Purpose, objectives, projects and indicators, linked on one page. Your assistant keeps it current with you; you decide.
Free for one matrix. The demo opens without an account.
Example shown: Hoshimeca, a fictional 120-person manufacturer. The structure is the same for any trade.
You. Write me a progress summary of the strategy for the committee. Claude. Summary ready, objective by objective. One point to watch: production lead time. The project meant to cut it is behind: changeover takes 41 minutes against a target of 25. Shall I share the page with the committee as a read-only link?
New to Hoshin Kanri? It is a management method for deploying a strategy: leadership chooses a few changes for the year, each level answers with its own objectives and means, and regular reviews check that the work still serves the direction. The X-Matrix puts it on one page. Hoshin Kanri explained, with one example
Your organization already has its AI agent. What it lacks is your strategy.
It was decided at an offsite, then filed in a slide deck. Three months later everyone follows their own priorities, projects no longer tie back to anything, and meetings go to status instead of decisions. Your AI assistant has never read it: it helps you with everything except what matters most.
The purpose of the company, the vision for the cycle, where you start from and what you give up. A short text, which you validate. Your assistant reads it before anything else.
Three-to-five-year objectives, this year’s objectives, projects and indicators, linked on one page. A project linked to nothing shows, and so does an objective whose projects have no indicator. A score says what is missing.
Reviews on fixed dates, prepared by your assistant: readings up to date, points to decide. Each decision is kept with the state of the strategy when the review is recorded.
A name, an email address, the name of the company. You confirm by a link sent to your inbox. Free for one matrix.
In Claude, add the address below as a connector, then sign in. ChatGPT, Cursor and others document the same kind of connection; we have not tried them. Conditions per agent on the dedicated page.
https://hoshin.app/mcpEvery agent and its conditions
Your workspace gives it to you when you first open it. The assistant interviews you, starting with the why, and writes the strategy into your workspace. The written why stays a draft until you validate it in the app; every line can be corrected or deleted.
The assistant prepares and keeps things current. The decisions stay yours.
Hoshimeca is a fictional 120-person machining company: a manufacturing example, and the structure is the same for any trade. Click an element to light up what it serves, then ask the assistant:
Get its completeness score without an account. We keep no copy of your file.
Three fictional companies, each with its demo. The exchanges below come from them.
Quality, lead times, plants: every shop-floor deviation ties back to the objective it threatens.
Recurring revenue, churn, onboarding, releases: what the product has to prove, on the same page as the strategy.
YouGoing live still takes 26 days. What does it touch?
AssistantThe 15-day objective and, behind it, churn: your customers leave mostly in their first six months. The guided onboarding project is halfway; importing existing customers is still manual. Shall I note it for the next meeting?
Earthworks, building, utilities: safety, job preparation, equipment and hiring tied to the partners’ objectives.
YouThree jobs started without a complete file this month. What does it touch?
AssistantThe standard preparation project, which carries the objective of 80% of jobs delivered on the contract date (66% today) and weighs on safety. You decided no job starts without its file: making an exception or holding the rule is yours to decide. Shall I note it for the next meeting?
One matrix per site, linked to the group’s: what each level takes up, and what nobody takes up.
One workspace per client, reviews prepared by the assistant, a method that outlives the engagement.
Hoshin Kanri is the method lean practitioners use to deploy strategy. Its single page is called the X-Matrix. You do not need to know it to start: your assistant asks the questions in order. If you already practice it, everything is here: correlations, a completeness score, cascade from group to site, reviews at five levels.
3 to 5 breakthrough objectives for the next 3-5 years — the North of the X-Matrix. Few enough to mean something.
What must be true this year for the vision to stay credible (East). Negotiated with the teams — the catchball — not decreed.
Improvement initiatives (South) and the indicators that will prove they work (West) — an owner on every row.
A 5-level meeting hierarchy, from strategic review to weekly sync. The review is where strategy deploys or dies.
Decisions traced, initiatives re-arbitrated, the matrix kept alive — hansei, the reflection that closes the loop.
Hoshin does not plug into your systems. Each agent keeps its access and its rhythm; they all read the same X-Matrix, and what they file is for people to decide.
Your systems, your rights
Prepares reviews and summaries
Ties the order book to the objectives
Watches indicators on its own schedule
Follows the site matrix, under the group’s
Objectives, initiatives, indicators and the links between them, read through MCP. Score, chains and cascade are computed, not generated.
No line from here to your systems: the agents make that link.
Proposals, in the inbox
The agent then acts in its own tools, if you allow it.
A steering tool has 5-10 active users — pricing by seat punishes adoption. Every paid tier has a 30-day free trial, no credit card. No sales call, no paid onboarding — your agent sets you up.
Early access: everything is unlocked during the beta — and early adopters keep an advantage when billing starts.
The X-Matrix is the central tool of Hoshin Kanri (policy deployment), a strategic planning method from Japanese quality management, since taken up by lean practitioners. It aligns four quadrants on one page: 3-5 year strategic objectives (North), annual objectives (East), improvement initiatives (South) and KPIs (West), with correlation marks showing how each level supports the next. It makes the chain from vision to daily work visible — and auditable.
Yes, with a frame. On its own it forgets between sessions. Connected through MCP, it reads the objectives, links and KPIs, prepares reviews and flags gaps. People decide. Tried by us: Claude. Other agents that speak MCP document the same connection.
For one person, honestly: yes, at first — the way Excel is fine for a first cycle. Our free local mode is exactly that, done well: structured JSON on your machine, the same MCP tools, completeness score included. DIY breaks when other people join: a file has no identities — no per-user rights for your CFO, your consultant or an intern's agent, no validation that objectives actually chain down to KPIs, no review cadences carrying decisions and action items. A skill tells an agent how to work; it cannot be the shared, permissioned state a leadership team works on. That is the part you pay for.
MCP (Model Context Protocol) is the open standard that lets AI assistants like Claude use external tools. Hoshin Kanri ships an MCP server: an agent can read your matrix, capture ideas mid-conversation, update initiative statuses, diagnose how well a strategic theme is deployed, and run a weekly sync against your execution tool. You connect it to claude.ai with OAuth in about two minutes.
No. The agent does, through the system's own MCP server or a bridge your IT team builds, under the rights that system grants. Hoshin holds no connector and no access to your systems. An agent that reads both can tell you which objective a drifting indicator serves.
It can connect with an API token, on its own schedule. The schedule runs on your side: Hoshin runs no loop for you. We have not run a 24-hour test yet. Two limits today: filing an idea requires a write token, which can also change the matrix and publish it on a public page, and nothing removes duplicates. The manufacturing page explains how to work with them.
No, by design. Your systems keep the work orders, non-conformities, tasks and tickets. Hoshin Kanri keeps the strategy: objectives, initiatives, KPIs and reviews. An initiative can carry a link (external_ref) to its project, and the weekly_sync agent prompt keeps the two aligned without double data entry.
In the hosted version, your matrices are stored in France (OVH), every query is scoped to your organization, and everything is encrypted in transit. Hoshin Kanri calls no AI model on your matrices: your strategy reaches a model only when your own agent reads it, under your contract with that model's provider. We use your data for nothing but the service: no training, no analysis, no resale. Like any host, we keep technical access to the database; our admin console shows the names of your matrices, not their content. If your business requires that nobody but you can read it, use the local mode: the matrix lives in files on your machine and never passes through our servers.
We build what verified organizations ask for, in the order they ask. One click records the request; we only come back to you if we start building it.
The assistant prepares and keeps things current. The page tells it why, what comes first and how success is measured. People decide. Free for one matrix.
Start with my assistant