How to Respond to a Buying Signal Without Sounding Like You Have Been Watching

The fastest way to waste a good buying signal is to open with it. "Saw you just raised your Series B, congrats!" lands in a pile of fifty identical emails, and "I noticed you're hiring three integration analysts" reads like you've been watching. The signal is why you're writing. It is almost never what you should lead with. Here's the structure that works, plus the line between public and creepy and how long you actually have before a signal goes cold.

The line between public and creepy

Nobody minds you knowing something they announced. People mind you knowing something they didn't.

That's the whole rule, and it holds up in practice better than any personalization framework I've used.

It is also a higher bar than the legal one. The FTC's CAN-SPAM compliance guide governs commercial email with rules on identification and opt-out; nothing in it stops you from referencing inferred browsing behavior. The law permits plenty that still costs you the reply.

Fair game: a press release, a funding announcement, a job posting, an RFP, a conference talk, a regulatory filing, a published review. The company chose to put it in the world. Referencing it is no different from having read their website.

Not fair game: anything inferred about an individual's behavior. "I saw you visited our pricing page twice this week." "Our data shows your team has been researching warehouse automation." Both are technically available to you. Both make the recipient feel watched, and the feeling is the point. You've told them you have a file on them.

This is the practical reason to prefer public events over inferred intent, separate from the accuracy argument. Public events are quotable back to the person. You can say where you saw it. Try writing that sentence about an intent score.

If you want the broader principle this sits inside, our guide to ethical prospecting works through value-first outreach at length.

The structure: event, pain, relevance, small ask

Four moves. The event is the smallest of them. (This assumes you've already found an event that genuinely connects to the pain you solve. The guide to B2B buying signals covers that derivation, and getting it wrong is a targeting problem no amount of copywriting fixes.)

1. The event: one clause, not a sentence. Enough to establish why you're writing, and no more. "Since the Austin office opened..." Not "I saw the exciting news about your expansion into Texas!"

2. The pain it implies. This is your opening line's real job. Don't tell them what happened; tell them what you think it means, and be specific enough to be wrong. "A new office usually means a revenue target in a market where nobody knows you yet." That sentence earns a reply if it's right and a correction if it's wrong, and both are conversations. This is where personalization stops being decoration: McKinsey finds the fastest-growing companies drive 40% more of their revenue from personalization than slower peers, and the mechanism is exactly this specificity.

3. Your relevance: one sentence, no feature list. How you've helped with that specific problem. Not what your product does.

4. A small ask. Not a demo. A question they can answer in one line, or a five-minute conversation. The signal earned you attention, not commitment.

Here's the shape, for a pipeline-generation vendor writing to a firm that opened a new metro office:

Since the Dallas office opened, I'd guess the hard part isn't finding work. It's finding it fast enough to justify the hire before the year turns.

>

We work with advisory firms in exactly that window: the first two quarters in a new market, where the referral base doesn't exist yet and outbound has to carry the pipeline.

>

Worth a five-minute conversation about how the Dallas pipeline is building?

The event appears once, in a subordinate clause. Everything else is about their problem.

What makes signal-led outreach fail

Four failure modes, in rough order of how often I see them.

Leading with the event. The subject line "Congrats on the Series B" tells the reader you're one of fifty. Worse, it makes the signal the point, when the signal is only the reason you were confident enough to write.

Congratulating. Nobody wants congratulations from a stranger who wants something. It reads as a transaction with a bow on it.

Assuming intent. An event means a pain is likely live. It doesn't mean they've decided to buy anything. Write "since you're evaluating new storage vendors" when all you know is that they issued an RFP for tape storage, and you'll get corrected. That burns the credibility the signal just bought you.

Stacking every signal you found. If you know four things about the company, use one. Listing all four is the surveillance move, and it converts an insight into a threat.

How long you actually have

Signal decay gets discussed as though the whole game is speed. Our data says otherwise: the median signal we act on is two months old, and the 90th percentile is around 13 months.

Two months is not slow. For most families it's about right, because the pain a signal reveals takes time to turn into budget and priority. A company that announced an office in May has felt the pipeline gap by July. In the announcement week, everyone was posting congratulations and nobody had felt anything yet.

The windows do differ by family:

The general principle: the more procedural the event, the tighter the window. RFPs have deadlines. Culture and capability changes don't.

Match the channel to the conviction

Not every signal deserves the same effort, and treating them equally is how teams burn the good ones.

High-conviction, low-volume signals like regulatory actions, RFPs and repeated public complaints should route to a person. A researched call and a considered LinkedIn approach. These fire rarely and they're close to unambiguous; putting one into an automated sequence wastes the best input you'll get all quarter. If the touch is a cold call, know the rules before dialing: the TCPA carries statutory damages of $500 to $1,500 per violating call, and our TCPA compliance guide covers what a sales team actually has to check.

Medium-conviction, higher-volume signals like hiring, expansion and leadership changes are where a well-written email cadence earns its keep, with the message shaped by the signal family rather than the individual event.

Low-conviction signals like funding and general growth news are prioritization input, not a reason to write. Let them move an account up the list; find a better reason to make contact.

This is also where the human half of the work matters most. AI can find the event, connect it to the pain and draft the message. Judging whether a regulator's enforcement notice calls for a careful phone call or no contact at all is a person's job, and it should stay one. Which families warrant which treatment is worth working out before you build the cadence.

Measure replies, not opens

One measurement note, because signal-led outreach gets judged on the wrong number.

The point of working from public evidence isn't a higher open rate. It's a higher rate of useful replies, the ones that say "yes, that's exactly the problem" or "no, we solved that last year, but here's what we're actually dealing with." Both move you forward. Both are impossible to get from a message that didn't say anything specific enough to agree or disagree with.

Track the share of replies that engage with your premise. Selling time is too scarce to spend on polite dead ends: Salesforce's research puts the average rep at about 28% of the week actually selling. If people are replying "not interested" without addressing what you said, the pain link was wrong. You found a real event and drew the wrong conclusion from it. That's a targeting problem, not a copy problem, and rewriting the email won't fix it.

And keep the goal honest. A well-worked signal earns you an interested lead, a person who replied with genuine interest in the problem. Turning that into a meeting, and the meeting into a deal, is still the rep's job. The signal just means you arrived while it was still true.