B2B Buying Signals: The Complete Guide to Spotting Intent

What is a B2B buying signal?

A B2B buying signal is an observable event or behavior that suggests a company, or a person at it, may be about to evaluate or buy something. It is evidence, not a prediction. A funding round, a pricing page revisit and a spike in research on a topic are all signals, and they differ enormously in how much you can trust them.

Why do they matter at all? Because most of your market is not buying right now. In a paper for LinkedIn's B2B Institute, Professor John Dawes of the Ehrenberg-Bass Institute notes that corporations change providers such as their principal bank or law firm around once every five years on average. That implies about 20% of business buyers are in the market over a year, and something like 5% in a quarter. He is explicit that the 95% figure "is not meant to be a precise rule" but a heuristic. Read it as a shape: at any moment, most of your market is not shopping, and a signal is how you find the minority that is.

The second reason is timing. 6sense's 2025 Buyer Experience Report surveyed nearly 4,000 buyers across North America, APAC and EMEA, each with a minimum spend of $25,000 over two years. By the time buyers first engage a seller, much of the decision work is already done.

The same report found the point of first contact with a seller moved from about 69% of the journey to 61%. Put together, the lesson is not "signals fix everything." Signals tell you who is moving and when. They cannot put you on a shortlist that was formed before anyone saw your name, which is a visibility problem covered in our AEO work and measurable with the AI visibility checker.

What are the main types of buying signals?

There are four families: first-party signals from your own properties, trigger events at the target company, third-party intent from research elsewhere on the web, and engagement with your own outreach. They are listed here from most to least direct.

First-party signals

These are the most direct signals because they come from the prospect's own behavior with you. Examples: a return to the pricing page, several service pages in one session, a stalled demo request, a question in chat. The limitation is identity. Most visitors are anonymous, and what you can identify depends on the law and the visitor's consent, which we cover below. Our visitor identification documentation explains how DeliChat identifies, scores and flags these moments, and says plainly that we do not claim a match rate.

First-party signals also decay fastest. A Harvard Business Review team audited 2,241 US companies on how quickly each answered a web-generated test lead.

In a separate analysis of 1.25 million leads across 29 B2C and 13 B2B companies, firms that tried to contact a lead within an hour were nearly seven times as likely to qualify it (a meaningful conversation with a key decision maker) as firms that tried even an hour later, and more than 60 times as likely as firms that waited 24 hours or longer. The data is old and covers inbound web leads, so treat the multiples as directional. The direction is the point: a hot first-party signal loses value by the hour.

Trigger events

Trigger events are things that happen to the company rather than things it does with you: a funding round, a new executive, an acquisition, a new office, a regulatory deadline, a job posting for a role that implies a coming need. Their strength is verifiability. You can check them, which means you can say them out loud to the prospect without hedging. Their weakness is that everyone with the same feed sees them at the same time.

Third-party intent

Gartner Digital Markets defines third-party B2B buyer intent data as behavioral information about an individual's or company's online activities collected across the broader business internet, usually through publishing co-ops, advertising networks and individual websites. Bombora, a well-known provider, describes its Company Surge data as coming from a consent-driven tag on sites in its data co-op, and measures "the most recent three weeks of activity against a 12-week historical baseline" for each account and topic. That is the general shape: a company is "surging" when its research on a topic rises above its own normal level.

Review sites are a second kind. G2, for example, describes its buyer intent as first-party activity on its own properties, such as profile, pricing, category, comparison and alternatives page views, delivered at the account level. A company viewing your pricing page on a review site is a stronger signal than a topic surge, but you still do not know who it was.

Engagement signals

These are responses to your own outreach. A reply is the strongest signal in this whole guide. An accepted connection is modest. Opens are close to worthless: Apple states that Mail Privacy Protection "stops senders from using invisible pixels to collect information about the user" and prevents senders from knowing when they open an email, so open data is unreliable for recipients who use it.

Which buying signals are reliable?

Reliability follows three rules: verifiable beats inferred, person-level beats account-level, and recent beats stale. A funding round filed with a regulator is a fact. "This account is researching your category" is an inference about a company made from the behavior of an unknown number of unknown people.

There is no independent published benchmark that ranks these by conversion. We went looking, and the reply-rate multiples vendors repeat do not trace to a disclosed study, which we covered in our breakdown of signal-based outreach. The closest independent evidence is Forrester's Global B2B Intent Data Survey, fielded in Q1 2023. It found over 85% of companies using intent data had achieved business benefits, but more companies expected benefits than achieved them across every benefit surveyed, especially in sales use cases and revenue growth. The top execution challenge was identifying specific contacts within accounts showing intent. Forrester's write-up does not state the sample size.

