Why the same problem comes back with another name

A building-maintenance services company, 300 people, 60 active contracts (a mid-size Brazilian company, figures in Brazilian reais). In February, two large clients complain about delays. Leadership creates a task force: two coordinators step away from their own routines for six weeks to closely track those contracts. It works. By April, both clients are satisfied.

In July, four other contracts fall behind — precisely the ones left without coordination while the two were assigned to the task force. New task force. In November, the cycle repeats, now under a different name: critical contracts recovery committee. Nobody, in any of those meetings, asks the question that matters: why does this happen three times a year?

01The idea: structure produces behavior #

Peter Senge published The Fifth Discipline in 1990. The fifth discipline in the title is systems thinking, and it organizes the other four — personal mastery, mental models, shared vision and team learning. The work draws on the system dynamics developed by Jay Forrester at MIT since the 1960s, which described how feedback loops and delays produce counterintuitive behavior in industrial systems.

Senge's central argument is that an organization's repeated behavior stems from the structure people operate in — not from their character. Replacing the people while keeping the structure tends to reproduce the same behavior, because the incentives, delays and feedback loops stay the same.

Senge also popularized systems archetypes: patterns that reappear in very different contexts. The one in the story above has a name — fixes that fail. A symptom appears; a correction is applied that relieves it quickly; the relief is real, but the correction generates, with a delay, a consequence that worsens the original symptom. Because the relief is immediate and the consequence is distant, the organization learns exactly the wrong lesson: that the fix works.

In building maintenance, the loop is visible once drawn: pulling coordinators from their routines resolves the contracts in crisis and neglects the rest; the neglected ones enter crisis the following quarter; the company concludes it needs more task forces. Each round consumes more coordination capacity and leaves the operation more fragile than the round before.

02Mental models: what keeps the loop invisible #

If the pattern is so clear on paper, why does nobody see it? Senge attributes this to mental models — the generalizations we carry about how things work, which we rarely put to the test. In the example company, the prevailing model is simple: a contract falls behind for lack of close follow-up. It's true in each isolated case and false as an explanation of the whole.

Two practical obstacles feed this. The first is the delay: five months pass between February's task force and July's delay, and average organizational memory doesn't cover that span. The second is that nobody logs the connection. The February and July meetings exist in separate documents, likely in different presentations, made by different people.

That's why organizational learning isn't a training topic but a record-keeping one. Without a history of what was decided and what happened afterward, every quarter starts from zero — and the company is condemned to rediscover the same thing. It's the same distinction between noise and signal we described in variation is not error: without a historical series, every point looks like a new event.

03The cost of the cycle that repeats #

At the scale of the 300-person company, with 60 contracts. Stated assumptions, to redo with your own numbers:

Want to see how this looks inside a real operation? Explore the platform.

The time math: 2 coordinators × 6 weeks × 40 hours give 480 hours per round; × 3 rounds, 1,440 hours a year; at a fully loaded R$ 150 an hour, about R$ 216,000. The second line uses the company's own experience: two discounts per round, averaging R$ 18,000, plus one unrenewed contract for the year, with estimated annual margin of R$ 110,000. The numbers are estimates; the direction isn't.

And there's a cost that doesn't make the spreadsheet: the organization learns that crisis is the only way to get attention. Once learned, that lesson changes everyone's behavior — including those who would rather have flagged it early.

04Where this breaks down in practice #

1. Every crisis is treated as unprecedented. There's no record linking today's incident to the one five months ago, and the repetition becomes coincidence.
2. The quick fix gets rewarded. Whoever puts out the fire gets noticed; whoever prevents the fire has nothing to show in the meeting.
3. The side effect lands in another area. The task force is measured by the contract it saved, never by the ones left uncovered.
4. The event gets discussed, not the pattern. The meeting analyzes the specific case and never the series — which blocks any conclusion about structure.
5. The person gets blamed. The coordinator on the problem contract is replaced, and the new one delivers the same result six months later.

05What management with data answers #

Organizational learning depends on memory, and memory depends on logging where the work happens — not in minutes nobody reopens. Incident with a date, cause on record, decision with an owner, and the outcome observed afterward: with those four things, the pattern that today takes five months to notice shows up on the first query.

It also changes what gets discussed. With history available, the meeting stops spending time reconstructing what happened and starts looking at the series. It's the direct complement to learning how to learn and to knowledge that doesn't circulate isn't an asset, and it connects to what culture reveals under pressure in the three levels of culture.

At Relevanti, the process, operations and intelligence modules keep that history with no extra work — see the platform, the breakdown by area in solutions, or bring a concrete case to a conversation.

If the same problem comes back every quarter, it isn't a problem. It's the expected outcome of the way work is organized today.