MQL Nurture Agent
Find the dead MQLs that just came back to life.
Re-screens your historical MQL backlog for new jobs, fundraises, and product launches, then drafts a tailored opener for the contacts that came back to life.
Who it's for: Marketing and revenue leaders sitting on a backlog of MQLs that never converted. Most are dead, but the live ones are buying signals you're missing every week.
How it works
01 · PULL
Slice the backlog
Iterates a date-range slice of your historical MQLs from your CRM. Lifecycle stage = MQL, regardless of whether they ever signed up or converted. The whole backlog is in scope.
02 · CHECK FOR LIFE
Look for what's changed
For each contact, looks at recent CRM activity, current job and company on the contact, and the public web for the company. Has anything actually changed since the original MQL date?
03 · SCORE
Be ruthless
Grades A/B/C/skip on whether there's a real, datable re-engagement signal — new role, new fundraise, new product launch, repeat content engagement. Most are skip. That's the right answer. Quality over quantity.
04 · DRAFT
Reference the actual change
For A/B candidates, drafts a tailored opener that names the specific signal — 'Congrats on the Series B; last time we spoke you were evaluating X, the picture has changed.' Routes to the right rep for a fresh outbound attempt.
Signals it can watch
Recent CRM activity (page views, opens, clicks)
Job changes on the contact
Company fundraises
Product launches
Repeat content engagement
Industry / segment shifts
What you give it
A date range of historical MQLs
Pick the original-MQL window you want to scan. The agent iterates that slice in batches, dedupes against past runs, and ignores anyone already worked.
Your ICP definition (as a skill file)
What 'still a fit' looks like today. Companies and roles drift; the agent re-grades against current ICP, not the one from when they first downloaded a report.
Your writing voice (as skill files)
How you write to MQLs vs. cold prospects. The agent reads these on every run and drafts against them, not a generic LLM voice.
What it produces
A filtered list of live re-engagement candidates
Most of the batch is correctly skipped. The output is the small, high-quality slice with a real, datable reason to reach out now.
Per-candidate signal + suggested play
What changed (the re-engagement signal), what play fits (fresh demo / congrats-on-funding / feature recap / etc.), and which rep gets it.
A drafted tailored opener
Opener that names the actual change, not a generic 'circling back.' Pushed to your outbound platform, ready for human review or auto-send.
How we use it
Imports historical MQLs in date-range batches.
Growth Engineer imports historical MQLs from a date range and runs each through this agent. Most come back as 'still cold' — that's the right answer. The 5-10% with a real signal (new role, new round, fresh activity) come back with a tailored opener and route to the right rep, ready for outbound.
See the skill files it uses →
Tools used
CRMWeb researchOutbound platform
Skills it uses
Want this running for your team?
We deploy this agent against your stack, wire up the routing, and turn it on. You keep the keys, the data, and the rep relationships.
Book intro call