Your Strategy Didn’t Fail. You Had Fifty-Six Of Them.

Most plans I read are not wrong. They are too long to survive a normal week.

You know the day. Everyone off site, phones down, butcher’s paper on the wall, the good coffee. By four in the afternoon there are fourteen priorities on the board and the room feels genuinely great, because for one day you were working on the business instead of inside it. Somebody types it up. It goes in a folder.

Nine months later, three of those things are done. Four are half-started. The rest are dead and nobody has said so out loud, which is its own particular kind of drag on a team. And the conclusion the owner usually reaches is that the business has an execution problem, or a people problem, or that everyone just needs to lift.

I think it is almost always a length problem.

The Numbers Are Uncomfortably Clear

There is research tracking thousands of strategic plans from 2017 through to 2024, and the headline is grim. Across that period, 84.5 per cent of strategic projects never reached completion. Only 5.7 per cent of organisations finished three quarters or more of what they said they would do.

The interesting part is what predicts which side of that line you land on. Plans with fewer than 20 moving parts produced high performers 68 per cent of the time. Plans with 40 to 60 elements completed at 15 per cent. Plans with 60 or more completed at 8 per cent.

Same people. Same market. Same intent. The variable that moved the result most was how much was on the list.

The part that stopped me was what happened next. Over those same years the average plan nearly doubled in size, from about 33 milestones to 56. Businesses responded to a decade of not finishing things by writing longer lists.

That data comes from larger organisations with dedicated planning people. In an SME it is worse, because there is nobody whose job is the plan. There is you, and there are eleven people who all have a day job.

Why The List Always Gets Long

Nobody puts fourteen priorities on a board by accident. It happens for reasons that feel responsible in the moment.

Every single item is a good idea. That is what makes cutting so hard. You are never choosing between a good idea and a bad one, you are choosing between six good ones, and the only honest reason to drop five is capacity. Capacity is a boring argument to make at four in the afternoon when everyone is excited.

A long list also protects you from choosing. If fourteen things are priorities then you have not backed anything, and if the year goes sideways you can point at the list and say you tried on all fronts. A three-item plan is exposed. If those three do not work you have to look at your own judgement, and that is precisely why a three-item plan is worth more.

Then there is the room itself. Everyone brought something. Leaving someone’s idea off the board feels like telling them their area does not matter, so it goes on, and now the plan is a diplomatic document rather than a plan of work.

What you end up with is a record of good thinking. Which is a genuinely useful artefact, just not the same thing as a decision about what the business will do next.

The Plan Is Competing For People Who Are Already Full

Look at any list of priorities in a small business and follow the names. The new CRM, the onboarding rebuild, the second location, the quoting system, the hiring push. Four of the five land on the same operations manager, who is also running operations.

The plan was written as if a second business existed to deliver it. It does not. The people who will execute the strategy are the same people currently delivering the work that pays for everything, and their week is already full. Nothing on that board gets done in found time, because there is no found time.

So the strategy loses. Every week, quietly, to the customer on the phone and the job that has to go out on Thursday. Which is the correct call in the moment and a terrible one repeated fifty times.

This is why I ask for capacity before I ask for ambition. Not what do you want to achieve this year. Who is doing it, in what hours, and what are they no longer doing to make room. If a priority cannot survive that question, it was never a priority. It was a wish with a deadline attached.

Agreement Is Not Ownership

Everyone nodding in the room is the most reliably misleading signal in business. Agreement costs nothing. It is the cheapest thing a person can offer at 3.45pm on a planning day.

A priority owned by the leadership team is owned by nobody. A priority owned by two people is owned by neither, because both will assume the other is closer to it. One name. One date. One written line describing what done looks like, agreed in advance, so that finishing is a fact rather than an opinion.

And that owner needs the authority to actually move. Half the stalled initiatives I see are stalled because the person responsible is waiting on a decision only the owner can make, and the owner has been in back-to-back client work for a fortnight. That is the same bottleneck that shows up everywhere else in the business, wearing a strategy hat. Delegating the task and keeping the decision is not delegation. It is a queue.

The Meeting Where Nothing Gets Decided

Most businesses do have a rhythm around the plan. It is usually a monthly meeting where each person reports what they have been doing, everyone listens politely, and the meeting ends. No decision is made. Nothing is dropped. The list stays fourteen long and everybody leaves feeling slightly worse.

Twenty minutes a week, with three questions, does more than an hour a month of reporting.

What moved since last week. What is stuck, and specifically what would unstick it. What do we drop, because something has to give this week and I would rather choose than discover.

A monthly cadence means a priority can drift for four weeks before anyone notices, and by then the excuse has hardened into a story about why it was never going to work. A week is short enough that being stuck is still an event rather than a condition.

How To Cut It Back

The exercise is subtraction, and it is uncomfortable. Do it anyway.

