How to find leads showing buying signals, without buying a database

Observed signals beat bought intent scores

Third-party intent data (Bombora, 6sense and the like) guesses which companies are researching a topic, mostly from what gets read at their IP addresses. It names the company, rarely the person, and you can't check it. Published accuracy estimates run from about 15% to 81% depending on who's measuring, and every one of them comes from a vendor or a rival. I wouldn't lean on any of them.

An observed signal is something you can read: the Head of RevOps at a company you'd love to sell to, commenting on Tuesday that her team is “finally replacing the spreadsheet”. You can judge it and decide whether to write. If you already pay for intent data, use it to choose accounts, then look inside them for a person who did something you can see.

Seven places to look, all free

1. LinkedIn post search. Pick the Posts filter, set “Date posted” to the past week and search the phrases people use when they're shopping: "anyone recommend", "alternatives to" with a competitor's name, "looking for a" with your category. Quoted phrases and OR work in the search bar.
2. Comments on competitors' posts, and on posts by the people your market listens to. Read the comments and skip the likes. Someone who writes “we tried this and onboarding was painful” has told you more than a hundred reactions.
3. Job boards. Search Google for site:boards.greenhouse.io, site:jobs.lever.co or site:jobs.ashbyhq.com plus the role that implies they need you, in quotes. Three openings in one team say more than one. On LinkedIn Jobs, save the search with the past-week filter and LinkedIn emails you new matches.
4. Past customers and champions changing jobs. UserGems says about 20% of customers and champions move each year, and that deals involving a former customer are won 114% more often (their data, and they sell the tracking). The free version is a spreadsheet of people who liked your product and a monthly look at their profiles.
5. Your own profile viewers and post engagers. A free account shows only your few most recent viewers, by most reports, and Premium shows the last 90 days. These are some of the strongest signals you'll get, and ones you should never mention (more on that below).
6. Funding news, from Crunchbase's free news pages, local startup newsletters and LinkedIn announcements. The round itself is a crowded moment. The marketing or sales hire that follows a month or two later is usually the better person to write to.
7. Unhappy reviews and recommendation threads. One- to three-star reviews of a competitor on G2 or Capterra often give the reviewer's role and company size, and a Reddit or Slack-community thread asking for a tool is about as close to a raised hand as you'll find.

Rank by fit first, then freshness

A signal from someone who could never buy is noise, however strong it looks. A great-fit prospect with no signal is still worth a message, with less urgency. Bingle scores leads that way: score = 100 × fit × (0.3 + 0.7 × intent). A perfect fit with no signal at all still scores 30, and a hot signal from a poor fit stays near the bottom.

Freshness is harder, because nobody has published a real decay curve. The 24-hour and 72-hour rules you'll see are rules of thumb, even on vendor blogs that quote confident numbers. What does hold up is that signals age at different speeds. A comment asking for recommendations goes cold in days. A new leader or a funding round stays relevant for months, because the budget and the mandate don't disappear. Bingle's own half-lives run from a week for a profile visit to four months for funding (the full table is here), and they're our judgement, not a measured curve.

You can measure your own. Log the date of each signal and the date of each reply, then bucket them: 0 to 3 days, 4 to 7, 8 to 14, 15 and over. After about a hundred messages you'll know more about decay in your market than any blog post can tell you, this one included.

Mention what they made public, and keep the rest to yourself

Some signals are things a person chose to publish: a post, a comment, a job ad, a funding announcement, a new-role update. Referring to those is normal. Others you only know because you were watching: a profile view, a visit to your site, an intent score, a like on a competitor's post. Those should decide who you write to and when, and never appear in the message. “I saw you looked at my profile” turns a buying signal into a surveillance disclosure, and it points at the weakest part of it, since plenty of profile views are idle curiosity.

A hiring post and a new role, used well:

“Saw you're hiring three AEs at once. Is ramping them the slow part?”

“Six weeks into a new sales role, most people I talk to are still deciding what to keep from the old setup. Is outbound on that list?”

Keep it short. Gong looked at more than a million executive sales cycles (January 2026) and found replies drop sharply past 100 words, with the best emails at 50 to 100. C-level buyers were 30.2% less likely to reply than everyone else, which is one more reason to give them a specific reason to.

Mistakes that waste a good signal

• Opening with "Congrats on the new role!" and nothing else. It's the message every vendor sends that week.
• Spending the signal in the connection note. Belkins found no difference in acceptance between invites with a note (26.37%) and without one (26.42%), so save the point for your first message after they accept.
• Treating every like as a lead. Recruiters, students and other vendors like posts too.
• Sending as many invites as you can. LinkedIn doesn't publish its limit (tool vendors put it around 100 a week), and a low acceptance rate seems to hurt an account more than volume does.

Where Bingle fits

Everything above works by hand, and the hard part is doing it every day. Bingle covers the LinkedIn half: it watches posts, comments, job changes, hiring, funding and your profile viewers, ranks each person by fit and freshness, and drafts the first message for your OK. Its writer follows the rule above, naming a job ad or a new role when it helps and never mentioning a profile view or a like on a competitor's post. Review sites, Reddit threads and Greenhouse searches are still yours.

Running an agency? Here's which signals predict an agency hire, and when.

There are guides for recruiters, consultants, SaaS founders and MSPs, and one for each signal, in the guides.

Questions

Where the numbers come from

1. Gong, Do execs really reply to cold email?. 29 January 2026
2. UserGems, the business case for champion tracking. vendor data
3. Abmatic, intent data providers compared. 2026, a competitor of the providers it rates
4. MarketBetter, Bombora review. 2026, vendor
5. Bombora on intent data myths. vendor
6. Valley, hiring signals for prospecting. September 2026, vendor
7. Fuzzy AI on post engagers, citing Belkins' invite tests. 29 September 2026, vendor
8. Valley, LinkedIn invitation limits in 2026. vendor; LinkedIn publishes no figure
9. HubSpot on who can see profile viewers. third party