"Signals" is the fastest-growing buzzword in B2B GTM. Nobody agrees what it means.

How signal-based GTM actually evolved since 2025

I've watched signal-based GTM rotate three times in the last eighteen months. Each rotation followed the same pattern. A new "signal" became the meta. Early movers built pipelines around it. Three to six months later, everyone copied them. The arbitrage closed. The market rotated again.

The receipts:

1. Before Jun 2025Luck-based
   You sat on LinkedIn 24/7 and hoped a relevant news item would land in your feed. Most of it didn't. The rest you saw three days late, after someone else had already replied to the founder.
2. Jun to Sep 2025Engagement signals
   We bet hard on this before it was even a conceptualized category. Short-form video to drive pipeline. Tracking who commented on what. Replying to viral posts within minutes of them going up. The teams that ran this play well got real outbound conversion. The teams that copied it three months later got noise.
3. Sep to Oct 2025Funding, hiring, job changes
   The market rotated. These three became the top of every signal-based outbound play. This is when 'signal-based GTM' went from niche vocabulary to actual category. The earliest movers had already built the pipelines. The rest of the market started buying.
4. Nov 2025 to Jan 2026Social signals at scale
   Everyone caught up. The same plays that worked for early movers six months earlier became table stakes. Response rates dropped. The arbitrage closed.
5. Now · May 2026No new meta
   The 'new signal as meta' is finally gone. There is no next thing to copy. Which means one thing: you actually have to put in the work.

Not just collect signals fast. Verify them. Attribute them to a bigger company picture. Use them as contextual pieces in a buyer's story, not as silver bullets to close the next deal.

I've interviewed 30+ operators on how to build a signal-based motion since we started Signalbase. We've shipped 700+ demos in the same window. Here's the part nobody wants to say out loud.

Signal-based go-to-market is not sending a message when someone comments on a Reddit thread. It is not sending a message when someone follows a person on LinkedIn. Those are signals in the most technical sense and worthless in practice. The barrier to action is so low that the noise drowns the value.

Every vendor in this category is bending reality to fit their product. Most of the "thought leadership" hitting your feed right now is a Claude-generated lead magnet with a vendor logo at the bottom. Be skeptical of anyone selling signal-based outbound who can't tell you which signals matter, which don't, and why.

Intent data vs signal-based GTM: two infrastructure categories, one buzzword

Marketing teams and GTM Engineering teams both talk about "signals" now. Put a job description from each side by side and that's the only word they share. We pulled 1,378 active GTM job posts to map how far apart the two definitions actually are.

Marketing teams talk about "intent data" and "intent signals." Honestly, I can't blame the marketers out there that fall into the intent data trap. The whole industry was built on legacy vendors selling "getting ready to buy leads" (third-party content consumption, ABM scoring) and marketing it as buying intent. In reality, it was just a simple website visit.

The vocabulary is settled. GTM Engineering teams talk about "signal-based" and "buying signals," but the vocabulary is younger, the stack is younger, and the motion is genuinely new.

We pulled 1,378 active GTM job posts via the Signalbase API across four functions (Marketing, GTM Engineering, Sales, RevOps) to figure out who's actually talking about signals and what they mean by it. 75 posts (5.4%) used signal or intent language. We read every one. The word fractured cleanly into two infrastructure categories.

Same word. Two completely different infrastructure categories. This is why every conversation about "signal-based GTM" goes sideways. Marketing hears intent data 2.0. GTM Engineers know you mean something else entirely. Live events from the public web, captured the moment they happen, fed into workflows that fire outbound the same hour. Both can be right, but they're describing different things.

The clearest tell isn't the vocabulary. It's the stack.

Zero meaningful tool overlap between the two. A senior demand gen manager who has never used Clay? Not unusual. Most haven't. A senior GTM Engineer who hasn't touched a 6sense dashboard in a year? Normal. These are different jobs that share a word.

Where intent data and signal-based GTM are actually growing

When testing our hypothesis, we saw that 75 of 1,378 posts used signal or intent language. That's the 5.4% baseline, but the aggregate is misleading. The growth is concentrated in two functions and absent in two others.

Marketing is moving fastest. Posts mentioning intent data, intent signals, or signal-based vocabulary went from 6.5% in February to 7.4% in March to 15.1% in April.

2.3x increase in 60 days. Caveat for the data nerds: this measures postings still active at pull date, so survivorship bias is in play. The trend direction holds across cohorts, but absolute numbers in earlier months understate slightly.

The interesting part isn't that the language is growing. It's where it's growing. Here's the cross-function picture:

GTM Engineering passed 15% months ago and now sits closer to 20%. The function was built around this category, so this is the floor, not the ceiling. Marketing is following but mostly renaming: intent data adoption is migrating into "signal" vocabulary while the underlying motion (nurture, ABM scoring, slow burn) stays the same.

Sales and RevOps mention signals at rates close to random noise. If you've been hearing more about signal-based GTM from those two on LinkedIn lately, that's the algorithm talking, not the hiring data. They haven't woken up yet.

From buying intent data to building signal infrastructure

What's more interesting than the vocabulary growth is the type of role being created. Teams aren't just adding "intent data" to a tool wishlist; they're hiring people whose entire job is to build signal infrastructure from scratch.

Three examples pulled live from the active pool:

What stood out across these listings is that signals are the job, not a tool the new hire happens to use. None of them say "we bought 6sense, come interpret the dashboards." They say "we need someone to build the signal layer from scratch." Different role, different KPI, different ask.

And this isn't happening at the Fortune 500. The Fortune 500 is still negotiating its annual intent data renewal. The teams actually building it are mid-stage, with no legacy intent stack to defend and no five-year contract to honor.

Skip building your GTM signal infrastructure.

Intent data and signal-based GTM aren't competing technologies. They're competing definitions of what "signal" means. A 6sense subscription doesn't replace a Clay workflow, and a Clay workflow doesn't substitute for an account scoring layer. Pick the category deliberately.

And if you pick signal-based GTM, decide the next question carefully too: do you build the infrastructure yourself, or buy it?

Every week, a GTM tooling founder asks me the same thing on a call: who do you resell to get this much data? Same answer every time: nobody. Real-time signal infrastructure isn't a database play. It's pulled from social media, news, investor datasets, advanced web scraping, and a lot of in-product logic that turns one detected event into three more.

That sourcing is the easy part. The trap is what comes after.

Building it is cheap. Maintaining it is the trap.

The raw API call is a penny a fetch, live in an afternoon. Then comes everything nobody demos: deduplicating across sources, verifying each event against its origin, attaching a source URL and a confidence score, and monitoring it all in real time. That's not a sprint. It's a pipeline your engineers babysit forever, one that breaks every time a source changes its markup.

You'd burn two quarters rebuilding what's already running today, and every one of those weeks is a week your team didn't spend on the product your customers actually pay you for.