The Twin
The Strategy Twin
A living model of your strategic world, from six domains to governed actions.
The Twin · Evidence and trust
What makes AI strategy board-grade
Why would a board trust strategy made with AI? Because every number links to evidence with a confidence grade, every plan is red-teamed, and licensed Domain Fellows sign the parts that carry liability.
In short
What makes AI-built strategy evidence-based?
Evidence-based strategy means every conclusion can be traced to its sources. strategy1 makes AI-built strategy board-grade with five controls: an evidence ledger that links every number to a source and date, a confidence grade on every key figure, a red-team report on every plan, sign-off by licensed Domain Fellows, and client-isolated Strategy Twins built for India's DPDP Act.
Key takeaways
- Every number and claim links to a source, a date and the agent that produced it, in an evidence ledger whose entries are chained.
- Every key figure carries a confidence grade, and a conclusion that rests on a model estimate is labelled as an estimate.
- Every plan ships with a red-team report: its strongest counter-arguments and the conditions under which it fails.
- Licensed Domain Fellows review and sign the parts of the work that carry professional liability.
- Nothing is written to a client's systems without a policy check, the owner's approval of a dry-run and a hash-linked audit record.
Defined in the glossary: Evidence ledgerConfidence gradeRed-team reportDomain FellowsGoverned writebackAll strategy terms
Five pillars
Evidence ledger, confidence grades, red team, expert sign-off, data protection.
Pillar 01 · Evidence ledger
Each number and claim links to a source, a date and the agent that produced it.
You can click from any slide to the evidence. The ledger is sealed in a hash chain, so an entry cannot be changed without the change showing.
Source · Date · Grade · Produced by
Slide figure · sample
Growth in adjacent segments, a year
Click the figure for its source.
Ledger entry
- Source
- Industry survey (fictional)
- Date
- Sep 2026
- Grade
- Credible secondary source
- Produced by
- Research agent
Fig 3.3A figure on a slide linked to its ledger entry, which records the source, date, confidence grade and producing agent. Entries are chained, so a change shows.
From decision to action
Nothing is written to your systems without approval.
Approved decisions become CRM opportunities, budgets, briefs and run-sheets — after your named owner accepts them. Before any write, the change passes a policy check, your owner approves a dry-run of exactly what will change, and the approval is recorded in a hash-linked audit trail. Outcomes then flow back into the Twin.
Signed
1Decision memo
Approved at gate H2.
Check
2Policy check
The change is tested against your policies.
Gate H3
Owner approves dry-run
Exactly what will change, before it changes.
dry-run ›
+ create opportunity · Segment B pilot accountRecord
4Audit record
Hash-linked to the approval.
Write
5CRM pipeline
Written only after approval.
Outcomes then flow back into the Twin.
Targets · choose one to replay the path
Dry-run lines are a sample, shown to illustrate the format.
AI-use disclosure
We say what AI did and who approved it.
Every deliverable names the model and the people who approved the work.
Specimen · sample disclosure
Prepared by strategy1 with AI assistance (Claude models). Reviewed and approved by [named Strategy Partner]; sections marked ✓ signed by [Domain Fellow]. Every figure links to its source and confidence grade.
The engagement letter carries the same disclosure before work starts.
FAQ
Questions about evidence and trust
How does strategy1 check AI-generated numbers?
Every number comes from a source recorded in the evidence ledger or from strategy1's own deterministic calculation code. The language model explains numbers but does not calculate them, each key figure carries a confidence grade, and the red team tests the conclusions that rest on them.
Who signs financial, technical and legal conclusions?
Licensed Domain Fellows. strategy1 provides decision support and blueprints; licensed professionals sign technical, financial and legal conclusions. Sections a Domain Fellow has signed are marked in the deliverable with their name.
Can a board see how any number in a strategy was made?
Yes. Every figure in a strategy1 deliverable links to its evidence ledger entry, which shows its source, date, confidence grade and the agent that produced it. In a Bootcamp, the client sees the ledger, the grades and the red-team report on their own decision.
What does the AI-use disclosure on a deliverable say?
It says what AI did and who approved the work. Every strategy1 deliverable names the model used and the people who approved it, including the Strategy Partner and any Domain Fellow who signed sections. The engagement letter carries the same disclosure before work starts.
How is client data protected in a Strategy Twin?
Each client's Strategy Twin is isolated from every other client's, with role-based access, and the controls are built for India's DPDP Act. Security, privacy and AI-use policies are published in the strategy1 Trust centre.
Keep reading
Related
Method
The S1 Method
Sense, Model, Simulate, Decide, Deploy, Learn. Six stages, three human gates.
Agents
Agent Mesh
The agents inside every engine, each with a guardrail, and a red team that argues against every plan.
Trust centre
Trust and privacy policies
Security, privacy, DPDP and how we use AI, in one place.
Proof
Compression Index
How we measure compression: baseline time against our time, method published.
Glossary
Strategy glossary
Plain definitions of the Strategy Twin, the S1 Method, evidence ledgers, red-team reports and more.
Ask us how any number was made.
In a Bootcamp you will see the evidence ledger, the grades and the red-team report on your own decision.