Skip to content
← All posts

Strategy

Signal-based outbound: the buying signals worth acting on (and how to build the system)

A practical guide to signal-based outbound: which buying signals predict deals, how to detect them, score them and turn them into timely, relevant outreach.

Victor Mendonça
Victor Mendonça

Founder, OretonGTM · ·10 min read

Most outbound fails for a boring reason: the right message reaches the right company at the wrong time. The account fits your ICP, the buyer has the problem you solve, but this quarter they’re busy with something else, so your email gets ignored.

Signal-based outbound is about fixing timing. Instead of working through a static list in alphabetical order, you contact accounts when something observable happens that makes your offer more relevant now. Done well, it raises reply rates and gives every email a natural, specific reason to exist.

Done badly, it becomes “Congrats on the funding!” sent to 400 companies a week. This guide covers how to do it well: which signals matter, how to detect them, how to score them and how to write messages around them.

What a buying signal is

A buying signal is an observable event or behavior, tied to a specific account, that increases the odds that the account needs what you sell right now.

Three parts of that definition matter:

  • Observable. You can see it in a data source: a job board, a funding announcement, a LinkedIn profile, your website analytics.
  • Tied to an account. “Interest in sales automation is growing” is a market trend. “Acme posted three SDR roles last week” is a signal.
  • Increases the odds. A signal is evidence, not proof. It shifts probability; it doesn’t guarantee a buying cycle.

Intent data is one category of buying signal, focused on research behavior. I break down the difference in intent data vs. buying signals.

Why signal-based outbound works

Two reasons, and they compound.

Timing. Buying windows open and close. A company that just hired its first VP of Sales is about to rethink its sales process; six months later, the new process is set and changing it is expensive. Contacting accounts inside those windows means more of your emails reach people who are actually open to change.

Relevance. A signal gives you something specific to say. “We help B2B startups book more meetings” is generic. “Teams that add three SDRs at once usually hit a data and deliverability wall in month two; here’s how we’ve seen it handled” is specific to what’s happening at that account. Specific emails get read.

My take: The main benefit of signals isn’t the clever first line. It’s that they force you to have a reason for every email. If you can’t explain why this account, why now, you shouldn’t be sending.

The signals worth acting on

Not all signals are equal. Here are the main categories, roughly in order of how often they predict real deals for B2B startups.

1. First-party engagement

Activity on your own properties: pricing-page visits, comparison-page visits, product sign-ups, email clicks, replies like “not now, try me in Q2”, webinar attendance.

  • Why it’s strong: the account is interacting with you, not just your category.
  • Shelf life: days for website visits; whatever date they gave you for “not now” replies.
  • How to detect: website visitor identification, sequencer webhooks, product events, reply classification.

2. New leaders in the function you sell to

A new VP, head or director in the department that buys your product.

  • Why it’s strong: new leaders have a mandate to change things, usually in the first months, and often bring tools and vendors they trusted before.
  • Shelf life: most receptive in roughly their first 90 days.
  • How to detect: LinkedIn job change data, enrichment providers that track role changes, Clay signal monitoring.

3. Champion job changes

A person who used or bought your product at a previous company moves to a new company that fits your ICP.

  • Why it’s strong: they already know the value. This is often the highest-converting outbound signal there is.
  • Shelf life: a few months, while they’re setting up their new team.
  • How to detect: track your customers’ and past users’ contacts over time. More in job change signals.

4. Hiring for roles your product supports or replaces

Job posts that reveal a priority: an account hiring five SDRs, a first RevOps hire, a data engineer for a migration you can help with.

  • Why it’s strong: job posts are public statements of priority and budget. The job description often tells you which tools they use and which problems they’re trying to solve.
  • Shelf life: while the role is open, plus a few weeks.
  • How to detect: job board scraping or a job-post signal provider. More in hiring signals.

5. Funding rounds

A new round, especially Seed to Series B.

  • Why it’s useful: new capital comes with growth targets, hiring plans and budget.
  • The catch: everybody watches funding announcements. Newly funded companies get flooded with vendor emails in the following weeks. The signal works best combined with a second one (funding and hiring for the role you support).
  • Shelf life: one to three months.
  • How to detect: funding databases, news feeds, Clay. More in funding signals.

6. Technology changes

Adopting, dropping or outgrowing a tool in your ecosystem: switching CRMs, adding a tool that integrates with yours, removing a competitor.

