Mei L.
Spots the accounts slipping — usage drop, sentiment turn, champion change — and runs the save play.
Mei watches the customer cohort for risk signals. Usage drop, NPS dip, support sentiment turn, champion job change. When the risk score crosses a threshold, drafts the save play and routes it to the AE.
What Mei shipped
Mei produces real artifacts. Not summaries.
Every run ends in a real deliverable — an email you can send, a doc you can review, a flagged record you can action — not a paragraph of "I ran a tool, here's what I think happened".
Identified · top match ready
Helix · $48k ARR
Customer · 4 mo · usage −38% · renewal in 47d
At risk · 0.81Usage trend
−38% MoM · last login 11d ago
Sentiment
Last 2 tickets · low-confidence resolution · 🤔
Champion
Cara still at Helix · last touch 21d ago
Renewal
May 17 · 47 days · auto-renew
Save play drafted · routed to AE · proof: Helix-style cohort retention case.
Skills
How Mei works.
Daily risk score
Top decile risk per cohort. Combined: usage trend, sentiment, last touch, champion status.
Save play drafted
Specific to the risk reason. AE walks in with the email already written and the proof attached.
Champion-change alarm
When LinkedIn shows a champion left, Mei pings before the call gets cold.
Renewal radar
Renewal date − 90d trigger lights up. Risk-adjusted forecast lands in the AE’s queue.
Tools Mei uses
Native, no Zapier in the middle. Tool calls land in the audit log.
HubSpotSalesforceMixpanelLinkedInSlack
One real run
Watch Mei work.
Mei L.
Churn Rescue · Customer Success
Anything to save this week?
done · 7.2s · $0.10
3 accounts at risk. Top: Helix (usage −38%, renewal in 47d). Save play queued. AE briefed.
audit trail · 4 steps[show]
→ usage.fetch · cohort active
→ sentiment.scan · last 14d
→ score · combined
→ draft · save play helix
Mei · today
Accounts watched
142
At-risk surfaced
3+1
Saves last 90d
11 / 14
Same team
Mei works with the rest of Customer Success.
Mei is spinning up. Get on the list.
Early access opens shortly. We onboard Customer Success customers in cohorts so the bar stays high.