Health-score read & churn-save play
Builds on Topic 7.
What you'll produce
A health-score read and churn-save play for Brightline Logistics: a one-page assessment that rolls four signal categories — usage, engagement, support, and sentiment — into a single red/amber/green score with a written rationale, the early-warning signs you're watching, and one concrete save play triggered by a real at-risk signal (find the real reason → intervention → escalation plan → follow-up). This is the artifact that proves you can do the central CSM job from Topic 7: use metrics as an early-warning radar instead of a post-mortem. Anyone can read a green dashboard; a paid CSM is worth it because they catch the amber-going-red account before the customer has quietly decided to leave — and then runs a calm, coordinated save instead of a panicked renewal-week scramble. For Brightline specifically, the renewal is 12 months out and the CFO wants ROI proof, so a documented health read is also your evidence trail: it shows leadership you saw the risk early and acted, which is exactly what makes you a strategic partner rather than a ticket-router.
Instructions
- List your signal sources first. Before scoring anything, write down where each signal actually comes from — your CS platform / product analytics (logins, feature usage), your CRM and calendar (meeting attendance, email replies), your support/ticketing tool (ticket volume, severity, CSAT), and survey data (NPS, last QBR sentiment). If a signal is a guess, label it a guess. Real CSMs score on data they can point to, not vibes.
- Build a 4-category scorecard. Use the four standard health dimensions and weight them: Usage (40%), Engagement (25%), Support (20%), Sentiment (15%). Usage gets the most weight because, per Topic 7, quietly declining usage is the strongest predictor of churn. For each category, write the raw signal, then rate it red / amber / green.
- Roll up to one overall color — and apply the override rule. Compute a rough weighted color, but then apply the rule every good CSM uses: any single red in a high-weight category caps the overall score at amber, and a departing champion or a buyer-level threat caps it at red regardless of usage. A health score is a floor, not an average — one fatal signal outranks three healthy ones.
- Write the rationale in 2–4 sentences. State the color, the one or two signals driving it, and what it means for the renewal. This is what you'd paste into your CRM and what your manager reads in a forecast review. No jargon-as-decoration: name the number.
- List 3–5 early-warning signs you're actively watching — the specific, measurable thresholds that would move Brightline from amber toward red (e.g., "logins per dispatcher drop below X," "champion goes two cycles without replying"). These are your tripwires, not vague worries.
- Pick one real at-risk trigger and build the save play. Choose a trigger that's plausibly already firing for Brightline. Then write the play in five parts, mirroring the Topic 8 pattern: (a) find the real reason — the diagnostic step, because the stated reason usually hides the true one; (b) the intervention — what you do, with whom, by when; (c) the escalation plan — who you bring in, the exact context they need, and the trigger that fires it; (d) the follow-up — how you confirm the save held and feed it back into the health score; (e) the success measure that tells you the account moved back to green.
- Tie it to revenue. Close by naming the dollars and the metric at stake (the $48k ARR and its effect on NRR) so the read connects to what leadership cares about — not just a feelings check.
Worked example
(Account: Brightline Logistics — $48k/yr RouteIQ contract, month 5 of 12. Champion: Maya, Head of Operations. Economic buyer: the CFO. 15 dispatchers. Renewal in 7 months.)
Overall health: AMBER (trending red) 🟠
Scorecard
| Category (weight) | Signal (the data) | Rating |
|---|---|---|
| Usage (40%) | 9 of 15 dispatchers logged in last week (60% seat activation, down from 80% in month 3). Route-optimization feature — the one tied to empty-mile reduction — used on only 41% of dispatches, vs. the 70% target set in onboarding. Daily active dispatchers slid from 11 → 9 over six weeks. | 🔴 Red |
| Engagement (25%) | Maya attended the month-3 QBR and replies within a day, but the last two check-ins were rescheduled by her team and one was a no-show. CFO has never joined a call. | 🟠 Amber |
| Support (20%) | 4 tickets in 60 days, all low-severity ("how do I…" questions); none about outages. Average CSAT on tickets 4.6/5. Low volume here is actually healthy. | 🟢 Green |
| Sentiment (15%) | NPS from Maya: 7 (passive). Verbatim from last QBR: "The dispatchers who use it like it, but I'm fighting to get the rest to switch off their spreadsheets." No buyer-level sentiment captured yet. | 🟠 Amber |
Rationale (for the CRM / forecast review): Brightline is amber, trending red. Support and CSAT are fine, but the signal that matters most — usage — is red: only 60% of dispatchers are active and the core empty-mile feature is at 41% adoption against a 70% goal, so we are not yet delivering the ROI the CFO will demand at renewal. The single red in our highest-weight category caps this at amber under our override rule; if route-optimization usage keeps sliding we will have no value story in 7 months and the renewal becomes a real loss. The relationship with Maya is intact, which is our lever to fix it now.
Early-warning signs I'm watching (my tripwires)
- Seat activation drops below 55% (8 of 15) or any further week-over-week decline → move to red.
- Route-optimization adoption stays under 50% at the month-6 mark → no defensible ROI for the QBR.