• Three active priorities. That is the whole list. Not three per person. Three for the business. If that feels thin, remember the businesses finishing 68 per cent of their plans are the ones running short lists.
• Park the rest where people can see them. A visible later list is very different from a deleted idea. It tells the person who raised it that their thinking survived, just not this quarter, and it stops the same six items being re-argued at every meeting.
• Name a person, a date and a definition of done. Written down. “Improve onboarding” is not a priority. “Every new client gets the same five-step start, documented, running by 31 October, owned by Sarah” is.
• Say what stops. Every new priority displaces something. Name it. If nothing is being displaced, you have not added a priority, you have added overtime, and that is a bill that comes due in people leaving or an owner burning out.
• Sequence instead of running everything at once. Three things finished by March, three more by June, beats twelve things at 40 per cent in December. Finished work compounds. Half-finished work just sits there costing attention.
• Cut formally, once a quarter. Killing an initiative should be a decision someone makes out loud, not something that happens through neglect. Teams can handle “we are stopping this”. What corrodes them is watching things die quietly and learning that the plan does not mean much.

Where Technology Helps, And Where It Absolutely Does Not

There is a version of this problem people try to solve with software, and it deserves a fair hearing. AI is genuinely good at the admin that surrounds execution. It will pull a status summary out of the tools your team already uses, draft the update nobody wants to write on a Friday, and tell you which priority has not been touched in three weeks. That is real time back, and it is time currently being spent on reporting rather than doing.

What no model will do is choose. Deciding that the second location waits and the quoting system goes first is a judgement about the business you are building, made with information that mostly lives in your head. Point AI at the tracking and keep the deciding. That is what amplifying intelligence looks like in practice, and it is the same reason most AI pilots impress everyone and change nothing. A tool that reports on fourteen stalled priorities has given you a very tidy view of a problem you have not addressed.

What A Real Plan Looks Like

One page. Three lines on it. A name and a date against each. A short list underneath of what the business is deliberately not doing this quarter, which is usually the more revealing half of the page.

It looks embarrassingly small next to the version with fourteen items and a colour-coded matrix. It is also the one that gets done, and a team that finishes three things develops an appetite for finishing the next three. That appetite is the actual asset. Momentum is not a mood, it is evidence, and people only believe the plan once they have watched something on it get finished.

This is the middle of the arc I work through with owners. Clarity on the strategy, execution with momentum, then growth. Clarity is the part everyone enjoys. Execution is where the year is actually won or lost, and it is mostly won by having the nerve to run a shorter list than feels safe.

Go and find the document from your last planning day. Count the items on it. Then count the ones that are genuinely finished. Whatever gap sits between those two numbers, that is the real conversation, and it is a much better use of an hour than writing a new plan on top of the old one.

Frequently Asked Questions

What is the strategy execution gap?

The strategy execution gap is the distance between what a business decided to do and what it actually did. It shows up as a plan that everyone agreed with and nobody finished. Research tracking strategic plans between 2017 and 2024 found that 84.5 per cent of strategic projects never reached completion, and only 5.7 per cent of organisations completed three quarters or more of what they set out to do. In most cases the thinking was sound. The plan simply asked for more work than the business had capacity to deliver, and no single person owned any given piece of it.

How many priorities should a small business work on at once?

Fewer than most owners are comfortable with. The completion data is blunt about it: plans with fewer than 20 elements produced high performers 68 per cent of the time, plans with 40 to 60 elements completed at 15 per cent, and plans with 60 or more completed at 8 per cent. For a small or medium business with one leadership team and no spare capacity, three active priorities at a time is a sensible working limit. Everything else goes on a visible later list so it is parked rather than lost.

Why do business plans fail to get executed?

Usually because the plan was written as though a second business existed to deliver it. The same people who run operations are the ones expected to build the new system, fix the onboarding and open the second location, on top of a full week. Add to that a lack of single-person ownership, no agreed definition of done, and a weekly meeting that reports on status rather than making decisions, and the plan quietly loses to the work in front of everyone.

How do you actually make a strategic plan happen?

Cut the list to three things, give each one a named person, a date and a written definition of done, and say out loud what stops or slows to make room for them. Then meet weekly for twenty minutes with three questions: what moved, what is stuck, what do we drop. Sequence the rest rather than running everything in parallel. Momentum on three finished things beats activity across fourteen unfinished ones, and finishing is what builds the appetite to keep going.

Can AI help with strategy execution?

It helps with the admin around execution rather than the execution itself. AI can pull a status summary out of the tools your team already uses, draft the update nobody wants to write, and flag a priority that has not been touched in three weeks. What it cannot do is decide what matters. Choosing which three things to run, and what to stop in order to run them, is a judgement call that depends on the business you are trying to build. The expert stays in the loop and makes the call, which is the whole point of using AI to amplify people rather than replace their thinking.