Guide / Buying Signals

Signal-based selling, from first principles

Most outbound still works the old way: build a big list, write one message, send it to everyone on the same day. Signal-based selling is the opposite. You wait for a reason, then you write. This is the guide I wish I had a decade ago, when I was running a volume agency and could not understand why response rates kept falling. It pulls together the whole approach: what a signal is, why timing beats volume now, the four signals we lean on, and how to turn them into a working loop.

What signal-based selling actually is

Signal-based selling means you trigger outreach off an observable event that suggests a company is in a buying window, instead of blasting a static list on a schedule. The signal is the reason the message is timely. A funding round, a key hire, a tech-stack change, a new executive in a buying seat: each one tells you something changed this quarter, and change is when budgets move.

The contrast that matters is list-based versus signal-based. List-based outbound asks one question: does this company fit my ICP? Signal-based outbound asks two: does it fit, and is something happening right now that makes the message land? The second question is the entire difference. It is the difference between knocking on every door on the street and knocking on the door of the house that just put up a renovation permit.

Why it works now and did not have to before

Three things changed at once, and they are the reason this approach went from a nice-to-have to the default. Buying signals became cheap and accessible: hiring data, funding announcements, tech-stack detection and social activity are all available without an enterprise contract. AI got good enough to do the research, drafting and reply triage that used to need a room full of people. And the buyer stopped responding to generic outreach, because the buyer now gets a hundred generic messages a week and has learned to delete on sight.

I have a particular reason to believe this. I started a cold email agency in 2014, grew it past a hundred people and a few thousand clients, and the entire model rested on volume forgiving sloppiness. Send enough, and the math worked. That era is over. Reply rates on volume outbound have fallen by more than half over the last several years, while signal-driven, properly personalized outreach runs multiples higher. The market repriced sloppiness, and signal-based selling is what is left standing.

Anatomy of a signal

Not every event is a signal. A useful signal has three properties. It is observable, meaning you can detect it without insider access. It is timely, meaning it opens a window that closes, so acting this week beats acting next quarter. And it is predictive, meaning it correlates with a real buying decision rather than just being interesting trivia.

A company changing its logo is observable and timely but not predictive of anything you can sell against. A company hiring its third QA engineer in a quarter is all three. The discipline of signal-based selling is mostly the discipline of throwing out events that fail the third test, no matter how easy they are to detect.

The four signals we lean on most

There are dozens of possible signals. In practice four of them carry most of the weight across the B2B motions we run. Each one has its own guide, because each behaves differently in timing, sourcing and message framing.

• Hiring signals. A public job post is a budget announcement in disguise. The role tells you which budget just unlocked and roughly when it closes.
• Funding signals. A round does not change a company on announcement day. It changes it over the following two or three quarters, in a predictable order you can time against.
• Tech stack signals. What a company already runs tells you its budget level, its maturity, what integrates cleanly, and where the displacement openings sit.
• Job change signals. A new executive in a buying seat has roughly a quarter of political license to replace what she inherited. The highest-density window we know of.

They overlap, and the overlap is where the strongest plays live. A funding round produces a hiring spree, which produces the job posts. A new VP arrives and reviews the stack she inherited. The signals are not independent events, they are different views of the same underlying motion inside a company.

Scoring and sequencing signals

Once you have more than one signal firing, you need a way to decide what gets the high-intent workflow this week and what waits on nurture. We do not use a fancy model for this. We use a simple additive score: fit first, then signal strength, then signal recency, then signal stacking.

1. Fit is the gate. A signal on a non-ICP account is not a signal, it is a distraction. Firmographic fit comes first, always.
2. Signal strength is the multiplier. A new VP Sales outranks a single junior job post. Weight signals by how reliably they predict a real decision.
3. Recency decays the score. A signal from six months ago is closer to noise than to intent. Date-stamp everything and expire it.
4. Stacking boosts it. An account with two signals firing in the same month gets prioritised over one with a single signal, because two views of the same motion is stronger evidence than one.

We run this in our own platform, usegrit.io, where signals, enrichment and scoring live in one place. Clay does the same job well and is the alternative we point people at. The tool matters less than the rule: fit gates, signals prioritise, and stale signals get expired without sentiment.

Building the loop

Signal-based selling is a loop, not a campaign. A signal fires. Enrichment runs on the account. An agent produces a short research brief. A draft message gets written in the assigned voice, referencing the implication of the signal rather than the signal itself. A human approves the batch. The sequence runs across email and LinkedIn. Replies get classified, and the positive ones reach a human the same hour. What worked feeds back into which signals you weight next week.

The single most common mistake is leading the message with the signal. "I saw you posted a role for a Senior QA Engineer" reads exactly like the automation it is. Lead with the implication instead: the pain the person in that new seat is about to inherit. The signal is why you are writing this week. It does not have to be the reason you give in the message.

Where this sits in the bigger picture

Signal-based selling is the engine room of GTM engineering. If you want the wider discipline it lives inside, start with what is GTM engineering. If you want the step-by-step build, the how to build a GTM system guide shows where signals plug into the seven-step model. And if you want to go one level deeper on the data side, buying intent data covers what is real and what is vendor noise.

The honest summary, from someone who spent years on the other side of this: volume was never the point, it was just the cheapest thing to measure. Signal-based selling puts the point back where it belongs, on timing and relevance, and lets the boring parts run themselves.