Warmly + Make: Signal-Based Outbound Workflow Guide

Most outbound teams are still working off static lists pulled weekly from Apollo or ZoomInfo, blasting sequences at leads who went cold days ago. The teams actually building pipeline right now do the opposite. They trigger outreach the moment a high-fit account lands on the pricing page, at exactly the moment that person is thinking about the problem you solve. That is signal-based outbound. And Warmly plus Make is one of the cleanest implementations I have seen hold up in production.

Signal-based outbound is not about personalization at scale. It is about timing. The same message sent two minutes after a pricing-page visit converts at a materially different rate than the same message sent two weeks later.

The Stack and Why It Works

The core idea is simple. Warmly identifies a visitor, fires a webhook to Make, Make enriches and routes the lead, and HubSpot enrolls them in the right sequence automatically. No rep manually checking a dashboard. No lead slipping through a 48-hour window.

Here is why this combination outperforms Warmly’s native automations alone. Warmly’s built-in playbooks are solid for Slack alerts and basic CRM writes. They do not give you multi-step conditional logic for ICP filtering, deduplication, and enrichment branching. Make fills that orchestration gap. Clay fills the data quality gap, running waterfall enrichment across Apollo, LinkedIn, and Clearbit in a single HTTP call from Make. The result is a visitor signal that arrives in HubSpot as a fully enriched, correctly routed contact ready for sequence enrollment, not just a company name and a URL.

I have built this stack for clients at Homegrown Growth Co. The first time it fires correctly in production is genuinely satisfying. A VP of Engineering visits your integration page, Warmly fires within seconds, Make routes them to a Clay table, Clay appends their mobile number and confirms they are in a 200-500 person ICP company, and HubSpot enrolls them in the technical sequence before the rep has finished their coffee.

Building the Workflow Step by Step

// Clay API response - name your columns cleanly
{
  "email_verified": "ian@example.com",
  "job_title": "VP of Engineering",
  "seniority": "vp",
  "company_headcount": "250",
  "mobile_phone": "+14155550123",
  "linkedin_url": "https://linkedin.com/in/example",
  "funding_stage": "Series B"
}

// Make HTTP module: map each key explicitly
// Do NOT rely on auto-mapping for Clay responses
// Set "Parse response" to Yes and reference
// {{1.email_verified}} not {{1.`Work Email Verified`}}

Choosing Your Sequencing Layer

The workflow above writes to HubSpot sequences. Depending on your stack, you might route elsewhere. Here is how I think about the sequencing layer choice for this specific workflow. Skip Outreach and Salesloft unless your team is already fully committed to them, the HubSpot native modules in Make are cleaner and you will spend far less time mapping fields.

Which sequencing tool fits this workflow?

Choose HubSpot Sequences if

• Your CRM is already HubSpot and reps live in it daily
• You need native contact enrollment with CRM property writes in the same step
• You want reply detection that auto-pauses the sequence without extra config

Included in Sales Hub Pro+ Try HubSpot →

Choose Smartlead if

• You are sending high volume and need multi-inbox rotation to protect deliverability
• Your team runs outbound outside HubSpot and syncs replies back via webhook
• You want per-mailbox warm-up managed automatically

From $39/mo Try Smartlead →

Choose Lemlist if

• Your sequences include LinkedIn steps, video, or dynamic image personalization
• You want built-in multichannel cadences without stitching separate tools
• Signal-triggered outreach will mix email and LinkedIn DM in the same sequence

From $59/seat/mo Try Lemlist →

What This Workflow Actually Produces

Once this is running, you get three things a manual or batch-based process cannot replicate. First, timing advantage: the average rep responds to a high-intent signal in somewhere between 30 minutes and 2 hours according to InsideSales research, while this workflow responds in under 60 seconds. Second, data quality: because Clay runs enrichment at trigger time rather than at list-pull time, the contact data is current, not 30 days stale. Third, rep focus: instead of triaging a Warmly Slack channel full of alerts, reps see contacts already in their HubSpot sequences with full context on what page was visited and why the outreach fired.

For teams thinking about building a broader signal-based lead scoring system on top of HubSpot, this workflow pairs naturally with Trigify or similar intent enrichment layers feeding the same score object.

The one thing I would not skip is the deduplication filter. Without the warmly_enrolled property check, a prospect who visits your pricing page three times in a week gets enrolled three times. I have seen this happen in a client’s HubSpot instance. It is the fastest way to get your domain flagged and your sales team embarrassed.

Ship It This Week

The tools are mature. The webhook infrastructure is reliable. Warmly’s de-anonymization layer combined with Clay’s enrichment waterfall gets you to sequence-ready in under five minutes per visitor. Make is the right orchestration layer here because it gives you visual debugging, native HubSpot modules, and enough conditional logic to handle routing without writing a line of custom code.

Don’t overthink the rollout. Build the simple version today. Measure reply rates against your baseline batch sequences for 30 days. The data will make the case better than any planning doc.