Tools

What You'll Build

A daily process that runs while you sleep and delivers 5-10 sales-ready leads to your inbox by the time you wake up. Each lead is:

  1. Triggered by a real signal (just hired a VP of Sales, just posted their first sales engineer role, just got funded, etc.) — not pulled from a static list.
  2. Matched to your ICP automatically. The agent only surfaces companies that fit your sweet spot.
  3. Enriched with a verified business email for the right contact.
  4. Ready to send. Either dropped into your sequencer as a pre-approved lead, or sitting in your inbox for human review and send.

You do not maintain a lead list. The signal feed is the list. The list changes every day because the signals change every day.

Why This Works

For ten years, outbound has been a list game. Buy the list (Apollo, ZoomInfo, Lusha). Enrich it (Clay, Prospeo). Send to it (Smartlead, Manyreach). The math has always been the same: thousands of cold emails, low single-digit reply rates, hope for a meeting.

The list game is breaking. Reply rates keep dropping. Inboxes keep filtering. Buyers know what a cold email looks like in the first sentence.

The signal game is different. Instead of emailing 1,000 strangers, you email 5 people who just took an action that makes them a buyer for what you sell.

A VP of Sales just got hired at a Series B company. The first 30 days, they need: a CRM consultant, a sales ops contractor, an outbound agency, a sales coaching firm. If you sell any of those, you have a 30-day window where this person is actively looking. A cold email that reaches them in that window has a 5-10x higher reply rate than the same email two months later when the new VP is heads-down executing.

The pipeline below catches those moments at scale.

How the Signal Layer Works

You have three options for sourcing signals:

Option 1: Trigify (purpose-built signal layer)

Trigify monitors LinkedIn at scale for specific signal types:

You configure the triggers once. Trigify runs daily and returns the people who matched each trigger. Plug-and-play, no scraper to maintain.

Option 2: Roll your own signal scraper

For specific public data sources (SEC filings, government tenders, news mentions, niche directories), you can build the signal layer yourself. An OpenClaw cron job scrapes the source daily, extracts the new entries, and feeds them into the same pipeline. Eric Nowoslawski's SEC Form 144 scraper is the canonical example: every day's insider stock-sale filings become tomorrow's outbound list for wealth advisors.

This route works when the signal you care about lives on a public site nobody else is watching. It does not work when the signal is "what's happening on LinkedIn." For LinkedIn, use Trigify.

Option 3: Hybrid

Use Trigify for the LinkedIn signals. Build a small scraper for the niche public data source specific to your business. Feed both into the same pipeline. The agent treats every input as another signal type with its own ICP rules.

For most teams, start with Trigify alone. Add the custom scraper only if there's a public source your buyers leave fingerprints on.

The Full Pipeline, Layer by Layer

Layer 1: Signal in

The signal tool delivers a daily file: who triggered, what they triggered on, and the raw data (name, company, role, the trigger event).

Layer 2: ICP filter

Not every signal matches your buyer profile. OpenClaw runs a small ICP filter:

The output is a smaller list: only the signals that match your ICP. Often 20-40% of the raw signal volume.

Layer 3: Contact identification

For each company that passed the ICP filter, identify the right contact:

OpenClaw does the role mapping using the Clay or Apollo data. Output: name, title, LinkedIn URL.

Layer 4: Email enrichment

Prospeo runs the email lookup for each contact. Two-pass approach:

  1. Primary lookup. Prospeo returns the most likely business email and a confidence score.
  2. Verification pass. MillionVerifier confirms the email is deliverable.

Only leads with a verified business email proceed.

Layer 5: Personalization context

For each lead, OpenClaw assembles a short context bundle the sequencer can use:

This context bundle becomes the "personalization variable" in the cold email. Your sequencer uses it to write the opening line of each email so every send feels specific to the recipient.

Layer 6: Delivery

You have two routing options:

After you trust the pipeline, switch to direct delivery. Until then, keep the human review gate.

