Buyer intent data shows up like a magic box. ABC Technology is in the market for VoIP, the box says. Why? How do you know? Those questions don't get answered, and a claim you can't poke at is a claim your rep can't use. Growth Orbit builds outreach on usable insight instead: research on a named company and a named person, source attached, specific enough that a seller can build an approach from it. Intent data hands you a verdict. Insight hands you something you can work.
Where buyer intent data comes from
Buyer intent data is a list of companies a vendor believes are researching a purchase right now. The ambition is fine. The problem is the mechanics, and what they leave out.
Intent vendors run networks of content properties. They log visitor IPs, match those IPs back to company records, and roll it up into a claim: these six companies are in the market for a new VoIP platform. You get the claim. You do not get the evidence.
You can't see which page was read, on what date, by whom, or why. And the behavior underneath is often thinner than the label. Someone read a white paper out of curiosity. Someone was cramming for a job interview. Neither is a buying process. Both can still show up as intent.
Growth Orbit wrote about this in 2023 as the Glengarry Glen Ross paradox: the premium leads everyone is desperate for, handed over sealed, supposedly the ticket. Who knows. That piece named the box. This one is about what has to be inside it.
Once you've bought the list, you cannot interrogate it.
The in-market list is already too late
"In market" means the deal is already contested, and someone else set the terms. Most people who buy intent data never price that in.
By the time a company is visibly shopping, a dozen vendors know it. Requirements are drafted. Evaluation criteria are written, usually with a preferred vendor's fingerprints on them. Whoever put the idea in the buyer's head already holds the credibility that decides these deals. Everyone who arrives after that is there to make the process look competitive and to press the price down.
You are not early. You are a quote.
Better intent data does not fix this. A perfectly accurate in-market list is a perfectly accurate list of deals you will fight on price against an incumbent-in-waiting.
And the data is often wrong
Two examples from Growth Orbit engagements, both paid for at list prices.
A client bought a list of companies said to be running AS/400 systems and shopping for a new financial services package. The team called in. Those companies had retired the AS/400 two years earlier.
Another client bought a VoIP in-market list. The team called in. They either were not in the market at all, or had made the change two years before.
Same story twice: a present-tense claim built on past-tense evidence, sold without a date. B2B contact and account data decays at roughly 30% a year. An intent feed with no visible timestamp gives you no way to tell decay from signal.
What usable insight looks like
Usable insight is research at the level of a named company and a named person, delivered with its source, specific enough that a rep can build an approach from it. Growth Orbit's working test: can you name the fact, open the source, and say it to a buyer without wincing? Fail any of those and you don't have insight. You have trivia with a citation, or a verdict with none.
A new CTO started four months ago. The company disclosed a security incident last year. The board has since asked for a defense plan. Each of those is a dated public fact with a source behind it, and a rep can do something with each one. Together they give a seller a reason to call this person, about this problem, now.
"In the market for VoIP" is also a claim about an account. Only one of those two claims tells a rep what to say on Monday.
Insight also doesn't claim the company is buying. That's the point. It says you've earned a reason to be in the conversation, and you have something to say when you get there.
This is how The Challenger Sale actually works. The sellers who win are the ones who change how a customer understands their own situation. The credibility from that reframe is what carries the deal. You can't reframe anyone's thinking with a list that says they are already shopping. You can only get in line.
Intent data still leaves the research to you
Assume the intent list is correct. You still don't know what triggered the flag, who inside the company acted, or what they were reading. You have a company name and a product category. Nothing in that record tells a rep what to open with.
So the research still has to happen. Intent data was never a substitute for insight. At best it's a sorting layer on top of work you were going to do anyway, and that's its honest, narrow job: it can suggest what order to work a list in. It cannot tell you what to say. You needed the insight before the list was useful.
How Growth Orbit does this at scale
Growth Orbit runs a research engine driven by a skill pack per market: a defined set of observations worth finding, and the public sources to find them in. Executive hires and departures, breach disclosures, regulatory filings, job postings, press releases, workforce notices, industry membership directories. Each finding is sourced, linkable, and dated. These are public signals, not black-box scores.
Four rules keep that from collapsing into an intent feed.
- No source, no score. Every finding has to carry a source URL. If it arrives without one, it gets downgraded to absent, scored zero, and written to the audit trail. An unsourced claim is worth nothing inside the system.
- The brief separates fact from inference. Each contact brief has five fields: the trigger event (the observation itself), the inferred problem, the hypothesis, the conversation angle, and the constraints. A reviewer can look at the observation alone and see whether everything downstream is grounded in it.
- The system is allowed to say it does not know. Every contact is classified rich, thin, or dark. Thin and dark contacts get no invented hook. They route to a simpler sequence or get suppressed. Personalizing badly is worse than not personalizing, and an intent vendor never returns "we don't know."
- Research has a shelf life. Findings carry a fetch date and get refreshed in staged waves against a decay policy. Insight that isn't current is just an expensive way to be wrong.
What to buy instead
Stop buying the shortcut. In most conversations with CEOs the request is for leads, and leads usually means someone with budget already trying to buy. Those buyers exist. What does not exist is a reliable way to purchase a list of them, and the ones a vendor can identify are the ones your competitors identified on the same feed.
You don't need more hustle. You need the research done the same way every time: source every claim, skip the invented hook when the evidence is thin, and refresh the pool as it decays. That's how this stays repeatable, instead of depending on whoever happens to be the hero that quarter.
Then judge any vendor, including this one, on one question. Not who is in market. Ask what your rep is holding when they pick up the phone, and whether they can tell the buyer where it came from.
Sales still takes work. The work is understanding the company you are calling, and bringing them something worth their time.