Is Your Forecasting Effort Worth the Effort?

I was asked to update the project forecast, something that is expected every month. Nothing unusual there. The existing forecast was interesting. The level of detail it went in to was intense.

The natural thing to do, is to repeat what everyone else has done. The natural thing to do, on a $10M programme, is to continue forecasting the cost of every resource into 15 minute intervals every week, showing the dollar and cents cost for each person, because that was what had been done before. That's the organisational norm.

Those who have worked with me know i like to shake up the norm. Particularly if it makes absolutely no sense.

What's the data being used for?

When you're into your forecasting routines, take a moment to think about what the data is being used for.

• In some organisations, forecast is used as a proxy for capacity. We know we have capacity for $10M of work per month. We total all the forecasts and that will tell us if we are trying to do more work than we can. Granular accuracy can be important here.

• In other places, forecasts are used for comparison to budgets and benefits and to validate if the “thing” still makes sense. Granularity here is just effort spent without value received.

• A middle ground could be where forecasts are used as measures of performance, or to understand where more support is required, or to make prioritisation calls.

What the data is used for should govern how carefully the data is collected, manipulated and reported and to what level of granularity.

What's the cost of data collection?

Just like in our business cases where we validate that the benefit is worth the cost, we should do the same for our forecasting cost.

Even if our forecast data is used as a proxy for capacity, and even if that is done at a team level (so granular accurate data has currency), if the cost of producing that data is worth more than the benefit, I would challenge validity of the exercise.

Is it really necessary to know the exact number of hours and the charge rate for each resource on your project? Or can we use an average charge rate and forecast usage at a team level? How we collect the data affects the time and the cost, but may not significantly affect the utility.

Finding the balance

As with sitting and standing, or eating salt and sugar, everything is OK within balance. Forecasting is OK as long as the cost and the benefit are in balance.

So take a look at your routine.

• What can you change and give an Order Of Magnitude answer for instead?

• Where can you assume averages and forecast in aggregate to save time?

• Where do forecasts rarely change and it's safe to assume continuity and not necessary to get into the detail every period?

Everyone is busy. Balance your forecast effort and value to create more time for the things that you care about.