  • Why it’s useful: tech changes often come with projects, budgets and integration needs.
  • Shelf life: weeks to a couple of months around the change.
  • How to detect: technographic providers, job posts that mention the stack, public integration pages. More in technographic targeting.

7. Company events

Expansion into a new country, a new product line, an acquisition, a new office, a regulatory deadline.

  • Why it’s useful: events create projects. Projects create budgets.
  • Shelf life: varies widely; tie it to the event’s timeline.
  • How to detect: news monitoring, press releases, AI research agents that read company news.

8. Third-party intent

Research surges on topics related to your category, aggregated by intent vendors.

  • Why it’s useful: reach. It can surface accounts that haven’t interacted with you.
  • The catch: it’s inferred, noisy and usually company-level only. Use it to prioritize, not as a trigger on its own.

Signals compared

Signal Strength Contactable person? Message angle Typical shelf life
Pricing/demo page visit Very high Not always Indirect (don’t mention it) Days
Champion job change Very high Yes Very strong Months
New leader in buying function High Yes Strong ~90 days
Hiring for supported roles High Hiring manager Strong While role is open
Tech change Medium–high Sometimes Strong if relevant Weeks
Funding round Medium Founders/leaders Overused 1–3 months
Company events Medium Sometimes Strong if relevant Varies
Third-party intent surge Low–medium Rarely Weak Weeks

How to pick your signals

Don’t start from a vendor’s list of 40 signals. Start from your own wins.

  1. Pull your last 15 to 20 closed-won deals (or your best-fit customers, if you have fewer deals).
  2. For each, write down what was happening in the 90 days before the first conversation. New leader? Raised? Hiring? Switching tools? Expanding?
  3. Look for patterns that repeat in at least three or four deals. Those are your candidate signals.
  4. Check detectability. Can you find this signal reliably across your TAM with available data? A signal you can’t detect at scale is a nice insight but not a workflow.
  5. Pick two or three to start. Add more only after you’ve seen the first ones work.

If you don’t have closed-won history yet, use your best hypothesis and the signals in the comparison table above, but treat the first two months as a test.

Building the signal system

The architecture I use for most startups has five layers. You can build it in Clay, n8n or a mix, depending on your stack.

Layer 1: The account universe

Signals only matter inside your ICP. Start with a fit-qualified list of target accounts: your TAM mapped and filtered by your ICP. Signals are then monitored against this list, not across the whole internet.

Layer 2: Detection

For each signal, set up a source and a refresh cadence:

  • Job posts: weekly scrape or provider feed for target accounts
  • Job changes: weekly check on key personas at target accounts, plus your champion list
  • Funding and news: daily or weekly feed
  • Website visits and email engagement: real-time, via webhooks
  • Tech changes: monthly technographic refresh

Layer 3: Qualification

Raw signals are noisy. A job post for “Sales Development Representative” at a 5,000-person company means something different than the same post at a 40-person startup. Add a qualification step, often an AI step, that answers:

  • Is this signal relevant to what we sell?
  • How strong is it?
  • Who at the account is the right person to contact because of it?

An LLM reading the job description or news article can classify this well if you give it clear criteria and ask it to quote its evidence. See AI lead qualification and how to stop hallucinations.

Layer 4: Scoring and routing

Combine fit and signal strength into a score with decay, so accounts rise when signals fire and fall as signals age. Then route:

  • Hot (first-party intent, champion move): straight to the account owner, with a task and a Slack alert. A human writes this one.
  • Warm (new leader, hiring, tech change): into a signal-specific sequence with personalized openers.
  • Cool (funding alone, intent surge alone): raise the account’s priority in the regular sequence, don’t trigger a special one.

I cover score design in the account scoring model guide.

Layer 5: Feedback

Track every signal as a field on the account and the contact. After each month, compare positive reply rate and meeting rate by signal type. Kill the signals that don’t outperform your baseline. Double down on the ones that do. Without this step you’ll never know which signals are real.

Writing messages around signals

A signal is context, not content. The goal is to use it to make an informed guess about what the buyer is dealing with, then offer something useful about that.

A simple structure:

  1. Signal, briefly. One clause, not a paragraph.
  2. Implication. What that event usually means for someone in their role.
  3. Relevant point of view or proof. What you’ve seen work, or how you help.
  4. Low-friction ask. An interest-based question, not a 30-minute meeting request.

