Data has regularly been dissed in the past for the conclusions it has led to. But data is data; it’s not really the villain, but the  victim of those who use – and abuse – it. What is clear in today’s world of vastly increased computer power is that data is a key engine of conclusions about a vast variety of issues. It’s becoming an integral part of life on Planet Earth.

If so, we need to know how to handle it, and how to leverage it in our businesses to get the most out of it. Competitiveness is increasingly being defined by how we use data to get the right signals, and make the necessary adjustments. Data is also key to a business function that is growing in practical usage and sophistication: forecasting.

Forecast: there will be more forecasts!

The practice of forecasting has had its ups and downs. The first models developed to predict the economy were greeted with great fanfare, only to prove a great disappointment, as they could not account for all the possible disruptions economies could (and did) encounter. Predictive models still have their problems, but advanced techniques are enabling models to self-correct, to automatically refine themselves, in effect to have on-board learning functions that continuously drive them to greater accuracy. There are applications in the financial world, but also in key artificial intelligence functions, machine learning and other such applications.

This can all sound daunting for the average small- and medium-sized business, as it looks like something only the big operators can afford. At the end of the day, it’s still the decision-maker that’s accountable for interpretation and use of the data, which brings me to a key – and pretty basic – point: there are still some simple data concepts that elude even the well-trained. Take as an example yearly growth. It is used in the economy, in business everywhere and in countless personal applications. It is calculated by taking the sum of this year’s activity over the sum of last year’s activity. Simple, right?

It’s so simple that it can lead to the conclusion that every year starts with a clean sheet, and that everything is based on the growth experienced in each month. Sounds reasonable – but it’s not 100 per cent correct. The calculation is actually average activity this year over average activity in the prior year. What that means is that if a business ends a year on a high note, and that level of activity is sustained through the next year, then last year’s in-year growth is determining this year’s forecast. Put another way, if last year ends with sales 5 per cent higher than average sales for the year, then with zero growth in any month this year, annual growth is actually 5 per cent. That’s what we call built-in growth.

What this reveals is that this year’s reported growth is actually being determined a lot sooner than is commonly thought. In fact, in a quarterly data series, next year’s growthrate starts being determined in the second quarter of the current year. In a monthly series, it’s even earlier – this year’s growth actually began getting cooked in February of last year!

Data interpretation makes all the difference

Why is this useful? Well, consider explaining to your boss that the sales team is going flat out, and the boss gives you the toss, because overall performance is still 10 per cent below last fiscal year. Both of you are right – monthly growth can be red-hot, but if last year’s sales tanked toward year-end, the year-on-year growth could indeed be quite negative.

It is amazing how often this simple mistake is made, and sometimes at the highest levels of power. Of course, if the growth path is linear (whether up or down), this whole argument is far less important, as even casual observers will come to the correct conclusions. Where this becomes critical is in times of disruption – which just happen to be make-it-or-break-it moments in business, government and personal finances.

The bottom line?

Data is not always what it seems to be, but getting it right is becoming even more critically important. As we increase in sophistication, it is important that the ultimate interpreters of this stuff keep a good handle on the basics.