Our stacks leapt into the future, and our signals went the other way. Here’s how to judge whether the signals underneath yours are worth relying on, and what to fix before you buy more.
I’ve spent most of my career on the buying side of intent data. Picking apart source quality and sitting through every sales pitch under the sun. That rigor wasn’t for kicks, either. The line item came up with my CEO every year, and I was on the hook to defend it.
So, you can imagine how thrown I was when my SDR manager at the time mentioned, in passing, that his team had stopped working the in-market list. No grand decision. They just slowed until they stopped, the way people do when something wastes their time.
The dashboard said those accounts were active. But when we called, nobody on the other end knew what we were talking about.
For a long time, I assumed this was our fault. We retuned the scoring twice and rewrote the nurture track. It helped for a quarter before results fell back, because none of it solved the real problem: the signals couldn’t tell us which person at the account had done anything, or why.
Back then, a real person worked every account by hand, so the damage stayed contained. But now, with AI in the picture? Nothing slows a bad signal down.
Smarter stacks, same old weak data
Our GTM stacks leapt forward. They score, route, and personalize in seconds. Yet the signals feeding them have been heading the other way for years.
TrustRadius finds 63% of B2B buyers use AI during research, making it one of the fastest-growing sources.1
More vetting is happening where we can’t see it. So, we panic and shovel scraped traffic and best-guess IP-to-company mapping into our sophisticated stacks. And still, all it tells us is that an account stirred (not who moved, or why).
What good signals look like (and how to test yours)
A decade of long checklists boiled down to two questions. Both answers have to be a resounding yes before a signal gets near my stack.
- Can you name the person? The actual person, not the company. “Someone at ACME was active this week” is a thin signal. You’ve got no clue whether it was the security architect or a contractor, and a seller is at a loss for what to open with.
- Did it actually happen? Directly observed means a known, permissioned person engaged with real content, and it was recorded. Inferred means a model guessed from an IP address and some lookalike math. Both show up in your systems looking identical, but only one leads to a high-impact conversation with sales.
When a signal can’t make it past both questions, examine how it was built. Most intent data gets assembled backwards, from a company guess down toward a name. Reverse it. Start with what a real person engaged with, resolve that to person-level intent, then roll up to the account and market.
Once the signal names people and records what they did, AI stops being a risk and becomes the gamechanger we were promised. It can sort thousands of engagement records into a mapped buying group, surface the stakeholders you’d have missed, and tell a seller what problem each one is trying to solve.
Great signals deserve a process built to act on them
Signal quality sets the bar. But your process determines how much of that value you’ll ever capture.
A strong signal fumbled by a weak handoff produces the same result as a bad one. So, as you raise the bar on the data coming in, raise it on what happens next, too.
At my last company, 80% of the prospects who took a meeting became real opportunities. But only 13% of flagged leads ever got to that meeting. We didn’t have a demand problem. We had a follow-up problem. The best intent data in the world only pays off when your team is swift and exact when it comes time to move on it.
That matters more now because preference forms early. Forrester finds 68% of B2B buyers already have a front-runner when the process starts, and that front-runner wins 80% of the time.2
The teams that read the buying group first and have airtight processes to act on it become the front-runner. The teams still working off an account score with shaky handoffs will be too late. AI can make intent data the advantage it was always supposed to be. Or it can get you to lackluster outcomes at breakneck speed. The quality of your signal and the processes around it determine which. To truly outpace your competitors and ensure your big-bet investments in AI pay off in spades, make it your business to fix both.
With Informa TechTarget’s comprehensive Buyer Intelligence revealing the who, what, and why behind every signal, you can finally fuel your GTM stack with data that’s just as smart and sophisticated as the stack itself.
12026 B2B Buying Disconnect Report, TrustRadius
2Building Preference Is The Key To Winning B2B Buyers, Forrester
About the Author