Where can you find buying signals for free?

You can find a lot for free: your own analytics and CRM, public regulatory filings, search alerts and company careers pages. These cover a focused list of accounts well, and they teach you which signals matter before you spend on data.

• Your own data. Web analytics, CRM history, chat conversations and email replies. This is the cheapest and most direct source, and it is the one most teams under-use. See how DeliHub holds this in one record.
• SEC EDGAR (US). The SEC provides free public access to filings. A company that relies on a Regulation D exemption must file Form D within 15 calendar days after the first sale, and the filing is publicly available on EDGAR. Public companies report events such as the appointment or departure of principal officers on Form 8-K, generally within four business days (Item 5.02).
• Companies House (UK). Free search shows company details and current and resigned officers, and companies must tell Companies House within 14 days of changes to directors.
• Google Alerts. Google says you can get emails when new results for a topic show up in Google Search. Set alerts on target company names plus words such as "appoints", "raises" and "expands".
• Careers pages and job boards. A role that only makes sense if a company is about to need what you sell. Check that it is a real, current posting.
• Job-change notifications from professional networks, for contacts you already know.

These rely on legal filing deadlines, so there is always some lag between the event and the record. Manual monitoring also stops scaling quickly. The research engine documentation covers how we automate the same market signals for signal-based outbound.

When is paid intent data worth it?

Paid intent data is worth it when you need to cover more accounts than a person can watch and you can act on what it surfaces within days. It adds scale and a view of research you cannot see yourself. It does not add certainty.

The paid categories, named by function rather than endorsement:

• Third-party topic intent from publisher co-ops, which flags accounts surging on topics you map to.
• Review-site intent, which flags accounts viewing your category, your profile or your competitors.
• Website visitor identification, which resolves your own traffic to companies, and to people where the law and consent allow.
• Job-change and trigger feeds, which monitor moves and events at scale.
• ABM platforms, which bundle several of these with scoring.

Before you buy, run a test. Take your last 20 closed-won accounts and ask each vendor whether their data showed a signal on those accounts in the weeks before the deal opened, and how many non-buying accounts showed the same signal. A vendor that cannot answer is asking you to buy on faith. Also ask how the data was collected, what consent it rests on, how often it refreshes and how topics are mapped. Plan for the contact-resolution step too, because account-level data does not tell you whom to call.

How do you score and decay buying signals?

Score each signal on three things: fit (is the account in your ideal customer profile), strength (how direct and verifiable the signal is) and recency (how long ago it happened). Fit is a gate, not a bonus. A pricing page revisit from an account you cannot serve is noise.

Decay matters because a signal's usefulness fades. HubSpot's documentation describes score decay as automatically reducing a score based on how long ago a scored event occurred, with the percentage and time period set by you, for example 50% every three months. That example suits slow CRM engagement. A pricing visit should fade in days.

Two refinements help. First, combine families: a trigger event plus first-party behavior on the same account within a short window is stronger than either alone. This is our judgment, not a published finding, so test it. Second, tier the output into three queues: act today, act this week, and watch. Anything first-party goes to a person quickly, in line with the response-time evidence above.

To back-test, list your last 20 closed-won deals and note which signals appeared on each account and how many days before the first conversation. The signals that repeatedly appear early, and the gap between them and the first conversation, are your starting weights and decay periods.

How do you turn a buying signal into a message?

Name something true, say why it matters, and ask for very little. The test is simple: could you say the opening line out loud, with the evidence on screen, without the recipient feeling watched?

Gartner's survey of 632 B2B buyers, run in August and September 2024, found 73% of buyers actively avoid suppliers that send irrelevant outreach. A signal-led message that is vague or creepy is just irrelevant outreach with extra steps.

The full method, including worked scenarios and an interactive tool, is in our signal-based outreach guide. For the copy itself, see annotated outbound templates. For how we run this as a service, see B2B lead generation services.

What causes false positives?

False positives come from automated traffic, privacy features, single people standing in for whole accounts, and signals that arrive after the decision is made. Know the main sources before you automate any response.

• Bots. Imperva's 2026 Bad Bot Report found automated traffic made up more than 53% of all web traffic in its latest data, up from 51% the year before. Raw visit counts are not buyers.
• Privacy features. Apple's Mail Privacy Protection, described above, makes opens unreliable.
• One person, one account. Forrester surveyed more than 16,000 business buyers and reports 13 people on average are involved in a buying decision. One visitor or one surge is one voice in a group.
• Wrong audience. Students, journalists, competitors, existing customers and job seekers read the same pages as buyers.
• Stale signals. A topic surge from last quarter or a job posting nobody removed.
• Late signals. Per the 6sense data, if the shortlist was set before the signal fired, the signal is real but arrives too late to win on speed alone.

