Everyone's Betting on Signals. Plan the Response First.

Kyle Poyar and Maja Voje launched the 2026 State of B2B GTM report at their summit on October 14, GTM being the sales and marketing side of a company. They asked 300 B2B software and AI companies what's working, and most of the report is about channels: which ones teams call core, which ones pay off, and where the money goes in 2027. I spoke at the summit, so I've had a couple of weeks with it. One pair of numbers keeps pulling me back.

Where the money is going

A signal is something a buyer just did, like coming back to your pricing page. Signal-based outbound is reaching out because of one, instead of because a name is on a list.

Ask where teams plan to spend more on sales and marketing in 2027 and two answers lead. AI search, getting your company into the answers AI assistants give, is first at 50%. Signal-based outbound is a point behind at 48%. Among companies with $10 million to $50 million in revenue, it's first, at 58%.

Now the other number. Asked which channels are core to their growth today, 33% of teams named signal-based outbound. That's the same share as last year. The report files it under "still experimental," below cold outreach, events and paid ads.

So the money is arriving ahead of the practice. Half the market plans to spend more on something only a third of it calls core today. I don't read that as a warning. I read it as a planning problem, and planning problems have a fix.

The respondents name the problem themselves. Pipeline, the deals a team is working toward a close, is the top challenge, picked by more teams than any other, and the quote the authors chose to illustrate it is the clearest sentence in the report: "the hard part is identifying the right buyer signals and channels that consistently convert into paying customers."

The signals teams act on are the ones they own

If the bet were about data, you'd expect teams to be short of it. They aren't. The report asked which signals teams act on. Among companies selling deals of $5,000 to $50,000 a year, which is about half the respondents, the top three are content engagement (54%), website visits (47%) and product usage (41%). Third-party intent data, bought lists of companies researching a topic, is at 20%.

The top three come from your own channels: your content, your site, your product. They're where a buyer tells you something only you can see, which is why I spent a whole series, starting with The Class You Own, on signals you don't have to rent. When I wrote my partner note for the report, I pointed at the website row: fewer than half of companies act on website visits, whatever their deal size.

So the 48% aren't waiting on a data purchase. Most already have the signal. What they're deciding is what to do when it fires.

Same outreach, opposite odds

The report rates each channel's odds of paying off, by deal size. Look at three rows that are, mechanically, the same thing: writing to someone who didn't write to you first.

Signal-based outbound rates good to very good at every deal size. Cold outreach rates OK to poor, and very poor on the smallest deals. Agentic campaigns, the report's name for AI tools that send cold email sequences, rate very poor everywhere.

Same outreach. What separates them is what started the message. Cold outreach starts from a list. Signal-based outbound starts from something the buyer did. An agent sending to a list is still sending to a list, faster, and the respondents who tried it said so on the report's "what didn't work" pages: "High volume, near-zero reply rate, and it hurt our sending reputation."

I don't take that as a verdict on agents. An agent that waits for a signal is running the first row, not the third. The trigger decides which row you're in.

Personalization is the reason you're writing

The same pages carry a line I could have written in August: "Cold email with superficial personalization." A few rows away sits another: "Straight cold outbound without filtering for prospects with the problem we can solve."

In The {FirstName} Fallacy I argued that a merge field isn't personalization. A respondent said it in five words. It's worth being exact about why, because a signal bet is a bet on getting this part right.

Most personalization stops at the bottom three layers: the first name, the company, the role and industry. Each is a fact about the person. None is a reason to write to them today. Only the top layer is: what they did, and when. That's the signal.

It doesn't mean telling them you watched. "I saw you on our pricing page" is the fastest way to lose the thread. The signal decides who hears from you, when, and what you send. Someone back on your integrations page for the third time this week gets the integration guide and an offer to walk through it. Someone who read three posts on one problem gets the next useful thing on that problem, not a meeting request. The reason does the work without being announced.

Write the response before you fund the bet

Here's what I'd do before any signal spend goes into a 2027 budget. For each signal you plan to act on, write three lines.

• The signal. What the buyer did, in plain words, and what makes it count. "A second visit to pricing in a week, from a company that looks like our customers" is a signal. "They're in market" isn't.
• The response. What goes out, from whom, and what it says. Write the first line of the actual message. If it opens with a first name and no reason, go back to Figure 4.
• The clock. How fast it has to happen. Some signals are worth a reply within minutes, while the buyer is still deciding. Others can wait for Monday's sales review. Pick on purpose.

If you can't write all three lines for a signal, it isn't ready for budget. Get them right and you're running the one kind of outbound the report rates good or better at every deal size. Start with the signals you already own, the teal rows in Figure 2, because you can test those this week without buying anything. Run each one by hand for a week before anything automates it. Ten real sends will teach you more about the response than any planning doc, and you'll find out whether your clock is honest.

Kyle and Maja close the report with what they call the GTM barbell: be exceptional at relationships, be exceptional at AI, and "beware of the space in between." The first line under it is "Signals can tell us who's most likely to buy." The relationship end is still the strongest in their data: network and intros is core for 62% of teams and rates excellent at every deal size. A signal doesn't replace that. It tells you when to show up, and what to bring.

Bryn is the agent we built at Civic to spot buyers on your site and run the follow-up you approved. Those three lines are what Bryn calls a Play: you write it once, Bryn runs it when the signal fires, and every run is logged. If you'd rather start with a spreadsheet and a calendar reminder, start there. The three lines are the same either way.

Source: The 2026 State of B2B GTM report, Kyle Poyar (Growth Unhinged) and Maja Voje (GTM Strategist), October 2026, 300 B2B software and AI companies. Civic was a partner on the report.

Stop watching signals. Start running them.