B2B buying signals: which ones to act on and how fast they go stale

What are B2B buying signals?

B2B buying signals are dated, checkable events, such as a demo request, a job post, a funding filing or a thread asking for recommendations, that give you a reason to contact a company now. A signal is a reason to look, not proof that anyone will buy.

Signals matter because buyers do most of the work before they talk to a seller. In a Gartner survey of 750 B2B buyers involved in complex purchases, buyers spent about two-thirds of their time researching, learning and meeting with peers independently of suppliers, and just 17% of it meeting suppliers. That is one survey of complex purchases, so read it as direction, not as a rule for every deal.

Vendors define the terms differently. 6sense calls review-site data such as G2's second-party. An Anton Consulting white paper hosted by TechTarget rates TechTarget's own registration-based data as strong and third-party data as weak to moderate, so treat it as an interested view. This guide uses three plain groups instead.

Intent signals for sales: first-party signals come first

A form fill, a demo request or a reply is the clearest signal, because the person came to you. It also goes cold quickly. A 2011 Harvard Business Review study of 1.25 million sales leads at 42 US companies found that firms that tried to reach a lead within an hour were nearly seven times as likely to qualify it, meaning to have a meaningful conversation with a key decision maker, as firms that tried an hour later. The data is old, it covers inbound leads, and one co-author led InsideSales.com, so read the multiple as direction, not a forecast.

A website visit is the weakest first-party signal. It shows interest in a page, not a person or a need. Company-level visitor identification matches an IP address to an organization. Warmly, a vendor that sells it, reports company-level match rates of 30% to 65% across its customers, with remote work, VPNs and mobile networks behind the misses. Some vendors also sell person-level matching built on cookies and identity graphs. Do not mention anything learned that way in outreach.

Treat a fit company on your pricing page as a reason to look at that account, and a visit to one blog post as noise.

Public signals for sales prospecting: hiring, funding, leadership, tech and news

Public signals are free, dated and checkable. Your competitors can see them too.

Hiring signals for sales prospecting

A job post is public, dated and specific, which makes it the most practical public signal. Pick roles next to your problem: a first SDR hire may point to a need for outbound help. A post shows a need the company chose to staff, not a need for your product.

Check the date before you cite it. Greenhouse, an applicant tracking system, documents a public Job Board API that needs no authentication for GET requests. Its job list shows updated_at, while the single-job view also shows first_published, so a listed date may be an edit date, not the day the role opened.

Funding trigger events

In the US, a company that sells securities under Regulation D must file a Form D with the SEC within 15 calendar days after the first sale. The SEC's staff FAQ defines the first sale as the date the first investor is irrevocably contractually committed to invest. Filings are public on EDGAR.

Two catches. The SEC says a Form D can be filed before any securities are sold, so a filing does not prove a round closed. And offerings under other exemptions, or raised outside the US, may never appear. Press announcements can land well before or after the filing.

Funding means new money and pressure to grow, not a need for your product. Any window you read for how soon after a round teams start buying is a rule of thumb, so test it on your own accounts.

Leadership change

A common claim says a new executive reviews vendors in the first 90 days. The research we checked is thinner than the claim. Deloitte's guide to CIO transitions says new CIOs will likely need to understand and assess their web of relationships and quickly address existing vendors and partners. A 2013 study of newly appointed CIOs (Gerth and Peppard) describes an entry phase of about 4 to 6 months and an overlapping stabilization phase of 9 to 12 months, and says the first 90 days matter but are not the whole story.

Both cover CIOs only, and neither measures vendor switching by month. Treat a leadership change as a hint that vendors may be reviewed, not as a window with a date.

Technology changes

Tools such as BuiltWith crawl public web pages and read the signals a site gives off. BuiltWith gives each detection a confidence score from 0 to 1 and says a lower score can mean a detection is outdated or was seen only once. Because these tools read public pages, expect front-end and hosting technology such as analytics, advertising and CMS, and little of what runs behind a login.

A removed competitor tag or a newly added complementary tool can be a real trigger. A tool installed for years is a filter for your list, not a trigger.

News and community threads

News goes stale in days to weeks, so read the date, not the headline.

A thread where someone asks Reddit or Hacker News for a tool recommendation is often the closest public signal to a stated need, and threads are usually active for days, not weeks (our rule of thumb). Reply in the thread with something useful, say who you work for, and post it yourself.

Intent data for small business: what paid feeds can and cannot tell you

Paid intent data reports that an account is reading more about a topic than usual, at company level. Here is how four common sources describe themselves.

• Publisher co-op (Bombora). Bombora says its data comes from member sites that run its consent-driven tag, across 25,300+ topics. An account is flagged when its last three weeks of activity on a topic rise above its own 12-week baseline.
• Review site (G2). G2's documentation says its Buyer Intent data covers actions on G2: profile, category, comparison and alternatives page views, plus pricing and sponsored content. Only people researching on G2 are visible.
• Keyword network (6sense). 6sense describes third-party intent as research activity across other organizations' sites, so you can see which keywords and topics your accounts research. It says it also partners with Bombora, TechTarget, TrustRadius and G2.
• Bidstream. Data from the bid requests passed between ad exchanges. The UK ICO's 2019 report lists IP address, cookie IDs and location among the usual contents, and said that in its view consent is the only lawful basis for routine real-time bidding processing of personal data. Bombora, which sells a competing product, describes bidstream intent as keywords scraped from pages with no baseline of past activity.

