We Built the Capacity. The Practice Never Changed.

Why institutions invest in units, training and systems, and still decide the way they always did.

Published 2 November 2026 · 5 min read · 6 movements · Harpi Advisory

The launch went well. That is rarely where the problem is.

A unit was established. The team was trained, the system installed, the new process documented and presented to a room that applauded it. The report to the board, the ministry or the funder recorded the milestone in green. Everyone involved can show what was built.

A year later, the same decisions are made the way they were made before. The same two or three people approve them. The trained team routes questions upward. The system holds data that nobody opens before a meeting. Nothing failed. Nothing is visibly broken. The capacity exists, in the organisation chart, in the budget line, in the slide deck. It is simply not how the organisation works.

This is a different failure from the earlier ones in this series. It is not the opportunity nobody acted on, or the decision that was made and never moved. It is the capability that was built and never absorbed. The capability is visible. The inertia is real. Why does building it not change how the organisation works?

Because building capacity and absorbing it are different jobs, and they are rarely given to the same people.

Building is a project. It has a budget, a timeline, a delivery team and a date on which it ends. Absorption is a change in daily habit, and nobody is assigned to it. When the project closes, the people who built the capability move on, and the people who were meant to use it return to a calendar that was already full.

Capacity is also easy to count, and use is not. People trained, units created, tools installed, processes documented: all of it can be reported precisely. Whether a single decision is now made differently takes sustained attention to see. Organisations report what they can count, and over time they start to believe the count is the result.

And the old route persists because it is faster and safer. The person who was trained is still quicker at the old way. The old way has known failure modes and known allies. The new way has neither yet, and the first person to rely on it carries the risk if it falters. In environments where reputation travels fast and relationships carry real weight, that risk is not abstract. Nobody refuses the new capability. They simply choose, one routine decision at a time, the path that has never embarrassed them.

It hides inside reporting that shows success.

The measures are all measures of delivery: trained, installed, launched, handed over. Nothing in them asks whether the work now travels differently. A programme can complete every milestone and change nothing about how a single decision is made.

Pilots deepen the illusion. A new process works while someone senior is watching it, because attention itself supplies the missing incentive. Once attention moves to the next priority, the process quietly reverts, and the pilot stays in the record as proof that it worked.

The tells are small:

It looks different by setting. In a family enterprise, the founder still approves what the new managers were hired to decide. In a public body, authority moves upward by habit, whatever the new structure says. In a donor-funded unit, the capability lasts as long as the funding cycle does. The mechanism is the same: the structure changed, and the decision rights did not.

Building the capacity is not the same as changing the practice.

The organisation pays twice.

It pays first to build the capacity: the training days, the systems, the unit’s salaries, the advisers’ fees. It pays again in the value that capacity was meant to unlock and never did. The second cost is rarely itemised, because it never appears as a line anywhere. It appears as slower approvals, as decisions concentrated in the same few people, as growth that still waits on the bottleneck it was meant to remove.

Credibility pays as well. People who were trained and then not used learn that the programme was an event, not a change. The next initiative inherits their scepticism, and the cost of building anything new quietly rises.

The third cost compounds. Growth keeps arriving, from new business, new volume, new obligations, and it lands on a structure assumed to be stronger than it is. The organisation plans around the capacity it built, not the capacity it uses. The gap shows up at the worst moment, when load is highest and there is least time to close it.

The costliest error is the diagnosis. The organisation concludes that the capability was the wrong one, or that the people were not up to it, and commissions another build. The real problem was never the capacity. It was that nobody was made responsible for using it.

Not another build. A fourth training programme on top of three unused ones simply lengthens the list.

Go back to launch day.

The capacity was real. So was the old habit. The programme built the first and assumed it would replace the second. It did not, because habits are not replaced by capability. They are replaced by a decision, made on purpose and owned by someone, to stop doing it the old way.

This is general commentary on patterns we see across organisations, not advice on any particular situation. Read the full disclaimer.