Weak (signal as gimmick):

Hi Maria, congrats on the Series A! I’d love to show you how we can help you grow. Do you have 30 minutes next week?

Better (signal as context):

Hi Maria — saw Acme is hiring four SDRs after the Series A. When teams ramp that many reps at once, the bottleneck is usually lists and inbox capacity, not headcount: reps end up building their own lists and burning new domains in month one.

We set up the data and sending infrastructure so new SDRs start with researched accounts on day one. Worth sending over how we’d structure it for a four-person team?

The second email would still make sense if you removed the funding round. The signal improves the timing and the opening; it doesn’t carry the whole message.

For more on structure and CTAs, see how to write a cold email and cold email CTAs.

Signal-based vs. list-based outbound

Signal-based outbound doesn’t replace a well-segmented, steady outbound motion. It complements it.

Signal-triggered track Steady ICP track
Volume Low, unpredictable Predictable
Reply rate Higher Lower
Personalization High, often human-written Segment-level plus AI research
Main job Catch accounts in buying windows Build awareness and pipeline across the ICP

Most teams I work with run both. The steady track covers the ICP over a few months; the signal track pulls accounts out of the queue when something happens.

Common mistakes

  1. Using signals without fit. A perfectly timed email to the wrong company is still a wasted email.
  2. Too many signals at once. Twelve signals with arbitrary weights produce a score nobody trusts.
  3. Relying only on funding. It’s the most watched signal there is. Pair it with something else.
  4. Ignoring first-party signals. Your pricing-page visitors and “not now” replies are your best leads, and they’re free.
  5. Leading with creepy signals. Never mention website visits or intent data in the email.
  6. No decay. A new leader from eight months ago is no longer new.
  7. No measurement. If you can’t show that signal-triggered accounts book more meetings, you’re running on faith.

How to start this month

If you want a small version running in a few weeks:

  1. Write down your ICP filters and build a target account list.
  2. Pick two signals from your closed-won history. For most startups, “new leader in the buying function” and “hiring for roles we support” are good defaults.
  3. Monitor them weekly against your account list (Clay is the fastest way; see the Clay tutorial).
  4. Route first-party signals (pricing-page visits, positive replies) to the owner the same day.
  5. Write one sequence per signal, with a human-reviewed opener.
  6. After six weeks, compare results with your regular outbound and decide what to keep.

That’s enough to know whether signals work for your market, without buying an expensive platform first.

Frequently asked questions

What is signal-based outbound?

An outbound approach where you prioritize and time outreach based on observable events at target accounts, like funding, hiring, leadership changes, technology changes or website visits, instead of contacting a static list in random order.

What are the best buying signals for B2B outbound?

The best signals are the ones that preceded your own closed-won deals. Common strong ones are a new leader in the function you sell to, hiring for roles your product supports, a recent funding round, a tech stack change in your ecosystem, a former customer champion changing jobs and first-party activity like pricing-page visits.

How do you find buying signals automatically?

Use a data orchestration tool such as Clay, or workflows in n8n, to monitor job posts, funding databases, LinkedIn job changes, technographic data and your own website and email activity across a target account list, then push qualified signals to your CRM or sequencer.

Is signal-based outbound better than list-based outbound?

It usually produces higher reply rates per email because timing and relevance are better, but it produces fewer accounts per week. Most teams run both: a signal-triggered track for hot accounts and a steady, well-segmented track for the rest of the ICP.

Should I mention the signal in my email?

Mention public signals (a funding round, a job post, a new role) when they're the reason the email is relevant, and connect them to a point of view. Don't mention inferred or private signals like website visits or intent data.

How long is a buying signal relevant?

Website visits and demo-page activity are hot for days. Hiring signals stay relevant while the job is open. New leaders are most receptive in roughly their first 90 days. Funding signals are most useful in the first one to three months after the announcement.

How many signals do I need to start?

Two or three. Start with the signals you can detect reliably and that appear in your closed-won history. Adding a dozen signals at once makes scoring opaque and messaging generic.

Ready to stop doing outbound by hand?

Book a strategy call. We’ll map your ICP, review what’s working (and what isn’t) and show you exactly what we’d automate first.

Book a strategy call

Keep reading