Step-by-Step Setup

Step 1: Define Your ICP Exactly

Before you touch any tool, write your ICP down with concrete filters:

This document becomes the ICP filter prompt. The pipeline rejects anything that does not match.

Step 2: Pick Your Signals

Write down the 3-5 signals that mean "this person is a buyer right now":

Configure Trigify (or your custom scraper) to watch for these signals exclusively. Resist the urge to track 12 signals. Three is sharper than twelve.

Step 3: Wire OpenClaw

Set up an OpenClaw cron job that runs once daily:

  1. Pull the signal file from Trigify.
  2. Run the ICP filter.
  3. Run the contact identification.
  4. Run the email enrichment and verification.
  5. Assemble the personalization context.
  6. Output the day's leads to your chosen destination.

The whole pipeline is one OpenClaw skill that takes the signal file as input and produces the lead file as output.

Step 4: Configure the Sequencer

Set up one Manyreach (or Smartlead) campaign per signal type. Why one per type: the email copy is different for each signal. A "you just got hired" email is a different message than a "your company just raised" email. Different campaign, different copy, same pipeline feeding both.

Each campaign needs:

Step 5: Run the Pipeline for a Week in Review Mode

For the first week, do not auto-send. Let the pipeline deliver leads to your inbox or Slack. Review each one:

Tune. Iterate. The first week is configuration. The second week is calibration. By week three, the pipeline is producing leads you'd send.

Step 6: Flip on Auto-Send

Once you trust the pipeline, route the output directly into Manyreach. The agent fills your campaigns automatically every morning. You walk in at 9 AM, your campaigns are full, your day starts.

Step 7: Add a Bounce-Rate Watchdog

Eric Nowoslawski's second build in the same video is a bounce-rate watchdog. Run the same OpenClaw pattern but on the sequencer's side: every few hours, check campaign bounce rates. If any campaign goes above your threshold (typically 3-4%), pause it, re-verify the lead emails, and resume.

This protects your sender reputation. The signal pipeline can ingest bad data; the watchdog catches the bad data before it kills your inbox placement.

Adapting This for Your Business

The pattern works for any B2B service or product where a specific event predicts a buying moment.

Outbound agencies. Signal: "company posted a sales role." Triggered outreach: "before you hire, do you want fractional outbound for 90 days?"

HR consultancies and recruiters. Signal: "company posted multiple roles in the same function." Triggered outreach: "if you're hiring a team, here's how we cut your time-to-fill."

Marketing agencies. Signal: "company just raised funding." Triggered outreach: "the first 90 days post-raise are when most companies waste their marketing budget. Here's how we'd structure yours."

Sales coaching firms. Signal: "new VP of Sales hired." Triggered outreach: "if you're the new VP rebuilding the team, here's the playbook we run with our other VP clients."

Real estate investors and brokers. Signal: "owner filed Form 144 to sell securities." Triggered outreach: "if you're about to come into liquidity, here's how three of our clients deployed theirs into real estate."

Wealth managers and financial planners. Same Form 144 signal. Different message.

Compliance and security consultants. Signal: "company published a public privacy or security incident." Triggered outreach: "we help companies that just had an incident rebuild their program in 60 days."

The pattern is always the same. Find the signal that means "buyer right now." Configure the filter. Let OpenClaw run the daily pipeline. Personalize on the signal. Send.

Gotchas and Tips

What This Replaces

Before this stack:

After this stack:

The reply-rate jump is the whole story. A 4-5x improvement in reply rate is the difference between an outbound channel that limps along and one that compounds.

If your business already has cold email infrastructure and a sequencer running, this pipeline is a layer on top of what you have. If you're starting from scratch, the cold email infrastructure (domains, mailboxes, DNS, warm-up) is its own setup that has to come first. Either way, the signal layer is what makes the difference between outbound that converts and outbound that just creates inbox fatigue.

Keep Reading