Guide · 10 minute read

Why strategy execution fails

Almost every organisation has the same experience: a strategy that made sense on the day it was approved, and a year later a room full of capable people explaining why it hasn't happened. The usual answer — more commitment, another restructure, a bigger communication campaign — rarely changes anything. Execution doesn't fail randomly. It fails at seven predictable points, and each one has a specific fix. This guide shows you how to find which one is yours.

What strategy execution actually means

Strategy execution is the work of turning a chosen direction into results that are verifiable. It has three parts, and a plan usually breaks in the gaps between them:

• Translation — turning the strategy into a small number of concrete results, each with an owner and a date.
• Operation — running those results week to week, and clearing the obstacles that appear.
• Feedback — knowing whether the result is actually closer than it was, and adjusting while it still matters.

Note what isn't in that list: the quality of the document. Most strategies that fail were perfectly reasonable on paper. The failure is in the machinery between the paper and the world.

The seven break points

Read each one and notice which makes you nod. That instinct is your diagnostic — you already know more than your progress report says.

1. 1. The strategy was never translated into work

   The tell: People can repeat the vision word for word but cannot say what changes in their own week because of it.

   The fix: Rewrite each strategic goal as a result with a date, then name the two or three actions that move it and who does them. If a goal has no owner and no action attached, it is a wish, not a strategy.

   Example: a district development plan promised "improved livelihoods". Translated, it became "300 households in two sub-counties selling maize through a registered cooperative by July", with the DTDO owning it. Only then could anyone tell whether the week had moved.

2. 2. Ownership is shared, which means it is nobody's

   The tell: A committee is accountable for the result. Three people say "we're responsible" and nobody's calendar is shaped by it.

   The fix: Name one person per result in writing, with the authority that comes with it. Committees review and unblock; individuals deliver. Shared accountability feels safe and produces orphans.

   Example: an NGO's immunisation outreach sat with a "task team" for a year. Naming one programme officer as owner — and giving her the right to release fuel money — took coverage from a plateau to a climb in one quarter.

3. 3. There is no feedback loop on the result

   The tell: You can produce a beautiful activity report but cannot say, today, whether the result is closer or further away than it was last month.

   The fix: Pick one number that means the result, agree where it comes from, and look at it weekly. Activity data tells you you were busy; result data tells you whether you got anywhere.

   Example: a training programme reported 1,200 people trained and called it success. Once they tracked "graduates in verified paid work 90 days later", the number was 8%. The feedback loop didn't just measure the gap — it exposed it.

4. 4. Decisions are blocked below the level they should be made

   The tell: Work waits for approvals that take longer than the work itself, and the people closest to the problem cannot act.

   The fix: Set a decision threshold and delegate everything under it. Publish who can decide what, and give the owner a way to escalate fast when something genuinely needs a higher call.

   Example: a county roads programme stalled on variations under five million shillings, each needing a quarterly committee. One delegated authority later, works restarted within a fortnight on the same budget.

5. 5. Everything is a priority

   The tell: The plan lists fifteen priorities, all marked high. Every new request gets added to the same list, and nothing gets dropped.

   The fix: Name the one binding constraint and let it decide what waits. A priority list with no items marked "not now" is a wish list; the cost of a new commitment is an old one.

   Example: a manufacturer added a second product line while its first still had stockouts. Dropping the launch for one quarter to fix supply raised both lines' service level; running both had been sinking the better one.

6. 6. A capability gap is being read as a motivation problem

   The tell: The same task keeps slipping and the response is reminders, pep talks or reorganisation — not training, tools or hiring.

   The fix: Ask whether the person has the skill, the tool and the time. If any is missing, supply it before concluding they don't care. Effort cannot substitute for capability.

   Example: a health team's data quality kept failing audits. It wasn't laziness — two officers shared one laptop and the reporting tool needed a feature their phones couldn't run. Equipment, not exhortation, closed it.

7. 7. Incentives point somewhere else

   The tell: What the strategy asks for and what gets rewarded are different things — usually reporting upward beats delivering outward.

   The fix: Follow the reward: what is asked for in reviews, what gets promoted, what gets funded. Change one of those signals to match the result, and behaviour follows faster than any memo.

