What problem is worth solving?
Teaching scenario: sales has public-style material for five fictional companies but no clear priority or reason to contact them now. Build a meeting judgment, not an automated email blast.
Reduce unfocused preparation and prioritize opportunities worth validating.
- WORKING ROLES
- Sales, domain experts, sales operations and FDE
How does an FDE turn symptoms into a solvable problem?
The following discovery and implementation exercise uses a synthetic scenario.
Inspect a rejected brief
Ask a role-playing salesperson why a brief fails: wrong entity, stale evidence, no business angle or news mistaken for buying intent.
Separate fit and timing
Industry fit supports a hypothesis. Hiring or expansion signals raise questions but do not prove budget or purchasing timing.
How does the solution enter the workflow?
- 01
Resolve
Check names, domains and subsidiary relationships.
- 02
Gather
Record dates, entity and source for every signal.
- 03
Frame
Link a signal to a testable business question and counterargument.
- 04
Prepare
Create a one-page brief, three questions and next validation; do not send automatically.
Which decisions make the solution reliable?
Unknown beats false precision
Explain scores and missing fields; leave buying timing unknown without evidence.
Calibrate with acceptance reasons
Use edits and acceptance reasons to fix entity or domain errors, not brief-open counts alone.
Turn expansion news into a useful question
Fictional Northstar announces a new service region without disclosing tools, budgets or procurement plans.
- Known
- Coverage has changed; the source is dated.
- Hypothesis
- Does cross-region service require consistent knowledge and response standards?
- Meeting question
- Which handoffs repeat, and who owns service quality?
What should you do and deliver?
Rank five fictional accounts
- OUTPUT
- Priorities + exclusions
- PASS CONDITION
- Fit, timing and confidence are separate
Design discovery questions
- OUTPUT
- One-page meeting brief
- PASS CONDITION
- Questions can disprove the initial thesis
Define briefing acceptance
- OUTPUT
- Quality rubric + time baseline
- PASS CONDITION
- Wrong entities and stale sources fail
How do you prove completion and reuse the work?
Acceptance checks
Review entity accuracy, freshness, specificity and acceptance reasons using fictional or authorized data.
Reusable assets
Reuse entity rules, signal taxonomy and briefing templates with ongoing domain review.
Further reading and original sources
- Building effective agents ↗Anthropic · 2024-12-19
- Demystifying evals for AI agents ↗Anthropic · 2026-01-09