- Maya misses or reschedules a third consecutive touchpoint → champion disengaging; risk she's lost the internal fight.
- Any dispatcher reverts publicly to spreadsheets or a "do we still need this?" comment surfaces from the CFO's office → buyer-level doubt, escalate immediately.
- A key contact leaves (Maya changes role, or a power-user dispatcher exits) → champion/relationship risk, re-baseline the account.
Save play — Trigger: route-optimization adoption stuck at 41%, six dispatchers effectively inactive
(a) Find the real reason. The stated reason ("dispatchers are busy / wary of new tools") is rarely the whole story. I will not assume it's a training gap. Within 5 business days I'll: (1) pull the per-dispatcher usage report to see who the six inactive users are; (2) ask Maya for a 20-minute call to learn what she's hearing internally; and (3) request 15 minutes each with two of the inactive dispatchers to watch them actually dispatch a load. Hypothesis to test: the spreadsheet-bound dispatchers aren't refusing the tool — they hit a specific friction (the route-optimization output doesn't account for one customer's delivery-window quirks, so they don't trust it and fall back to the sheet). If true, this is a workflow/trust problem, not a "people won't change" problem — and it's fixable.
(b) The intervention. Once I've confirmed the real blocker: (1) run a focused 45-minute working session with the six inactive dispatchers — not a generic re-train, but solving their actual dispatch on the tool, live; (2) if the delivery-window quirk is real, configure RouteIQ's constraint settings to handle it (or log a precise product request with the data); (3) appoint the most enthusiastic power-user dispatcher as an internal peer-champion so adoption is driven by a colleague, not just the vendor; (4) set a concrete 30-day target with Maya: route-optimization adoption from 41% → 65%, seat activation back to 80%. I'll send Maya a one-page recap so she has ammunition for her internal fight.
(c) Escalation plan. Trigger to escalate: if the real blocker is a genuine product gap I can't configure around, OR if after the working session adoption hasn't moved by the next check-in. Who I bring in: (1) a RouteIQ product/solutions engineer for the constraint-handling gap, and (2) my CS manager for visibility on a strategic at-risk account. Exactly what they need from me (so they can help fast): the account is $48k ARR renewing in 7 months; champion is Maya (Head of Ops), buyer is a skeptical CFO who wants ROI proof; the blocker is [specific delivery-window constraint] affecting 6 of 15 dispatchers and capping the empty-mile ROI; here's the usage data and the dispatch example that breaks; what I need is [a config path or a committed fix date] within 2 weeks so I can hold the renewal value story. I stay the single point of contact — Maya never gets passed around.
(d) Follow-up. I'll re-pull the usage report weekly for 30 days and confirm the trend, then send Maya a short "here's where adoption moved" note. At day 30 I update the health score in the CRM with the new numbers and a fresh rationale. Critically, I bank this as QBR evidence: "we found 6 dispatchers blocked by X, fixed it, and moved empty-mile-relevant adoption from 41% to 65%" — the exact ROI proof the CFO needs at renewal.
(e) Success measure. Account moves amber → green when: route-optimization adoption ≥ 65%, seat activation ≥ 80% and stable for 3 weeks, and Maya's NPS moves from 7 toward 9. The real prize: a defensible empty-mile-reduction number to put in front of the CFO.
Revenue tie-in. This is $48k of ARR on the line. Saved and on track, Brightline renews and becomes an expansion candidate (more seats / the analytics module) — pushing our Net Revenue Retention above 100%. Lost, it's a direct hit to NRR and a reference customer we'll never get. The cost of this save play is about three hours of my time over two weeks; the math is not close.
Rubric
The app's AI scores the learner's submission against these criteria and gives feedback. Levels: Needs work (1) / Solid (2) / Excellent (3). Passing = every criterion at Solid or above.
- Multi-signal health score with rationale — 1: a single gut-feel color or unweighted list · 2: rolls usage/engagement/support/sentiment into one red/amber/green with reasons · 3: weighted scorecard with real data per signal, a stated override rule (one red caps the score), and a crisp rationale a manager could paste into a forecast.
- Usage prioritized as the lead churn signal — 1: usage treated as one of many or ignored · 2: usage included and weighted heavily · 3: usage is the dominant signal with concrete numbers (activation %, core-feature adoption vs. target) and explicitly tied to the value/ROI story.
- Specific, measurable early-warning signs — 1: vague worries ("watch engagement") · 2: 3–5 named signals · 3: 3–5 tripwires with concrete thresholds that would actually move the color and trigger action.
- Save play finds the real reason — 1: jumps straight to a generic fix · 2: includes a diagnostic step · 3: a real diagnostic plan (who/what/by when) that tests a specific hypothesis and separates the stated reason from the true blocker.
- Intervention, escalation, and follow-up are concrete — 1: vague good intentions · 2: an intervention plus an escalation and a follow-up · 3: named people, exact escalation context and trigger, owner, dates, a success measure, and a follow-up that feeds back into the health score.
- Tied to revenue and the renewal — 1: a feelings-only check disconnected from money · 2: mentions the renewal/value · 3: names the ARR at stake and its impact on NRR, framing the save as protecting renewal and seeding expansion.