Intent data is evidence that a person or company is getting ready to buy, based on something they did instead of something they are. A company with 40 employees is a fact. A company that just posted a job for a facilities coordinator is a signal.
Software vendors sell intent data as a feed, and for large sales teams it can be worth the money. But most of the signals that matter to a small business are public, and you can read them for free in about 45 minutes a week. This article lists six of them, shows where to look, and gives you the opening line each one supports.
Intent data in plain words
There are two kinds of information about a prospect. The first kind describes them: industry, size, city, job title. Sales people call this firmographic data. It tells you who could buy. The second kind records behavior: a search, a job post, a permit filing, a complaint about a competitor. That is intent data, and it tells you who might buy now.
The reason it matters is timing. A property management company with 200 doors is always a possible customer for a landscaper, but it picks a landscaper maybe once every few years. Intent data narrows the list to the weeks when that company is actually deciding.
You will also hear intent data split into first-party and third-party. First-party means behavior on your own site or in your own inbox: who opened an email three times, who looked at your pricing page. Third-party means behavior somewhere else, such as which companies are reading articles about a topic. The free signals below are neither of those. They are public events that happen to other people, and you only have to notice them.
What the paid intent products sell
Vendors such as Bombora, 6sense, ZoomInfo and G2 sell intent data. In general terms, they watch research activity across networks of websites and tell you which companies seem to be reading about a topic more than usual. Some feeds are scored by an algorithm, some are tied to visits to a review site. Pricing varies and is mostly not listed on their sites, so check their current pages if you are curious.
Three honest limits apply to those feeds, and they are the same limits for any buyer. The signal is usually about a company, not a named person, so you still have to find who to write to. A topic like "payroll" is broad, so a company reading about it might be a buyer, a student, or someone comparing options for their cousin. And the feed is the same one every other customer of that vendor sees, so a dozen sellers can write to the same company in the same week.
For a team with a marketing department and a large market, that trade can still make sense. For a two-person agency or a local contractor, the free signals below fit better because they point at specific events in a specific town.
Six buying signals you can check for free
1. A job posting that matches what you sell
A company hiring a bookkeeper has a bookkeeping problem. A company hiring a recruiter, a dispatcher or a facilities coordinator has the same kind of problem in a different department. Search a job board for the titles your service replaces or supports, filtered to your region, and sort by newest.
What it tells you: there is a gap, and someone has budget approved to fill it. Try: "I saw you're hiring a bookkeeper. Until that person starts, would a few hours a week of outside help be useful?"
2. A permit or a new business license
City and county permit portals list building permits, tenant improvements and new business licenses, often with the owner's name and address. A restaurant that pulled a permit to renovate is about to need a sign, a cleaning crew, a pest contract and a payment processor. Many cities publish these lists, searchable by date.
What it tells you: a project exists, it has a start date, and the owner is making a pile of purchasing decisions. One way to open: "I saw the permit for the renovation on Pine Street. Do you have a cleaning contractor lined up for after the build?"
3. A new location or an expansion announcement
Local business journals, press releases, a "now open" post on a company's own page, and a new Google Maps listing all announce a second location. A new location doubles every recurring need the first one had.
What it tells you: the company is growing and has not yet settled its vendors for the new site. You could write: "Congratulations on opening the second shop. Who handles your signage and window cleaning at the new address?"
4. A funding round or an acquisition
Funding news is public, and it goes stale fast. A company that just closed a round or bought a competitor often has a short window in which it hires, buys software and replaces vendors that did not scale.
What it tells you: there is money and pressure to spend it. For example: "Saw the announcement about the new round. Teams that grow this fast often outgrow the payroll setup they started with. Is that on your list?"
5. A bad review of a competitor
Read the one-star and two-star Google reviews of the companies that do what you do. A customer who writes "never called back" or "missed three appointments" has told you exactly what they will pay for next time. Be careful with this one. Do not quote the review or mention that you read it in a way that sounds like you were watching them. Offer the thing they said was missing.
What it tells you: a customer is unhappy and probably shopping. A line that fits: "Do you have a contractor who answers the phone the same day? That is the main thing we build our week around."
6. A public question asking for a recommendation
"Can anyone recommend a good accountant for a small restaurant?" posted in a neighborhood group, on Reddit, or on a local forum is the strongest signal on this list, because the person said it themselves. The catch is that it expires in a day or two, and these communities have rules about self-promotion. Reply with help, say you own the business, and follow the group's rules. On Reddit, Conversations finds threads like this and drafts a reply you post from your own account, and the keyword alerts guide shows the manual version.
How long each signal stays useful
Signals decay at different speeds, and that decides how often you check each one.
- A public question for a recommendation: 24 to 48 hours. Check daily.
- A bad review of a competitor: about two weeks, and fresher is better. Check weekly.
- A job posting: until it is filled, often three to six weeks. Check weekly.
- A permit: a few weeks before work starts, and for months of the project after. Check weekly or every other week.
- A new location or a funding announcement: roughly one to two months. Check weekly.
One more point is that signals stack. A company with a new permit and an open job posting is a better target than either alone.
Turning a signal into a first line
Every opening line built from a signal has the same three parts: name the event in a few words, say what it usually means for them, and ask one small question. Keep it to two or three sentences. Do not say "I noticed your company is in the market for..." because it sounds like surveillance, and do not use anything the person has not made public.
Here is how a week looks for a fictional two-person commercial cleaning company. Dana spends 15 minutes on job boards for "facilities" and "office manager" postings, 15 minutes on the city permit portal for tenant build-outs, and 15 minutes reading one-star reviews of three competing cleaners. That is 45 minutes. She finds 8 job postings, 4 permits and 3 reviews, which is 15 signals. She sends 15 emails, each starting with the specific event. She does not know in advance how many will answer, and she should not assume it will be many. What she does know is that all 15 of those people, this week, had a reason to be thinking about it.
For the full writing process, how to write a cold email that gets replies covers subject lines and length, and personalisation at low volume covers how much research is worth doing per email. If you want to go from signals to a repeatable weekly routine, outbound prospecting for small teams splits the work into pick, find, reach and follow up. For the list-building side, see how to find companies that are about to buy and how to find new businesses in your area to sell to.
Rules to follow when you use public signals
Reading a public permit list is fine. Scraping a job board in bulk usually breaks its terms of service, so read the page and type the name into your list. Emailing the people you find is regulated: in the US, CAN-SPAM requires an accurate From line and subject, a physical address and a working opt-out. Check state rules too, and if you sell financial, insurance or legal services, check the solicitation rules for your field.
What to do when you can't check them all by hand
Forty-five minutes a week works for one person selling to a few hundred local companies. It breaks when the market is larger or you stop doing it in a busy month. That is where an AI worker helps: it finds the buyers by name, writes the first email, sends it from your own Gmail, follows up, and books the meeting on its own. You still choose which signals matter and what the offer is.
Intent data is only useful if someone acts on it while it is still true. Rocketship finds the buyers, emails them from your own Gmail, follows up, and books the meeting on its own. Plans that get you customers start at $24.99 a month (Launch), and every monthly plan starts with a 7-day free trial, card required.
