Brain Readiness Score
Purpose: This document explains Summit's Brain Readiness Score of 87/100: what the score measures, where the points came from, where the missing points went, and why the score is meant to move over time. Dave owns this number; it gets re-scored at each quarterly review.
The score: 87/100
87 means the Brain is ready for daily use. The core of the business is documented well enough that the Hermes agent can answer most internal questions and draft most routine communications from the vault alone. It does not mean the Brain is finished. The remaining 13 points map directly to the six gaps in the Knowledge Gaps Report, and closing them is the point of the ongoing optimization work in the Implementation Roadmap.
What the score measures
Five dimensions, weighted toward the ones that make an agent safe and useful on day one.
| Dimension | What it asks | Weight |
|---|---|---|
| Coverage | Is every part of the business documented: offers, sales, ops, marketing, team, tools? | 30 |
| Consistency | Do documents agree with each other on prices, thresholds, names, and process steps? | 20 |
| Actionability | Are rules written as decisions a person or agent can actually execute, not vague intentions? | 20 |
| Agent-readiness | Can Hermes ground answers in these docs, with clear permission boundaries? | 20 |
| Freshness | Is there an owner and a review date on every document, and a cadence to keep them true? | 10 |
Where Summit scored well
- Coverage (27/30). All four service lines are documented with real numbers: the $12,000–$38,000 roof replacement range, the $650 repair minimum, the $450 emergency dispatch fee, the 30% remodel deposit. The Offer Library, Lead Handling Process, the SOP core in Operations, the Brand Voice Guide, Team Structure, and Tool Stack Map are complete.
- Consistency (18/20). The anchors hold everywhere: 15-minute speed-to-lead in business hours, estimates within 3 business days of the site visit, discounts over 5% needing Tara or Dave, and the GHL pipeline stages (New Lead through Won or Lost) named identically in every document that touches them.
- Actionability (17/20). Most rules are written as decisions. "Only Tara or Dave approves discounts over 5%" is executable. "Ranges only until a site visit or photo review" is executable. The points lost here trace to escalation calls that still live in Dave's head.
- Agent-readiness (18/20). The Approved AI Use Cases, Human Approval Rules, and Prohibited Agent Actions give Hermes hard boundaries: never send external messages, change CRM stages, publish, or quote firm prices without a human. The agent config in folder 09 is grounded in the same vault it answers from.
- Freshness (7/10). Every document has an owner and a last-reviewed date from the Q2 2026 engagement. What is not yet proven is the habit: the quarterly re-review cadence has not run its first full cycle.
Where the 13 points went
Each missing point maps to a known gap, detailed one by one in the Knowledge Gaps Report:
- Profitability by service line is not documented (Coverage).
- Field photo standards are drafted but await Pete's final approval (Coverage, Freshness).
- Some escalation rules still resolve to "ask Dave" instead of a written rule (Actionability).
- CRM stage definitions exist on paper but are not tightly enforced in GHL itself (Consistency, Agent-readiness).
- Warranty handling differs by service line and is not unified in one document (Consistency).
- Marketing attribution is noisy because agency UTM discipline is inconsistent (Coverage, Freshness).
None of these block daily use. All of them limit what the Brain and the agent can do next: an agent cannot prioritize leads by margin when margin by service line is undocumented, and it cannot summarize pipeline health accurately when stages are applied loosely.
Why the score is designed to move
A Company Brain decays if nobody maintains it. Prices change, people change roles, tools get replaced. The score exists so that decay is visible and improvement is measurable:
- The six gaps have owners and closure criteria. Closing all six is worth roughly the missing 13 points.
- Freshness only earns full marks once the quarterly review cadence has actually run, not just been scheduled.
- New gaps will appear as Summit grows. That is normal. The score going up and down over time is the system working, not failing.
Target for the next scoring pass: 93+ after the First 30 Days plan and the month-one items in the Implementation Roadmap are done.
Related
Want documentation like this for your business?
Every Company Brain engagement produces documents at this level of detail — captured from your team, organized, and wired into an agent.