OPEN PRACTICE • VERSIONED • REVIEWABLE

Community Sales Playbook Registry

Reusable sales practices should be inspectable, measurable and improvable. Each playbook states the problem, evidence needed, steps, success metrics, risks and human-review points so contributors can challenge and improve it.

Registry

Stable • VERSION 1.0.0

Responding to ‘We already have a CRM’

A discovery-first objection play that positions AI as a transparent intelligence and workflow layer rather than an automatic CRM replacement.

Author: Flame Connect Community Seed
Audience: Account executives and sales managers
License: Pending repository license decision

Problem this solves

Prospects may interpret AI sales tooling as duplicate software before the seller has diagnosed gaps in prioritization, follow-up, coaching or forecasting.

Evidence needed before use

  • Current CRM and adoption level
  • Where manual work remains
  • Forecast or follow-up pain
  • Decision criteria and stakeholders

Practice steps

  1. Acknowledge the existing CRM investment.
  2. Ask which parts of the sales workflow still depend on spreadsheets, memory or manual judgement.
  3. Confirm the business consequence of that gap.
  4. Position the proposed AI workflow as complementary unless replacement is explicitly justified.
  5. Offer one measurable pilot with a clear baseline and human approval points.

Measure whether it works

  • Discovery-to-next-step conversion
  • Pilot acceptance rate
  • Time saved in the chosen workflow
  • CRM activity completeness

Risks and failure modes

  • Overclaiming integration capability
  • Suggesting a replacement before understanding the current stack
  • Using vague ‘AI transformation’ language without a measurable job-to-be-done

Human-control points

  • Confirm all integration claims before sending them to a prospect.
  • Do not promise ROI or implementation timelines without validated evidence.
CONTRIBUTION STANDARD

Useful community knowledge must be testable.

A playbook contribution should define the sales problem, target user, prerequisites, steps, measurable outcomes, risks, human-review requirements and a version. Claims about performance need evidence; unsupported promises should not be merged.