Buying signals are the observable actions and events that suggest a person or company may be moving toward a purchase. A second visit to your pricing page counts, and so do a fresh funding round, a job ad for a role your product supports, a trial user hitting a plan limit and a former champion turning up at a new employer.

None of them proves anything on its own. They're evidence, and the conclusion you draw from them is buying intent, the readiness to buy that you never get to observe directly. One signal shifts the odds a little. A few of the right ones, at an account that fits, shift them enough to justify a rep's time this week.

Teams tend to get this wrong in one of two directions. Some ignore signals and keep working a static list from the top down. Others pipe every flicker into a sequence and call it personalization. The second group usually does more harm, because it teaches buyers to delete their emails and teaches reps to stop opening the alerts.

TL;DR

• Buying signals is the widest term in this corner of sales. It covers six families: first-party behavior, product usage, trigger events, technographic changes, third-party intent and relationship moves. Intent signals and trigger events are both subsets of it.

• The families arrive at different moments. Third-party research and trigger events tend to show up before a buyer has touched you, pricing visits and product usage show up once they're looking at you, and a champion's job move can land at any time.

• Weigh each signal on four things before anyone acts: how specific it is to you, how recent and repeated it is, who did it, and whether the company matches your ICP. Email opens barely count, because Apple Mail loads them in the background.

• Act on combinations. A workable starting rule is two signals from different families at a fitting account, with lone weak signals parked until something backs them up.

• Bought signals scout and owned signals confirm. Strong combinations go stale within days, so route them to a named rep the same working day, and keep every signal dated on the company record in your CRM.

• Judge the program by meetings per signal type, the share of triggered touches that go nowhere and the hours between signal and first touch. The number of signals captured tells you nothing.

• Nebor builds this signal layer in a client's own CRM and Clay workspace, tunes the thresholds against the meetings it actually produces, and leaves the rules written down for the team that runs it.

Buying signals, intent signals and trigger events: which word covers what

Buying signals is the umbrella term. It covers anything that changes your estimate of whether an account is in the market. Intent signals are the narrower group about active research, such as category searches, review-site visits and repeat reads of your comparison pages. A trigger event is a dated change inside the company itself, like a newly hired CRO or a funding round that has just closed, which says nothing about research but often comes right before it.

Each one asks for a different reaction. Research tells you what topic is on someone's desk. A trigger tells you why now might be different from last quarter. A pricing-page visit tells you they've found you specifically. Treat a funding announcement as if it were a demo request and you'll be pitching a company that hasn't started looking yet.

The six families of buying signals, and when each one appears

A program that reads only one family misses most of the accounts that are moving. These are the six groups we sort signals into.

• First-party behavior. What identified people and companies do on channels you own: repeat pricing visits, comparison or integration pages, replies, webinar attendance. It's aimed at you, which makes it the strongest family, and it's what a website visitor workflow is set up to catch.

• Product usage. In a free trial or freemium motion, the actions that tie someone to real value, like switching on a core feature, hitting a plan limit or inviting colleagues. These signals are hard to fake, which is why a product qualified lead usually beats a form fill.

• Trigger events. Funding, a leadership change, a hiring push, a new office or an acquisition. They don't show any research yet, but they create the budget or the mandate that research follows.

• Technographic changes. A company adds a tool that pairs with yours, or removes a competitor. Stack data from technographic data sources turns into a buying signal when it changes, not when it sits still.

• Third-party intent. Research done away from your site. Bombora's Company Surge scoring compares an account's last three weeks of reading on a topic against a 12-week baseline, and G2's Buyer Intent documentation separates views of a category page, your profile, your pricing and a head-to-head comparison page. Both report at company level, not person level.

• Relationship moves. A past champion lands at a new company, a former user of your product joins a target account, or someone keeps engaging with your team's posts, which is where social selling and LinkedIn outreach take over. People move more often than most CRMs assume: in the US, median tenure with a current employer was 4.1 years in January 2026, according to the Bureau of Labor Statistics. A moved champion is one of the warmest openings in B2B, and one of the least watched.

Timing is what separates them in practice. Third-party intent, trigger events and stack changes come early and tell you which accounts to watch before anyone there has heard of you. First-party behavior and product usage come late, once the account is looking at you, and they're the cue to move. Relationship moves follow no order at all, which is why champion tracking needs its own alerts.

Four questions that decide what a signal is worth

A demo request and a single blog visit both count as buying signals, and treating them the same is how alert channels fill up with noise. Before a signal changes anything, it has to answer four questions.

1. Is it about you or about the category? A comparison page that names your product sits far closer to a decision than a general article about the problem. Some first-party signals are weaker than they look, too. Apple's Mail Privacy Protection loads remote content in the background whether or not the person engages, so an email open from an Apple Mail user may only mean the email arrived.

2. How fresh is it, and has it happened more than once? This week beats last quarter, and three visits beat one. Repetition is what turns a flicker into a pattern.

3. Who did it? The same visit means more from the person who owns the budget than from a junior analyst gathering links. Most purchases involve several people, so read signals against the buying committee you'd expect at that account, not one contact at a time.

4. Does the company fit? This is the question that gets skipped. A loud signal from a company outside your ICP is still a poor opportunity. That's why you keep fit and intent as two scores: a burst of activity can't then pass for a good fit, and a perfect fit with no activity can't pass for a ready buyer.

Nine days at one account, signal by signal

Say you sell contract management software to mid-sized companies, and a 220-person logistics firm in Rotterdam is in your target list. Here is what the next nine days could look like from your side.

