Why Change Reverts Six Months After Go-Live

Go-live gets the celebration. Six months later, someone opens the dashboard and the numbers have quietly slid back toward where they started. The system is still running. The training still happened. And yet the organization is behaving almost exactly as it did before the project began.

This is the most common failure mode in transformation, and it is also the least examined. It doesn't announce itself. There is no crash, no incident report. Change just erodes, a little each week, until the results disappear.

The revert curve nobody plots

Most project plans end at go-live. The Gantt chart runs from design to build to launch, and then it stops. What actually happens after launch follows a predictable shape.

For the first few weeks, adoption looks strong. People are trying, managers are watching, the project team is still in the room answering questions. Behavior holds because attention is high.

Then attention moves. The steering committee closes the program. The budget is spent. The senior people who sponsored the work turn to the next priority. And the moment the pressure lifts, the organization starts optimizing for what is easiest under load, not for what the new process asked of it.

The old workaround comes back first because it is faster when someone is behind. The spreadsheet that was supposed to be retired quietly becomes the real source of truth again. A manager, stretched thin, stops asking about the new behavior in one-on-ones. None of these are decisions anyone made on purpose. They are the default the system falls into when no one is holding the new default in place.

By month six the drift is large enough to show in the metrics. By the time it shows in the metrics, it has been happening for months.

Why sustainment is the phase that gets starved

Reinforcement and sustainment is the phase that decides whether all the earlier investment holds. It is also the phase that gets the least funding and the least senior attention, and the two facts are related.

Design and go-live are visible. You can see the new system. You can point to the training numbers. You can take the photo. Sustainment produces nothing to point at while it's working. Its entire job is to make sure that a thing that already happened keeps happening. When it succeeds, the reward is that nothing bad occurs, which is difficult to put in a status report.

So the budget is written around the visible work. The consultants are scoped through launch and then they leave. The internal team that ran the program is reassigned. The one phase that determines the return on everything else is treated as a wind-down instead of a discipline.

What actually holds change in place

Sustainment is not a communications campaign or a follow-up email. It is a small set of concrete mechanisms that keep the new behavior from decaying.

Managers inspect the behavior

Behavior sticks when the people who run the day-to-day expect to see it. That means the new process shows up in one-on-ones, in team reviews, in how work gets checked. If a manager never asks about it, the message employees receive is that it doesn't matter, regardless of what the launch memo said.

Someone owns the metric after the team leaves

A metric with no owner is a metric that stops being read. Before go-live, name the person who will watch the adoption number in month four, month six, month nine. Give them the authority to raise a flag and the standing to be heard when they do.

Remove the old path

People revert to the old way because the old way is still available. The single most effective sustainment move is often to take it away. Decommission the legacy system. Delete the parallel spreadsheet. If the fallback exists, it will be used the first time the new process is inconvenient.

Keep a cadence past the excitement

Change needs to stay on the agenda after it stops being interesting. A short, regular review that survives past the launch buzz is what carries adoption through the quiet months where reversion actually happens.

Plan sustainment before you launch

The mistake is treating sustainment as something you turn to when the metrics slip. By then you are recovering lost ground, and the organization has already relearned its old habits.

I plan the sustainment phase during design, alongside the build. Who owns the behavior. Who owns the number. What gets shut off and when. What the review cadence is and who runs it after the project team is gone. This is the work that outlasts the engagement, and it is the reason the results survive contact with a normal, busy quarter.

Transformation lives or dies with the people who have to keep doing the new thing long after the launch was interesting. Fund the phase that keeps them doing it.