Start with one operating situation
I want a playbook to help a seller make a better decision on a real account. If it only works when the founder explains it, it is not ready to hand over.
The fictional product in this guide manages entitlements for B2B platforms. It is not a Ramp Year customer or a disguised account result. The play applies when a platform introduces differentiated enterprise access and the team may be managing changes manually. Public product information makes the situation worth investigating; it does not prove pain.
The play's objective is to determine whether a relevant buyer has a consequential access-management problem, whether the standard product can support it and whether a change decision exists. The objective is not to book every plausible account. A complete play includes the reason to stop.
Before using it, replace the product boundaries, roles and economics with your own. Have the product owner validate capability statements. Have the commercial owner define acceptance and authority. Review sender and jurisdiction-specific requirements before any external contact; this example does not authorize a campaign.
Save the account fact, source and date.
Let the buyer confirm, correct or reject the problem.
Progress, revise or stop with a reason and owner.
The account entry rule
Select a platform with a public enterprise tier or differentiated product access, a plausible engineering or product owner and a feasible standard integration. Save the URL and date for each observation. Reject known incompatible deployment requirements or a product model with no meaningful access complexity.
Record three separate statements: the observed fact, the possible consequence and an alternative explanation. For example, enterprise tiers are visible; manual administration may be costly; alternatively, the company may already have a reliable internal service. A researcher should never turn the second statement into a fact because the record feels incomplete.
| Decision | Evidence to use | What changes next |
|---|---|---|
| Entry observation | Public enterprise access or plan differentiation; source and date saved | Research current administration without claiming pain |
| Hard boundary | Unsupported deployment or no relevant access problem | Reject or request a bounded product review |
| Unknown | Current process, impact and decision owner not established | Prepare neutral discovery questions |
A message you can adapt
An illustrative opening: 'I saw your enterprise plan separates access by customer. How do plan changes reach the product today? We help teams address the workflow around this kind of access administration. If you already have it running reliably, that is useful to know.' The service statement must match your actual product; do not copy the entitlement claim if you sell something else.
The message has a source-backed observation, a relevant question and permission to correct the premise. It avoids a fabricated diagnosis or a forced demo request. Keep accurate sender identity and appropriate opt-out handling. Do not add unrelated personal facts merely to make the email appear customized.
If the contact asks a technical question, answer within approved boundaries and involve an expert where needed. If they refer another role, preserve context and approach responsibly. If they request removal, suppression controls the workflow. A mixed referral and opt-out is not simply a positive reply.
Discovery that can disprove the premise
Ask for a recent example of a plan or access change. Follow the workflow: who requested it, what happened, who handled it, how often it occurs and what consequence followed. Only then discuss the proposed alternative. Leading with an ROI number can make the buyer agree to a story you have not actually understood.
Use these five questions: How are changes handled today? Which exceptions are difficult? What effect does that have on the team or customer? Who owns a decision to change the process? What would a useful evaluation need to establish? Record answers in the buyer's terms and label remaining unknowns.
If the process is solved or the consequence is negligible, stop the pain narrative. If a problem exists but no active project does, retain that state honestly. An informative call can be useful without becoming qualified pipeline.
Buyer: 'We already built this internally.'
Ask whether the current system meets the relevant requirements and whether any unresolved issue exists. If it works well, record the rejected hypothesis and stop. Do not manufacture a new problem to preserve the meeting.
Buyer: 'It is painful, but not a priority.'
Ask what would make a change relevant and whether a review condition is useful. Confirm the pain without inventing urgency. Keep it separate from an active buying decision.
Buyer: 'Can you support our deployment model?'
Check the approved capability boundary. If unsupported, say so and involve the product owner only where a real alternative exists. Do not promise a roadmap date or custom implementation without authority.
An honest qualification and business-case record
An accepted opportunity needs the agreed minimum evidence and a receiving commercial owner. In this play, use operational fit, buyer-confirmed problem, feasible adoption and a meaningful mutual next action. Budget or approval details can remain unknown when clearly labeled with a validation action.
Illustrative note: 'Engineering manager confirmed 40 access exceptions per month. Estimated handling effort is 15 minutes each; inputs require validation. Product leadership will join a scoped evaluation discussion. Finance approval role unknown. Standard integration appears feasible, pending technical review. AE accepts the next action.'
