FP&A and performance management

A reliable forecast starts with the data

A forecast that is questioned every month rarely needs a smarter model. What is usually missing is one data foundation, clear drivers and ownership.

· · 3 min read

Many finance teams look for the answer to an unreliable forecast in a new planning tool or an AI prediction. Understandable, but the problem usually sits earlier. Sales, operations and finance each work with their own numbers, the forecast is a sum of spreadsheets, and every month the meeting is about which number is right instead of what we are going to do.

One data foundation

A forecast can only be trusted when actuals, order book and pipeline come from one source, with agreed definitions. What counts as revenue, when does an order count, which exchange rate do we use? As long as each department answers those questions differently, you are comparing apples with oranges and the forecast loses its authority.

Steer on drivers, not totals

A total says little about why it changes. Build the forecast from the drivers that really matter, such as volume, price, mix and conversion. The conversation then moves to the assumptions behind the numbers, and scenarios become a matter of adjusting a few drivers instead of rebuilding a spreadsheet.

Make ownership explicit

The business owns the input: sales owns volume and price, operations owns capacity. Finance guards consistency, the assumptions and the translation into results. Also measure forecast accuracy and discuss the variances. Not to blame anyone, but to learn from every cycle.

And AI?

AI can help: drafting a first version, flagging variances or preparing the commentary. But AI amplifies what is already there. On a messy data foundation an AI forecast goes just as wrong, only faster and more convincingly. So the order stays the same: first the data, then the drivers and ownership, and then automation.

Curious where your organisation stands? The data maturity check gives a first picture, and the FP&A and performance management page explains how I help.

Read also

Let's talk

Would you like to exchange ideas on data, analytics, AI or FP&A? I would be glad to connect.