Content
67%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A thorough, well-structured retention playbook with highly concrete guidance and clear sequenced workflows supported by real reference files. Its main weakness is length — some prose and detail could be trimmed or moved into the existing references to improve token efficiency and progressive disclosure.
Suggestions
Trim motivational prose (e.g., 'The best save happens before the customer ever clicks Cancel.') and move the full dunning email sequence plus retry-timing tables into references/dunning-playbook.md to tighten the body.
Add explicit inline validation checkpoints to the cancel-flow and dunning workflows (e.g., 'Verify save rate after 2 weeks; if below 25%, iterate on offer copy').
Flesh out the currently high-level 'Personal outreach' and 'Tool Integrations' sections with concrete steps or example commands, or move them into a reference file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with actionable tables but runs ~430 lines and includes prose that could be trimmed (e.g., 'The best save happens before the customer ever clicks Cancel.'), so it is mostly efficient but could be tightened. | 3 / 5 |
Actionability | Concrete guidance dominates — offer-to-reason mappings with specific discount ranges (20-30% for 2-3 months), retry timing, and a dunning email sequence — but sections like personal outreach and the tool-integrations list remain high-level with minor gaps. | 4 / 5 |
Workflow Clarity | Multi-step processes are clearly sequenced (Cancel Flow Steps 1-5, Dunning Stack) with measurement/A-B-testing serving as validation, though inline validation checkpoints are implied rather than explicit. This is a design/strategy skill, not a destructive batch operation, so the destructive cap does not apply. | 4 / 5 |
Progressive Disclosure | Two real, well-signaled one-level-deep references (cancel-flow-patterns.md, dunning-playbook.md) offload provider/industry specifics, but the body still retains substantial detail (UI mockups, full dunning sequence, health-score model) that could arguably live in the reference files. | 4 / 5 |
Total | 15 / 20 Passed |