   Example: field officers were told to grow user numbers but were appraised on forms submitted. They submitted forms. Switching one appraisal question to verified active users changed field behaviour within a cycle.

Fix the binding one first, not the loudest one

Every organisation has all seven problems to some degree, which is why the tempting move — a programme to fix everything — collapses. Progress moves at the speed of the weakest part, so the break point with the lowest score decides the outcome no matter how much you improve the others. A team with a clear result, strong skills and good funding will still stall if nobody owns the decision that unlocks the work; pouring money and talent into the parts that were never the limit is exactly what a well-resourced failure looks like.

The discipline is: name the lowest one, fix it, and watch what replaces it. Clear a blocked decision and a missing feedback loop usually surfaces next. That is not whack-a-mole; it is the system telling you where its limit has moved to. The binding-constraint guide walks through that loop in four twenty-minute steps.

The weekly rhythm that closes the gap

Executing organisations don't have better willpower; they have a lighter, tighter rhythm. Four small habits carry most of it:

• One number, once a week. Fifteen minutes on the single result number. Not a status round of fourteen updates — one number and the question "why this move?"
• One obstacle removed, every week. Ask owners what is blocking them and delete one thing. Execution improves fastest by subtraction.
• Two moves at a time, maximum. Anything more and you can't tell what worked when the number finally shifts.
• Close on evidence, not effort. A move is done when the thing you said you'd see is there — a letter signed, a number changed, a behaviour observed.

A full worked example

A municipal water utility had a five-year strategy with a flagship objective: cut non-revenue water — the treated water that vanishes before it is billed — from 38% to 20%. Three years in, it had moved two points. Running the seven-point check produced a stark profile: translation 4, capability 4, incentives 3, ownership 3, priorities 2, feedback 2, decisions 1.

• The binding break point: decisions. A leak repair needed a procurement sign-off that averaged eleven weeks; crews stood idle waiting for a valve.
• The move: a standing threshold letting the regional engineer release repair parts up to a set value, plus a one-page weekly loss map so crews could see where water was going.
• Evidence of the move working: median repair-authorisation time under five days, and twelve consecutive weekly loss maps.
• Then the next break point appeared: with repairs moving, the missing feedback loop bit — meter data was monthly and three weeks late, so nobody could tell whether a repaired zone was holding. That became the following quarter's single fix.

Nothing about the strategy changed. The machinery between the strategy and the ground did, and the objective that had crawled for three years began to move.

What good looks like, and when

• Within two weeks: one named owner per result, one agreed number, one obstacle actually removed.
• Within six weeks: the weekly number is moving, or has stopped moving and you know exactly why — both are progress against silence.
• Within one quarter: the break point has shifted, and the conversation in reviews is about the result rather than about effort and intentions.

Common traps

• Launching an execution programme — a second plan whose job is to deliver the first plan.
• Reading a full activity report as evidence that the result is on track.
• Naming "lack of funds" before checking ownership, decision rights and feedback.
• Fixing the break point that is comfortable to work on instead of the one that is binding.
• Announcing a restructure and expecting behaviour to change without anything being delegated.

Questions people ask

What is strategy execution?

The work of turning a chosen direction into verifiable results: translating the strategy into a few concrete results with owners and dates, running them week to week while clearing obstacles, and knowing from evidence — not from optimism — whether the result is closer than it was.

What is the difference between strategy implementation and execution?

Implementation usually means standing the pieces up: projects, budgets, structures, systems. Execution means the pieces actually producing the result — the weekly running, deciding and adjusting. You can finish implementation on time and still not execute, which is why "everything delivered, nothing changed" happens.

Who owns execution?

One named person per result, never a committee. Leaders own two things the individuals cannot: the decision thresholds and the priorities list. If everyone owns execution, the honest answer is that nobody does.

Why do strategies fail even with good funding?

Because funding flows to the parts of the plan that were never the limit. Money and talent improve the visible, comfortable parts while one unowned result, one slow approval or one missing feedback loop quietly decides the outcome. Progress moves at the speed of the weakest part, so funding cannot outrun a binding constraint.

How long before strategy execution improves?

The machinery can change in a fortnight — an owner named, a threshold delegated, one number on the table each week. The result number usually takes one to two quarters to move visibly. If nothing has changed in a quarter, the most likely cause is that the break point named was not the binding one.