Where forecasting fits on a long engagement.

A point forecast is close to useless on its own. The planning question is not "what will demand be" but "what should we prepare for", and that needs a range. A forecast of a thousand units means something quite different if the interval is nine hundred to eleven hundred than if it is four hundred to two thousand.

Simple methods are a genuinely hard baseline to beat. Seasonal naive — this month equals the same month last year — outperforms elaborate models on many business series, and a project that does not check against it has no evidence that its complexity earned anything.

What an assigned team does with forecasting.

Forecasts need monitoring after deployment, which is the step most often skipped. A model that was accurate at launch degrades as conditions change, and without tracked error nobody notices until a planning decision goes badly wrong.

That continuous evaluation is standing work rather than a delivery milestone, held within an agreed committed monthly capacity.

What we use forecasting for.

  • Intervals rather than point estimates A range to plan against, because the uncertainty is the decision-relevant part.
  • Complexity justified against a naive baseline Checking that a model beats seasonal naive before it is trusted.
  • Accuracy tracked after deployment Forecast error monitored, so degradation is caught before a plan depends on it.

How forecasting capacity is assigned.

Forecasting capacity is assigned under managed data services, with post-deployment accuracy tracking treated as part of delivery.

Tell us what your roadmap needs forecasting for.

A service delivery manager replies with the disciplines we would assign, the monthly capacity and what the first month looks like.

Loading the contact form… You can also email hello@azendo.co.

We reply within one working day. No obligation, and no newsletter.