PMAI PM Playbook

Sales Call CRM Assistant

Medium risk22 min full case
The verdictPilot candidate

Pilot after blockers resolved

What would unlock the pilot
  1. 01Define and execute consent verification for the pilot cohort.
  2. 02Deploy observability dashboards and alerting before pilot start.
  3. 03Calibrate confidence scoring and transcript quality scoring on 50+ examples each.
  4. 04Train pilot cohort on review workflow.
  5. 05Verify transcript retention and export controls for pilot data.

The case in four facts

The problem

Sales reps spend 45 min/day on post-call CRM updates. 30% of fields are empty or outdated. Pipeline forecasting suffers.

The AI's job

Extract structured CRM fields from sales call transcripts and present them for rep approval before writing to Salesforce.

Without AI

Reps manually update Salesforce from memory and notes after each call.

Expected outcome

Reduce time spent updating the five target fields by at least 50% during pilot. Increase field completion from 70% to 95% on those fields.

OwnerSarah Chen, VP ProductStageprototypeUsersEnterprise Account Executives · SDRs using Gong/Chorus with Salesforce

Readiness at a glance

3.07/ 5
Pilot candidate

The weighted score of 3.07 falls in the pilot candidate range. The problem fit, AI job definition, and suggest-only workflow justify preparing a controlled 12-rep pilot, but launch-critical gaps in data readiness (2/5), system behavior (2/5), and observability (2/5) must close before the pilot starts.

  • Problem fit
    4/5
  • Workflow fit
    4/5
  • AI job definition
    4/5
  • Data readiness
    2/5
  • Eval readiness
    3/5
  • System behavior
    2/5
  • Risk & safety
    3/5
  • Regulatory readiness
    3/5
  • Cost & business case
    4/5
  • Observability
    2/5
  • Launch & ops
    3/5

Where the score lands

3 dimensions to fix
  • Data readiness2/5

    18% of calls lack consent flag. Cannot process those calls.

    NextAudit consent flag coverage for pilot cohort. Build automated consent check. Implement transcript quality filtering.

  • System behavior2/5

    Confidence scores not validated against actual accuracy. May mislead reps.

    NextCalibrate confidence scoring against ground truth. Calibrate quality scoring on 50 manually-rated transcripts. Load test pipeline at 2x expected usage.

  • Observability2/5

    Dashboards not built. Cannot observe pilot without them.

    NextBuild dashboards and deploy logging to staging before pilot start. Configure alerting for fabrication and acceptance rate.

And before production

7 more gates after the pilot

  1. 01Automated consent verification flow not built.
  2. 02Eval set only 25% complete.
  3. 03Confidence scoring not calibrated against ground truth.
  4. 04Rep training not conducted.
  5. 05Observability dashboards and alerting not deployed.
  6. 06Transcript retention and export controls not verified in staging.
  7. 07Load testing not completed.
Alternative path

What the team could do instead

If consent flag coverage cannot reach 95%+, consider restricting to calls from the Gong-native dialer where consent is automatically flagged, rather than processing all call sources.

A medium-risk structured extraction workflow that updates Salesforce from sales call transcripts. The model accuracy is the easy part. The trust envelope around it — consent, calibration, and pilot scoping — is the actual PM job.

The situation

Sarah Chen is VP Product at a mid-market revenue tooling company. Her team's flagship integration sits in Salesforce. The CRO has been complaining for two quarters that pipeline forecasts are unreliable because reps don't keep their opportunities current. A time-tracking study on 40 AEs showed 45 minutes a day of post-call CRM updates, and audits showed 30% of pipeline fields were empty or stale within five days of a call.

The team prototyped an extractor on Claude Sonnet that reads Gong and Chorus transcripts and proposes updates to five fields: stage, next step, decision criteria, competitor mentioned, and budget signal. The rep sees the proposed updates inside Salesforce, accepts or edits, and the writes go through. The prototype is accurate enough at first glance that the CRO is asking when she can roll it out to "all 80 reps next month."

Sarah brought back a scoped 12-rep, 6-week pilot proposal instead. This case study is the reasoning.

What's working

  • The problem is real and recurring. 45 minutes a day per rep is 250+ hours a week across the org. Forecast accuracy improvements would compound quarterly.
  • The workflow is suggest-only with a human in the loop. Reps approve every write to Salesforce. No silent overwrites of pipeline data — the highest-blast-radius failure mode in a system like this — is structurally impossible in v1.
  • The AI job is sharp. Five named fields, schema-typed outputs, evidence requirement (every extraction must cite a transcript span). When the model has nothing to cite, the field stays blank instead of being guessed.
  • Cost is small for a sales tool. Per-call extraction comes in well under $0.10. At the pilot scale, the unit economics are not the question.

What's not yet ready

The model behaves well in the demo room. The pilot envelope around it has gaps you don't notice until you turn it on for 12 real reps with real customers.

  • Consent flagging is incomplete. The pilot will process recorded calls; the recording consent flag in Gong is reliable but the Chorus pipeline has 18% missing consent flags during the audit window. Processing a call without a verified consent flag is the kind of unforced error that becomes a legal write-up. Either restrict to Gong-native sources for v1 or do not start.
  • Calibration is unproven. The model returns a confidence score for each extraction, but the calibration of that score against true accuracy has not been measured. A 0.6 in this model may not mean what a 0.6 means in the next prompt revision. Without a calibration study, the "low confidence → suppress" rule is theatre.
  • Observability misses the only metric that matters. Acceptance is logged. Edits are not. A rep silently editing every extraction is materially different from accepting them as-is — that's how you find out the model is wrong in a way the accept rate hides.
  • Training is undefined. A rep who doesn't understand the evidence requirement will treat the proposals as facts. The pilot needs a 20-minute training session and a one-page job aid before day one.

The verdict, restated

Pilot, scoped, after blockers close. Not org-wide. Not in four weeks. The product is good enough that a 12-rep, 6-week controlled pilot is the right learning vehicle. The four blockers above are small but non-negotiable. Skipping them turns the pilot into a pilot of "did we get sued" instead of "did we save time."

The CRO's framing — "all 80 reps next month" — is a reasonable instinct from someone whose job is moving fast. The PM job is to translate that into the smallest experiment that actually decides whether to roll out. That experiment is 12 reps, six weeks, three success metrics, and one consent constraint.

What would change the verdict

A green light to scale beyond the 12-rep pilot requires all five:

  1. Consent flag coverage is verified at 95%+ across the data sources in scope, OR the pilot is restricted to Gong-native sources only.
  2. A 50-call calibration study shows the confidence score correlates with extraction accuracy at r > 0.6.
  3. The dashboard tracks accept_no_edit, accept_with_edit, and reject separately, with a weekly review.
  4. The 12-rep pilot completes 6 weeks with: edit rate < 25% on accepted extractions, zero confirmed bad writes to Salesforce, net-positive rep sentiment.
  5. Sales operations signs off on the rollout plan, the training plan, and the rollback procedure.

If any of the five misses, the pilot extends. The decision to scale doesn't move just because the calendar does.

Link copied