Limits shared by all of them:

• Company-level only. G2's documentation says its Buyer Intent data is company-level, so every person at an account shows the same activity in your CRM.
• A topic is not a purchase. Competitors, students and job seekers read about topics too.
• Accuracy claims come mostly from vendors. Vendors criticize each other's methods, as Bombora does with bidstream, and the sources above are published by parties with a stake in the answer.

For a small team, free first-party and public signals usually cover a lot of ground before a paid feed does. If you test a feed, follow Forrester's 2025 advice: ask for sample data, use narrow keywords and a window of weeks, expect the value of a signal to decay quickly, and check the sample against accounts you already know well. FoxKnock does not sell intent feeds, so weigh this advice accordingly.

Which signals justify outreach

Write when all five hold. If one fails, watch the account instead.

1. Fit. The company matches your ideal customer profile.
2. A chain. You can state, in one sentence, how the signal connects to a problem you solve.
3. Fresh. The signal is inside its window (next section).
4. Reachable. You have a plausible person with a verified email.
5. Sayable. You can say what you saw without the reader recoiling.

Whether signal-led outreach beats fit-only outreach for you is something to test, not assume. Split a list, send both halves in the same weeks and compare replies. Small samples mislead.

How fast signals go stale

Windows without a cited basis are our rules of thumb. A post older than the window may mean the role is filled or stuck, so check the posting date and the careers page before you mention it.

How to rank signals: fit x strength x freshness x reachability

Multiply four numbers for each account.

• Fit is how well the company matches your ICP, from 0 to 1.
• Strength is how directly the signal points at your problem, from 0 to 1. A form fill outranks a news mention.
• Freshness decays with age. A 7-day half-life is a reasonable start: a week-old signal counts half, two weeks a quarter.
• Reachability is 1 with a verified contact and lower without one (0.55 in the example).

All numbers here are invented for illustration:

Company D has the best fit, but a 20-day-old funding round sinks it. Company C has a good signal and no contact, so it waits for a verified email. Multiplying also means a zero anywhere gives zero: a hot signal at a company outside your ICP never surfaces.

Then cap the list. Someone who sends about 25 emails a day can work about 25 accounts, so show the top 25 and let the rest wait.

How to mention a signal without being creepy

The test is simple: would you be comfortable saying how you know? The ICO's direct marketing guidance says that if you find it hard to explain what you plan to do, or would rather not tell people because they might object, that is a sign you should rethink the activity.

Three habits help. Cite what the reader could verify in a minute, with the date (research a prospect shows a source-for-every-line method). Use public, company-level facts, never behavior inferred from tracking, IP matching or ad data. Connect the signal to a problem without claiming to know their pain.

Names change the legal picture. The ICO says that if you can identify an individual, the information is personal data even when the person acts in a business role, so the UK GDPR applies to a named contact at a company. The ICO also says data protection law does not necessarily stop you using publicly available personal information for direct marketing, but there may be restrictions: you need a lawful basis, the use must be fair, and you must tell people that you hold their data, what you will do with it and where you got it. Public does not mean agreed.

When the data comes from another source, the ICO says to give that privacy information within a reasonable period and no later than one month after obtaining it, or at your first message to the person if that comes sooner. The ICO's business-to-business guidance is under review after the UK's Data (Use and Access) Act, and rules differ by country. The GDPR cold email guide covers the details. This is general information, not legal advice.

How FoxKnock handles buying signals

FoxKnock's buyer intent signals apply the same structure. Credit prices below are defaults and may change (see pricing).

• Monitors: hiring, funding, news, tech change, competitor and custom search, run hourly, daily or weekly on up to 50 companies per run. Funding, news, competitor and custom search hits get AI triage. A new signal costs 12 credits at default prices.
• Other sources: form submissions (strength 90), company-level website visitor identification (needs a Snitcher or IPinfo key, sets no cookies), and relevant Reddit and Hacker News threads.
• Not covered: job changes, website changes, email opens or clicks, third-party browsing data and real-time alerts.
• Today queue: fit x strength x freshness (halves every 7 days; signals older than 45 days drop out) x reachability (0.55 without a usable contact), with a penalty for people contacted recently. One card per company, capped at 25.
• Drafting: drafts cite evidence cards with source links, and hiring, funding and news cards expire after 90 days. By default every email waits for your approval and sends from your own mailbox.

A signal is still a reason to look. Open the source link before you write to anyone.

To try this on your own accounts, create a free FoxKnock account. The Free plan costs nothing, needs no card and includes 500 one-time credits once you verify your email, enough to point FoxKnock at your website and research a handful of companies. The pricing page lists signal monitors and automated plays on the Growth plan.