Each of these is a reason to keep a person in the loop on any message that goes out. Our AI SDR analysis covers where automation holds up and where it does not.

Is it legal to identify visitors and act on signals?

It depends on the signal, the visitor's location and the consent behind the data. This is general information, not legal advice. Signals about companies from public records are low risk. Signals about identifiable people are personal data, and tracking technologies have their own rules.

• UK. The ICO says PECR applies to any technology that stores or accesses information on a user's device, including cookies, tracking pixels, link decoration, device fingerprinting, web storage and scripts. Consent is the default, with limited exceptions. The ICO's guidance also describes an exception for certain statistical analytics, where users must be able to object.
• EU. The GDPR and the ePrivacy rules apply and are enforced by each member state, so check each market you sell into rather than assuming one rule fits all.
• California. The Attorney General's CCPA page lists a right to opt out of the sale or sharing of personal information and a right to know what has been collected.
• Third-party data. Gartner notes that GDPR and CCPA now require overt acceptance of tracking cookies, which makes tracking buyers across the internet harder. Ask every vendor how its data was collected and on what consent.

For outreach itself, see cold email compliance under CAN-SPAM, GDPR and CASL. For our own approach to identification, see data ownership and privacy and the consent section of the visitor identification documentation.

How do you measure whether buying signals work?

Measure signal-to-meeting rate by signal type, compared with an untriggered baseline in the same ideal customer profile. A signal program that cannot show lift against that baseline is a reporting exercise.

To isolate the effect, hold out a random share of comparable accounts and contact them on your normal schedule, ignoring signals. Do not rely on vendor-quoted multiples. We make no claim of a specific reply-rate or meeting uplift from signals, because we have no source for one, and you should treat anyone who does with the same skepticism.

How do you build a buying signal program?

Start narrow, make the signals verifiable, and put a person on the response. A practical sequence:

1. Define fit first. Write down your ideal customer profile. Signals on accounts outside it are noise.
2. Pick two or three signals. One first-party, one trigger event, and optionally one third-party source if the budget is there.
3. Instrument what you own. Make sure pricing page visits, repeat visits and replies land on a CRM record. DeliChat and DeliHub do this for our clients.
4. Set up free monitoring on your top 50 to 100 accounts using filings, alerts and careers pages.
5. Write one message per signal, and decide in advance what you will and will not say.
6. Score, decay and tier. Use the model above, then calibrate on your closed-won deals.
7. Review weekly. Look at precision and signal-to-meeting rate, retire weak signals, and keep a hold-out group.

Sources

• John Dawes, Ehrenberg-Bass Institute, for the LinkedIn B2B Institute, Advertising effectiveness and the 95-5 rule: most B2B buyers are not in the market right now
• 6sense, The B2B Buyer Experience Report for 2025, nearly 4,000 buyers, minimum spend $25,000 over two years
• Oldroyd, McElheran and Elkington, Harvard Business Review, The Short Life of Online Sales Leads, 2011, 2,241 audited US companies and a separate study of 1.25 million leads
• Gartner, Gartner Sales Survey Finds 61% of B2B Buyers Prefer a Rep-Free Buying Experience, 632 B2B buyers, August to September 2024
• Gartner Digital Markets, 3 Best Practices for Using Third-Party B2B Buyer Intent Data in ABM
• Forrester, Intent Data Expectations Vs. Reality, Global B2B Intent Data Survey, Q1 2023, sample size not stated
• Forrester, The State Of Business Buying, more than 16,000 global business buyers, December 2024
• Bombora, Company Surge, methodology as published by the vendor
• G2, Buyer Intent documentation and Buyer Intent Data, vendor descriptions
• HubSpot, Build lead scores, documentation on score decay
• Apple, Apple advances its privacy leadership with iOS 15, Mail Privacy Protection
• Imperva, Bad Bot Report 2026: Bots in the Agentic Age
• SEC, Form D frequently asked questions, Form 8-K general instructions and EDGAR search and access
• GOV.UK, Company changes you must report and Get information about a company
• Google, Google Alerts help
• ICO, What are storage and access technologies?
• California Attorney General, California Consumer Privacy Act

FAQ

Related reading

• Signal-based outreach: how it actually works
• Outbound lead generation: the complete guide
• B2B lead generation: channels, costs, and what converts
• B2B outbound templates that actually work
• Cold email compliance: CAN-SPAM, GDPR, and CASL
• AI SDR: what it is, what it does, and where it fails
• Demand generation vs lead generation