Use Case 1
Weekly Lead Scoring Run
Trigger: Scheduled weekly scoring cycle
1.Pull all leads and current ICP profiles
2.Evaluate firmographic fit and engagement signals
3.Assign a 0–100 score and Hot/Warm/Cool band
4.Write a one-line rationale per lead
5.Track score movement since the prior run
6.Surface the top-ranked leads
✓
Reps work a ranked, rationale-backed lead list refreshed every week.
Use Case 2
Feedback-Loop Calibration
Trigger: A rep disputes a lead's score
1.Explain the score's drivers
2.Distinguish model signal from verified rep intent
3.Capture a disqualification tag with reason
4.Feed the outcome back into scoring
5.Down-weight similar misleading patterns
✓
Scoring improves over time as rep ground-truth corrects the model.
Integrations
CRM (Salesforce, HubSpot)
ICP Profiles
Engagement & Activity Signals
Dashboards & Charts
Lead Score Records
Industries
all
See It In Action
A sample conversation with this agent
Run this week's lead scoring.
Scoring all leads against the current ICP profiles and engagement signals—firmographic fit, recent activity, content engagement, and source quality. Each lead gets a 0–100 score, a Hot/Warm/Cool band, and a one-line rationale. I'm surfacing the top 10 ranked, plus the biggest score movers since last week. Nine leads jumped into Hot on new engagement; three cooled on inactivity. Want the full ranked list or just the Hot band?
A lead scored Hot but the rep says they're a tire-kicker—what gives?
The score reflects ICP fit and engagement signals, not intent quality the model can't see. This lead fits the ICP well and engaged heavily—high signal—but if the rep has direct evidence they're not buying, that's ground truth the model should learn from. I'll let you tag it 'disqualified—rep verified,' which feeds back so similar engagement patterns aren't over-weighted. The score is a prior, not a verdict.
Which leads moved the most and why?
Top movers: a mid-market lead jumped 38 points after downloading two assets and opening three emails this week—now Hot. An enterprise lead rose 22 on a new ICP match (we updated the EHR-environment weight). Two cooled 15+ on 30 days of silence. The movements track the signals, so reps can prioritize the rising leads rather than re-working a static list. Movement is often a better prompt than absolute score.