01 · DIAGNOSTIC FRAMEWORK

Turn the symptom into a testable decision.

A lower win rate can reflect worse execution, a harder opportunity mix or a changed definition. The diagnosis must reconstruct comparable cohorts before it judges people or process. The sequence below is designed to preserve definitions, expose alternative explanations and lead to a decision that can be validated.

  1. Audit the metricConfirm what counts as an opportunity, a win and a loss; identify changes in CRM fields, stage rules and open-deal handling.
  2. Rebuild cohortsGroup opportunities by qualification date and allow enough time for each cohort to mature. Avoid mixing recently opened deals with completed cohorts.
  3. Control for mixMatch region, company size, source, product, use case, price band and competitive intensity.
  4. Locate the stage breakMeasure entry, exit, elapsed time and re-entry for every stage. Read notes and activities around the break.
  5. Test a sales motionPilot one change—qualification rule, follow-up sequence, proof asset or pricing approval—in a bounded segment with pre-set success criteria.
02 · EVIDENCE

Ask for the minimum data that can change the answer.

Begin with read-only access and a field-level purpose. Reconcile samples before scaling extraction, preserve event time and source provenance, and record missingness rather than silently filling it.

CRM historyOpportunity snapshots, stage transitions, amounts, owners, products, sources, close dates and explicit losses.
Activity trailMeetings, emails, response delays, demos, trials, proposals and stakeholder coverage.
Commercial contextList price, discount, procurement steps, competitor, contract terms and approval time.
Customer evidenceCall notes, objections, product usage during trial, security review and reason for no decision.
03 · PROOF OF CONCEPT

Validate the claim before changing the operation.

Test the changed sales motion on one decision point

Select one stage with a measurable break and one segment with enough volume. Define eligibility before the test, hold pricing and lead source stable where possible, and compare conversion, cycle time and deal quality. If traffic is low, use matched historical cases plus blind review of calls and proposals; do not declare success from anecdotes or pipeline value alone.

04 · FAILURE MODES

What makes the diagnosis look right and still fail.

  • Comparing immature cohortsRecent opportunities have had less time to close and make the newest period look worse.
  • Using rep ranking as diagnosisTerritory, segment and lead-source mix can dominate individual results.
  • Trusting loss reasons literallyDropdown reasons are often selected late and inconsistently; validate them against notes and calls.
  • Changing five variablesA new script, price, qualification rule and incentive at once makes the result uninterpretable.
  • Optimizing win rate aloneRejecting more opportunities can raise win rate while reducing profitable revenue.
05 · SOURCE TRAIL

Primary and official references

These sources define the measurement, control or operating context. They do not replace validation on the company’s own data.

  1. Salesforce, State of Sales
  2. U.S. SEC, Guide to Financial Statements
  3. NIST, AI Risk Management Framework
06 · FAQ

Questions enterprise teams ask first.

What denominator should win rate use?

Use an explicit qualified-opportunity cohort and state how open, withdrawn and no-decision deals are treated.

How do we know if price caused the decline?

Compare matched opportunities, approval time, discount depth, competitor presence and objection evidence; price correlation alone is insufficient.

Can AI score sales calls reliably?

It can assist structured review, but a labeled rubric, human audit and segment-level error analysis are required before the score influences decisions.

What should the PoC optimize?

Choose profitable won revenue or another business outcome, with guardrails for cycle time, discount and customer fit.