On day one, your intent provider flags the company as reading more than usual about contract lifecycle management. On its own that's barely worth a note, since you don't know who's reading or why. On day three, the firm posts a vacancy for a legal operations manager, and the description mentions "replacing manual contract tracking". On day five, someone at the company opens the G2 page comparing you with your biggest competitor. On day eight, two visitors from the same company spend time on your pricing and security pages.

Alone, any one of these deserves a watch list at most. Together they come from four different families (third-party research, a trigger, a comparison and first-party behavior), they all point at the same problem, and the company fits your size and industry criteria. That's the moment the account should land in a rep's queue, with all four signals on one record.

The message that follows shouldn't mention the visits. It should talk about what the job ad already said out loud: the cost of tracking contracts by hand at a company growing that fast. The signals decided when to write, and the vacancy decided what to write about.

From single alerts to a rule the team trusts

The biggest improvement most teams can make is to stop reacting to individual signals and start acting on combinations. One data point is noise. Three that agree, from sources that have no reason to agree by accident, is a story worth a phone call.

Writing that down is what a sales signal stack is for: a short, agreed list of the signals you trust, each with a weight, plus a threshold that decides when an account goes to a person. Our usual starting rule is two signals from different families at an account that passes the fit check. Lone weak signals wait in a holding list and expire after a few weeks if nothing joins them.

Suppression matters as much as scoring. Filter out customers, open opportunities, competitors and disqualified companies before any alert fires, or reps learn to mute the channel within a week.

When the trigger for outreach is behavior on your own site, the motion has its own name, inbound-led outbound, and that entry covers the filters and the first message in detail. When any mix of signals decides who gets contacted and when, the practice is signal-based selling, and the stack is its rulebook.

Owned signals confirm, bought signals scout

Signals you collect yourself and signals you buy behave differently enough that they need different weights.

Owned signals come from your website, your product, your inbox and your CRM. They're about you, nobody else sees them, and they're the clearest evidence you'll get. Their weakness is timing: they only appear once an account is already paying you some attention.

Bought signals come from intent providers, review sites and data vendors. They catch accounts earlier and in far larger numbers, but they're noisier, they usually point to an account without naming anyone in it, and every competitor with the same subscription sees the same surge on the same Monday.

So use each for what it does well. Third-party data decides which accounts deserve a closer look, and first-party data confirms the interest is in you. The first-party versus third-party data entry covers the trade-offs in more depth.

Keep in mind that a lot of buying research never produces a signal of either kind. Peer recommendations, private chats and podcast listening happen in the dark funnel, so an account with no signals isn't necessarily an account with no interest.

How fast a buying signal goes stale

Signals lose value by the day. The buyer's attention moves on, the tab gets closed, and a competitor who saw the same surge may already be in the conversation.

The research behind this is older than intent data. When the authors of The Short Life of Online Sales Leads timed how companies answered web inquiries, their conclusion was that most weren't responding nearly fast enough. Their subject was people who had filled in a form, but the logic carries over to softer signals. The same principle sits behind speed to lead.

In practice this rules out the weekly signal report. A strong combination should reach a named rep the same working day, as a Slack message or CRM task with the evidence attached. On inbound, where the account has already raised its hand, an inbound meeting workflow does the same job for pricing visits and form fills. A perfect signal acted on a week late is just a well-documented miss.

Getting every signal onto one company record

Signals don't arrive in your CRM by themselves, and they don't arrive from one place. Each family has its own sources:

• Website behavior comes from website visitor deanonymization tools, which usually name the company rather than the person.

• Product usage comes from your product analytics and billing system.

• Funding, leadership changes and hiring come from news monitoring and job-post data, through tools like PredictLeads or a news monitoring workflow.

• Stack changes come from technology detection tools such as BuiltWith.

• Third-party research comes from providers like Bombora and G2.

• Champion moves come from job-change tracking with tools like UserGems, or a scheduled check against LinkedIn.

The hard part is pulling all of that onto the company record in the CRM, with a date on every signal, so it can be combined and scored. We usually run the collection and scoring in Clay and write the result back to HubSpot or Salesforce. A signal that lives in a vendor dashboard nobody opens might as well never have fired.

What to count once signals start reaching reps

Capturing more signals is easy and proves nothing. Three numbers tell you whether the program is producing pipeline or just producing alerts.

The first is meetings by signal type. If repeat pricing visits turn into meetings far more often than content downloads, the weights should say so. After a quarter or two, your thresholds should come from this data rather than from anyone's instinct.

The second is the false-positive rate: the share of signal-triggered touches that went nowhere because the account never fit or was never serious. When it climbs, reps stop trusting the alerts.

The third is time from signal to first touch. A program that picks the right accounts but reaches them four days later still loses most of the value.

High false positives mean the rule is too loose. Low volume with high conversion means it's probably too tight. Good accounts reached late is a routing problem, not a signal problem.

Recap

Buying signals are the raw evidence of buying intent. A signal is a hint, a pattern of hints from different families is a reason, and a reason at an account that fits is worth a rep's best hour.

The teams that struggle rarely lack data. They react to single data points, skip the fit check and let alerts age in a report, and more data only gives them more noise to react to.

The teams that do it well keep the rule short, keep every signal on the company record and look at which signals actually turned into meetings. Then they adjust. It's unglamorous work, and it's what makes outreach feel timely to the person receiving it.

If you'd like a signal layer like this, we set up the sources, the scoring and the routing in your own CRM and Clay workspace, and leave you a written rulebook that explains each weight, so the program stays yours to adjust.

Frequently asked questions