Posted on
Every few months a board asks why their “transformation” hasn’t delivered. Same slideware, same theatre: a wall of frameworks, some traffic lights, and a fresh promise that this OKR cycle will turn the ship. It won’t.
Not because your people are lazy, or the tech is bad, or the consultants are evil. It’s because you optimised the wrong thing: you optimised for ceremonies and artefacts, not decision quality and speed. ⚡
The uncomfortable bit
- You created an Agile operating model… and then buried it under quarter-by-quarter capex gates.
- You installed an AI platform… and then refused to change the KPIs that reward outputs over outcomes.
- You stood up a Portfolio Board… and then packed it with sponsors who can approve spend but can’t kill pet projects.
- You built a gorgeous Business Capability Model… and then treated it like wallpaper.
If that stings, good. Because the fix is not another tool or framework. It’s governance that produces fewer, faster, better decisions, and relentlessly de-risks the path from strategy to execution.
“But we’ve got the right frameworks…”
Frameworks are useful, until they become a crutch. TOGAF, SAFe, ITIL, OKRs, DevOps, FinOps… they’re all tools. The problem is when the tool becomes the goal. You start reporting maturity instead of value delivered. You reward adherence to process, not impact. You manage the map, not the territory.
Litmus test: if a smart cross-functional team can’t get a yes/no in under two weeks for something clearly aligned to strategy, you don’t have an Agile problem. You have a decision rights problem.
The three decisions that decide everything
- What will we stop doing?
Not “deprioritise”, stop. Shut it down. Archive the Confluence page. Reclaim the people. Until you can name the top three initiatives you killed this quarter and the capacity they freed, you are not running a portfolio, you’re running a museum. 🏛️ - What will we bet on, for how long, with what kill-switch?
Every strategic bet needs a timeboxed runway, explicit leading indicators, and a pre-agreed threshold where you pull the plug without drama. No post-hoc rationalisation. No “just one more sprint”. - Who owns the outcome, end-to-end?
Not ten people in a RACI. One accountable owner with the authority to trade scope, schedule and spend to hit the outcome. If they need five approvals to move one dollar, you’ve set them up to fail.
Capability first. Projects second.
Most orgs still run change as a conveyor belt of projects with brittle handovers to BAU. That’s why value leaks: you ship a thing, then the capability that should exploit the thing is under-funded, under-skilled, or structurally orphaned.
Flip it:
- Fund capabilities, not projects. Projects become timeboxed investments to lift a capability to a target level.
- Measure decision velocity and risk burn-down inside the capability, not just delivery throughput.
- Overlay tech on capability maps, so you can actually see duplication, technical debt, and where a single upgrade unlocks multiple outcomes.
Your data lake is a landfill (and other home truths)
- Data Lake ≠ Strategy. If the people who make the decisions don’t trust or use the data in weekly routines, you’ve built an expensive archive. Make one decision ritual (e.g., pricing, roster, inventory, triage) measurably better with data, then scale.
- AI won’t save bad governance. If an approval takes six weeks, an LLM that drafts the business case in six minutes just means you wait five weeks and six days with nicer prose.
- Platforms without product thinking are cost centres. If there’s no product owner with P&L-like accountability and a roadmap tied to capability outcomes, it’s a hobby.
Boring beats brilliant
The organisations that quietly outperform aren’t louder. They’re boringly consistent:
- Single source of truth for strategy → cascaded to value streams and capabilities.
- Quarterly portfolio reviews that actually cut, with capacity released the same week. ✂️
- Thin, repeatable decision patterns (buy vs build, vendor selection, integration, privacy) that teams can apply without a steering committee every time.
- Leaders who review outcomes weekly, not slides monthly. If it isn’t on a cadence, it isn’t a priority.
What to change this week
- Kill two initiatives. Reinvest the people into the top capability tied to this quarter’s outcomes. Announce it.
- Publish decision rights. For each critical decision type, state who decides, the SLA, and the minimum artefacts required (keep it to one page).
- Outcome clinic. For your highest-spend “programme”, rewrite the outcome in one sentence: “By 31 Dec, reduce X by Y% in Z segment at ≤$N per unit.” If you can’t, you’re not ready to spend.
- Stop measuring motion. Replace vanity metrics (story points, number of workshops, tool adoption) with two portfolio KPIs: decision lead time and risk burn-down to the outcome.
- Fund a capability, not a project. Shift one inflight project into a capability budget and give the owner authority to trade scope for outcome.
The heresy
You don’t need more Agile. Or more AI. Or more consultants. You need fewer, faster, better decisions that connect strategy to execution through capabilities, then the courage to stop the rest. That’s it.
Until then, you’ll keep getting exactly what you funded: shiny frameworks, impressive dashboards, and surprisingly little change. 🤷♂️
