice and GetApp, and updates daily. Each buyer gets a Buying Stage (Awareness, Consideration or Decision) and an Activity Level.

Third-party data is research tracked across sites you do not own. Bombora's vendor documentation describes Company Surge like this (Bombora, vendor documentation):

• It flags a business that is researching a topic more than its own historical baseline.
• The data comes from a co-op of 4,000+ B2B publishers, read across 6,900+ topics.
• It collects no personally identifiable information, so it names a company and never the person who was reading.

Bombora calls surge a strong indicator of purchase intent. Its documentation publishes no accuracy or conversion figures to back that up. What Bombora costs a buyer is a separate question.

Declared intent is the buyer telling you directly: a reply, a form, a call request. It is the rarest signal, and the one you can act on today. Outreach can produce it on purpose. Almost nothing else can, which is why I would rather start 50 conversations than wait for 5 signals.

Is a company showing intent the same as a person showing intent?

No.

An account surge tells you that someone at the company is reading about a topic. It does not tell you who that person is or whether they decide anything. It might be a junior analyst writing an internal report.

A person-level signal is a named reply or a demo form with a real name on it. These are rarer, and you can act on them today.

To turn an account signal into named contacts:

1. Go through your past deals and work out who actually signed off on each one. I start every list from that answer.
2. Search the flagged accounts using the exact titles that niche uses. I keep a bare "manager" title off the list.
3. Contact two or three of those roles at each account. One person's click does not speak for the group.

Then write to each role about the decision that role owns. The CFO decides cost and risk. The operations head decides disruption. The technical lead decides fit with what they already run.

Never mention the signal in the message. The prospect does not know they were tracked. With company-level data you cannot even be sure the research came from them.

Write to the problem the research suggests and ask who owns it. Open with a pitch and the message gets ignored. Period. The first message only has to sell the meeting, not your services.

Here is an invented case so you can see it: you sell ERP migration services, and a mid-size distributor surges on "ERP replacement".

Subject: who owns the ERP switch?

Hi Marta, distributors your size usually hit one question early in an ERP replacement: who keeps the warehouse running while the old system goes dark? In most firms that call sits with operations. If it sits with you, would a 15-minute call next week help you compare how other teams staged the switch? If it belongs to someone else, a name is just as useful. Tom

How fast does an intent signal go stale?

Faster than most dashboards suggest. Two vendors at least publish their own clocks, so start there.

G2's documentation links follow-up timing to Activity Level:

G2 publishes no outcome data showing that this timing works. Still, it is the vendor's own clock, and you should respect it if you pay for the signal.

HubSpot lets you set score decay to 1, 3, 6 or 12 months, and it counts a month as 30 days. The setting exists because an old click should not keep a lead at the top of the queue forever.

You will also see exact decay windows quoted in days with no method behind them. I ignore those.

Run every signal through this freshness checklist before a rep spends time on it:

1. Date the signal. If you cannot say when it happened, count it as a failed check.
2. Name the person. If you only have the account, map it to real decision-maker titles before writing.
3. Check the buying group. Is more than one person or role acting at the same account? One visitor is weaker than three.
4. Check the window. Is the signal still inside the vendor's stated timing (immediately for high, within a week for medium on G2), or inside your own decay setting for first-party scores?
5. Downgrade if it fails. A signal that fails two of the first four checks becomes a fit-only account and goes into your normal outreach.

What does the law say about tracking website visitors for intent?

This section covers UK rules only. This is not legal advice.

Regulation 6 of PECR (the UK rules on cookies and electronic marketing) was rewritten on 5 February 2026 by the Data (Use and Access) Act 2025. It stops anyone storing information on a user's device, or accessing information stored there (legislation.gov.uk). The new wording also covers collecting information the device sends out automatically. The rule is subject to the exceptions in Schedule A1.

The Information Commissioner's Office (ICO) explains what this means for cookies and similar technologies (ICO guidance):

• Consent needs a positive action. Simply continuing to use the website does not count.
• PECR applies to all cookies, including those that collect anonymous data.

The ICO says that page is based on its previous cookies guidance and that a revision is under way.

In practice, if your first-party intent comes from a tracking script on your site, that script needs consent from UK visitors before it runs, unless an exception applies. Turning the output into company names or anonymous scores does not take the tracking out of scope.

Other countries have their own rules, so check each market you track. This matters most for identifying website visitors.

Which intent source fits which team, and when should you skip intent data?

My rule is simple. We use intent data when it fits, and it fits when the total addressable market is large.

When the market is small, intent data adds little. Take 2,000 target companies with signals showing on about 10 of them. In that case we skip intent data and write one offer for every company in the market. Ten flagged accounts still leave 1,990 you have to reach some other way.

The small-market row is the expensive one to get wrong. I would rather a team commit to one offer for one audience for 12 months than chase whichever accounts lit up this week.

If you are weighing a purchase, the first question is whether a small team should pay for intent data at all.

Why can't you fill a pipeline from high intent leads alone?

Because the people who signal before you contact them are a thin slice of the people who will buy. Here is the campaign I mentioned at the top, in full.

One campaign, 30 days, 5 LinkedIn profiles plus email:

The 26 conversations in that table are declared intent, and my team's outreach is what produced them. Send 1 or 2 messages to a short list of ideal accounts and you never meet the buyers who would have replied on touch 5 or 6. They were busy or not ready that week. That is why I set the monthly goal at conversations started, not at calls booked. Calls come out of conversations.

The third follow-up usually brings the most replies. That holds only if real time has passed since the second one. Stop at two and you never see those replies.

Another client, 30 days, one LinkedIn account with email follow-ups:

A team working only 50 to 100 ideal-fit accounts a month is relying on a coin toss. Those people might be swamped this month, tied into a contract, or simply not thinking about the problem this quarter. None of that says anything about your offer.

More volume means more chances to win. Volume negates luck.

Here are the ranges I plan with, from our campaigns:

• Cold email reply rates usually sit around 1%, and at thousands of emails a day they are often closer to 0.5%.
• Roughly 10% to 50% of replies turn into meetings that get booked.
• Those averages slide as the volume rises.

These are ranges from our campaigns, not a promise for yours. The funnel above mixes LinkedIn and email, so do not read its reply rate against the email range.

An example for planning: about 10,000 people contacted in a month at about 1% -> about 100 replies that month -> 10 to 50 meetings.

The last piece is the inbox. Put one person on the replies so that nothing gets missed. A reply is the freshest intent signal you get: dated, named, and sent by choice. Leads who raise their own hand, like a demo request, are inbound prospects and deserve a prompt answer. Next, qualify the lead before it is booked. That way a fresh signal becomes a real meeting instead of a polite no-show.

Pull last month's replies and count how many came from accounts that showed any signal before you wrote to them. That number tells you how much intent data is worth paying for.

If few replies came from pre-signal accounts, keep your money on volume and fit lists. If many did, test a paid feed on one tier only.

Not sure intent signals are worth chasing in your market? Book a call and describe who you sell to. I will give you a straight answer.