Hoshin Kanri software: what it must actually do (2026 guide)

Strip away the marketing and four capabilities separate a hoshin tool from a drawing of a matrix:

• A real X-Matrix — four quadrants linked by correlation strength in all four corners, not a static table. Export and share for board meetings.
• Review cadences — a meeting hierarchy (strategic reviews down to stand-ups) with agendas, action items and decisions traced to matrix elements. This is where strategy deploys or dies.
• A completeness audit — automatic checks: is every quadrant covered? does every vision element chain to a KPI? are there orphans? is there a balance of leading and lagging indicators?
• Interoperability with execution — your initiatives' real progress lives in Linear, Jira or Asana. The tool must link to it (not duplicate it), so status updates do not depend on human re-typing.

Strategy-execution tools fail the same way regardless of vendor: they ask your team to re-type what already lives in the execution tools and the BI stack. Adoption survives the January workshop, decays by March, and by summer the tool is an expensive mirror of an outdated slide deck.

Evaluate any tool — including ours — on this single question: what keeps the data fresh between quarterly reviews? If the answer is "your team's discipline", you are buying the same failure again. If the answer is "an integration per tool", ask who maintains the long tail. The 2026 answer is different: the agent is the integration.

An agent-native hoshin tool exposes its matrix through the Model Context Protocol (MCP) — the open standard that lets assistants like Claude use external tools. Concretely, with our platform: you connect claude.ai, Claude Desktop or Cursor to the MCP server in about two minutes (OAuth 2.1), next to your Linear or Jira MCP.

Then one weekly prompt does the maintenance a strategy officer used to do: read the initiatives and their linked execution projects, pull progress and blockers, update statuses, flag drift (work tracked nowhere in the strategy, initiatives linked to nothing), and prepare the weekly review agenda. Humans open the matrix four times a year; agents visit it every week.

Most strategy-execution suites price per user — which quietly fights the method: hoshin works when the n-1 and n-2 levels see the matrix, and per-seat pricing makes every invitation a procurement question. Per-organization pricing (our model: free solo tier, €89/month for a full organization, €290/month for consultancies managing client organizations) aligns the incentive with deployment.

Also check the sovereignty question: where does your strategy — arguably your most confidential document — live? A local mode where the matrix stays in files on your machine, readable by your AI via MCP but never uploaded, is the strictest answer available; EU hosting with strict tenant isolation is the standard one.

Honest answer: for a first hoshin exercise in a team of five, Excel is fine — start with our free X-Matrix template. The spreadsheet breaks when reviews begin: version chaos, no correlation audit, no cadence management, no agent access. That threshold usually arrives with the second quarterly review — if yours keeps getting postponed, the problem is the tooling of the ritual, not the ritual.