That’s the first thing we look at. Before recommending a rebuild, we assess the model structure, data, and how it’s being used to understand what’s actually causing the problem. If starting again is the right option, we’ll say so. But in our experience, it often isn’t necessary.

Typically one to two weeks, depending on model size and how much documentation exists. You get a written read on what's working, what's holding it back, and what each option would involve.

A second opinion is always valuable. What looks like a model that needs replacing may sometimes be a structural issue that can be addressed. We take a fresh look at the model before recommending whether to improve what’s there or start again.

That's a real possibility at scale, and it's a different answer to a structural fix. Sparsity is where Classic gets expensive, and a model that's hit that ceiling needs Polaris rather than optimisation. We assess for it during diagnosis rather than discovering it halfway through a recovery.

Usually, yes. It’s a common reason for low adoption. Often, the problem isn’t training but how the model fits into the way people actually work. If the process feels difficult or disconnected from their day-to-day needs, more training can only go so far. Sometimes the model itself needs to change.

Almost always less, and often substantially. A recovery is typically weeks rather than months. The honest caveat is that we can't tell you the number before the assessment, which is why the assessment comes first.

Most clients move to a support arrangement while their team rebuilds confidence, then expand scope from there. That's what happened with the engagement above: four weeks of recovery became a support contract and then a full Polaris migration.

Our team handles the work directly. Every consultant is an EliteEPM employee, and the partner who scopes the engagement stays involved through go-live. We keep delivery within our own team, so you work with the people who understand your project from the start.