The arithmetic is 40 × 15 minutes = 10 hours per month. An assumed 50% reduction would free five hours, but not automatically create five hours of payroll cash savings. Add integration, maintenance and training effort. Let the buyer validate the inputs and decide whether the consequence justifies change. Do not write a guaranteed ROI conclusion.
| Decision | Evidence to use | What changes next |
|---|---|---|
| Confirmed | Buyer described exceptions and their workflow | Use accurate notes with date and appropriate source |
| Assumed | Frequency, effort or reduction still needs validation | Show input owner and confidence rather than false precision |
| Unresolved | Budget, scope or approval role is missing | Assign a next question and responsible seller |
Evaluation and the committee's next decision
Agree what the evaluation must demonstrate: one standard integration path, relevant access behavior, operational ownership and implementation constraints. Ask which roles review technical feasibility, product impact, security and spend. A champion can support the purchase without controlling every approval.
Write a mutual plan with action, purpose, owner, dependency and timing the buyer confirms. If security needs an approved document first, assign the internal expert and revise the sequence. Do not list 'sign Friday' while the relevant decisions remain open.
A sales room can hold the summary, scope, success criteria, buyer-validated assumptions and approved materials. Keep sensitive deal records private and access-appropriate. A public educational article should be indexed; a private committee room should not become an accidental search result.
Handoff, agent support and recovery
The SDR-to-AE record contains account rationale, source context, buyer text, meeting purpose, unknowns and the next owner. The AE accepts, returns or rejects with a reason. If the meeting moves, commercial ownership does not disappear with the old calendar invitation.
Agents may collect sources, prepare drafts, summarize with uncertainty and suggest routing. Humans retain consequential qualification, product commitments and exceptional terms. Test an ambiguous reply and a stale source, not only a clean happy path. Suppression must override a model's positive sentiment label.
For a routing failure, retain the response, alert an owner, verify account matching and replay the update without duplicate outreach. Document the recovery path. If only the original builder can repair it, the workflow has not been handed over.
| Decision | Evidence to use | What changes next |
|---|---|---|
| Normal handoff | Context, unknowns and receiving-owner acceptance | AE continues rather than restarts the conversation |
| Mixed reply | Referral plus removal request | Suppress the original contact and review any separate next action |
| Failed synchronization | Response retained; owner alerted; duplicate check | Recover with a documented, controlled update |
Prove the play survives handover
Name internal successors while the play is being built. Let them observe decisions, make reviewed attempts and then demonstrate on unfamiliar cases. A researcher should identify fit and disqualification; a seller should choose a responsible next buyer action; an administrator should recover a controlled failure.
The acceptance artifact is their work and reasoning, not attendance or a folder receipt. If a test fails, repair the explanation, product context or system ownership and coach the missing skill. Remove agency access only after continuity has been established.
Assign a maintainer and review triggers: repeated rejection, product boundary change, integration failure or a new buying pattern. Keep version and rationale, separate current rules from historical material and remove duplicates. The playbook stays useful when it changes the team's decisions, not when it becomes longer.
Write the exception, not just the happy path
A playbook becomes useful when it tells a seller what to do when the example stops fitting. Put the ordinary decision, its evidence and the escalation boundary together. The aim is consistent judgment, not identical sentences.
Rehearse an unfamiliar account and a difficult discovery answer. If the seller can quote the playbook but cannot explain why the next action fits the evidence, improve the decision rule before adding more pages.
| Part | Example instruction | Evidence of transfer |
|---|---|---|
| Trigger | Investigate a new approval requirement. | The seller can distinguish a signal from confirmed pain. |
| Question | Ask how the most recent handoff worked. | The seller follows the answer rather than a script. |
| Boundary | Escalate unsupported security or product commitments. | No promise is made outside agreed scope. |
| Stop rule | Stop when the buyer has no relevant problem or priority. | The seller can explain a disqualification. |
| Review | Compare the decision with its source evidence. | The coach can identify the specific judgment to improve. |
Connect this decision to the complete startup sales-motion plan.
Sources and research notes
- GitLabsales operating procedures
Company operating handbook
- GitLabcustomer-ready shadow program
Company operating handbook
- GitLabsales development handbook
Company operating handbook
- GoogleEmail sender guidelines
Mailbox-provider requirements
- Primary sourceFTC: CAN-SPAM business guide
Regulator guidance; US scope
- AttioBack up workspace data
Product documentation
Primary sources reviewed October 6, 2026. The operating recommendations and worked scenarios are Daavid’s analysis. Illustrative numbers are assumptions, not measured client results. Company marks identify sources and prior experience; they do not imply a customer relationship or endorsement.

