Clay playbook · Problem 5 of 5 · Data orchestration

Signal-Based Outbound: Timing Over Volume

Cold outreach replies at 3 to 5 percent; signal-timed, personalized outreach replies at 15 to 25 percent. The difference is not the copy, it is the moment. This playbook builds the Clay system that watches for buying signals and launches the right play while the window is open.

By Langley Erickson · CascadeGTM · July 2026

The problem

Spray and pray treats every account as equally ready, so everyone gets the same message at the wrong time. Meanwhile the actual buying moments, a new VP of Sales in seat, a funding round, a hiring spike, a competitor rip-out, pass unnoticed because no one is watching thousands of accounts by hand. The signal exists; there is just no system connecting it to a touch.

This playbook is part of the companion set to Data Orchestration in 2026: Why Clay Sits Under Every Modern GTM Stack. Read that first for the strategy; this is the build.

What you will build

A set of Clay signal watcher tables (job changes, hiring, funding, intent) feeding one In-Market Accounts table that scores signal strength, suppresses accounts already in play, has a Claygent draft the why-now angle, and routes each account to the matching sequence in your sender. Reps wake up to a short list of accounts with a reason to reach out today.

The stack

Clay (Growth plan)Signal tracking, Claygent, orchestration and sequencer push.
Clay Signals: job changes, promotions, hiring, newsNative trackers watch your named accounts and past champions.
Intent: Bombora or 6sense (optional but potent)Third-party intent feed imported to Clay; buy the feed, build the logic.
Sender: Outreach, Salesloft or SmartleadRuns the play; Clay decides who enters which sequence and why.
CRM: Salesforce or HubSpotSuppression source (open opps, active sequences, recent closed-lost) and destination for signal history.

Architecture

Many watchers, one brain: signals stream into their own tables, the In-Market table consolidates and scores them, suppression protects live deals, and only then does a play fire.

Signal watchersjobs / hiring / funding / intent
In-Market tablescore + dedupe
SuppressionCRM open-opp check
Claygentwhy-now angle
Sendermatched play
Many watchers, one brain, and suppression before anything fires.

Clay tables and connected objects:

Job Change Tracker (Clay table) · Clay signal on past champions and target personas; fires when a tracked person lands a new role.
Hiring & Funding Watcher (Clay table) · Signals on named accounts: open roles matching your pain profile, funding events, leadership news.
Intent Feed (Clay table) · Weekly Bombora or 6sense import: account, topic, surge score.
In-Market Accounts (Clay table) · The brain. One row per account with active signals, combined signal score, suppression status, Claygent angle, assigned play and sequence status.
CRM objects · Open Opportunity and recent activity checks power suppression; signal history writes back to the account timeline.

The build, step by step

1

Decide which signals actually predict revenue for you

Pull your last 30 closed-won deals and ask what was true in the 90 days before the first meeting. Common winners: a champion changed jobs (3x more likely to buy), the account was hiring for the roles your product serves, new funding, or a leadership change in the buying function. Pick 3 or 4 signals with evidence in your own history; more watchers than that just makes noise.

2

Stand up the watchers

In Clay, create the signal tables:

  • Job changes: import past champions and closed-won contacts, enable the job-change signal. When one moves, the row fires with the new company.
  • Hiring: track open roles on named accounts with a jobs source, filtered to titles that indicate your pain (for CascadeGTM clients that is often RevOps, GTM engineering, sales leadership).
  • Funding and news: Clay's company news signal on the target account list.
  • Intent: schedule the Bombora or 6sense CSV or API import weekly; keep only accounts above your surge threshold.
3

Consolidate and score in the In-Market table

Each watcher writes matched accounts into In-Market via lookup (dedupe on domain, courtesy of playbook 1 hygiene). Score signal strength additively:

Formula prompt: signal score
Start at 0. Add 40 for a champion job change into the account.
Add 25 for 3+ relevant open roles. Add 20 for funding in the last 60 days.
Add 15 for intent surge above threshold. Add 10 for each additional
distinct signal type. Cap at 100. Return the firing signals as a list.

Accounts at 60+ enter plays automatically; 30 to 59 go to a rep-review view; under 30 just accumulate history.

4

Suppress before you send

Nothing torches trust like a cold sequence hitting an account your AE is closing. Add CRM lookups and a suppression formula: suppress if open opportunity exists, OR any contact is in an active sequence, OR closed-lost within 90 days, OR the account is a customer. Suppressed rows stay visible with the reason, because AEs still want to know their deal just showed intent; that alert goes to the AE instead of a sequence.

5

Generate the why-now angle

A signal-timed email still has to name the signal well. Claygent column on play-eligible rows:

Claygent prompt: why-now angle
Account: {{Company}} ({{Domain}}). Firing signals: {{Signal List}} with
details: {{Signal Details}}. We sell {{Your Product, one line}} to {{ICP}}.

Verify the strongest signal from public sources, then write a 2-sentence
opening for a cold email. Sentence 1 references the verified signal
specifically (name the role they posted, the round they raised, or the
person who just joined; include the source). Sentence 2 connects that
moment to the problem we solve. Factual, specific, no flattery.
Return JSON: {"angle": "...", "signal_used": "...", "source_url": "...",
"confidence": 0.0-1.0}
6

Route to the matching play

Map signal type to sequence: job-change plays reference the new role, hiring plays reference the team they are building, funding plays reference the mandate. Push contacts to the sender with {{angle}} and {{signal_used}} as custom variables, and write the signal event to the CRM account timeline so every later touch has the context.

7

Review the misses, not just the hits

Monthly, pull closed-won deals that never appeared in In-Market and ask which signal would have caught them. That is how the watcher set evolves from your first guess into your actual buying pattern.

Measuring success

Judge this playbook on reply quality and meetings per hundred in-market accounts, against your cold baseline, and on how fast a signal becomes a touch.

MetricWhere to measureTargetYou know it is solved when
Reply rate on signal playsSender reports vs cold baseline15 to 25% vs 3 to 5% coldReplies reference the signal back to you
Signal-to-touch timeIn-Market timestamps: signal fired to sequence startUnder 48 hoursCompetitors show up to the same trigger a week after you
Meetings per 100 in-market accountsSender + CRM meeting outcomes3 to 8 depending on motionPipeline sourced-by-signal becomes its own reported category
Suppression accuracyAE complaints and audit of sends vs open oppsZero sequences into live dealsAEs ask for the signal alerts instead of asking you to stop

Staffing the workflow

The most involved build of the five, because it composes signals, suppression, Claygent and routing. It is also the one with the most visible revenue fingerprint.

PhaseHoursWho
Build: watchers, scoring, suppression, angles, routing30 to 40 hoursGTM engineer
Rollout: play design with sales, sequence templates, review views10 to 12 hoursGTM engineer + sales manager
Maintenance: signal tuning, suppression audits, miss reviews6 to 8 hours per monthGTM engineer

Maintenance and monitoring

Workflows drift as your GTM strategy, ICP and stack change. Keep this one honest with the following cadence.

Related

Want this running in your stack?

I build Clay-powered data orchestration systems for growth-stage GTM teams: this playbook and the four beside it, tuned to your ICP, CRM and motion.