In the world of business, barely a day goes by without an article popping up about how artificial intelligence, predictive analytics and machine learning can be used to support better business decision making.  Whilst I would consider myself to be pro-technology, I am also pro-good-decisions. So does one necessarily lead to the other? Here are a few considerations.

What is possible to predict?

Predictions are usually made by looking at historical data and then using this to estimate what might happen in the future. With so much historical data being created and stored these days, it makes complete sense to try and use it to make better predictive decisions.  And with computing power continuing to increase, there is almost no limit to the amount of data that can be crunched and analysed quickly.  Finding that needle in a haystack that leads to radical improvements in employee experience, customer experience and financial performance is what most Leaders dream about. 

If there is a problem, it is that in order to create a predictive model that is in any way useful, a very large amount of accurate data is required.  Predictive analytics is only really possible when this accurate historical data actually exists, and predictions are rarely possible when no data is available. This is great for regular-occurring and repeating circumstances, but not super useful for changing circumstances.  An example is the weather.  Whilst we now have extremely good historical weather records which can be used to predict the likelihood of rain in any given future period, the historical data can’t tell us whether a cyclone will hit at an exact point in time.  For those planning a holiday more than a few weeks in advance, packing for various weather conditions would still be required. Whilst we can assume that it will be hot in summer and cooler in winter, and the probability of rain might be higher or lower at certain times of year, a predictive algorithm is likely to miss a once-in-ten-year event. Relying on what happened this time last year may or may not be a good idea.

The point here of course is that for many companies and their leaders, whilst they have access to very good historical data, they don’t have the completeness of data that is required to make better decisions. Often, the impact of limited, ambiguous or unknown data can lead to worse decision making if relied on! Which the begs the question, what role should ‘gut feel’ based on experience and logic play? Data-driven is not necessarily going to create better decisions.

If there is another problem, it’s that focusing efforts on collecting substantial amounts of historical data can accidentally anchor everyone to the past. We call it becoming reactively predictive – waiting for the past to guide the future.

Is being proactive a better alternative? 

Let’s now, for argument’s sake, assume that we have no historical data whatsoever about something we wanted to make a decision about, how would we go about making that decision? Most likely, we would still try to gather whatever information is available, but rather than trying to rely on historical information (there is none), we would be forced into looking at things as they are right now. In other words, our attention would be focused on what our goals are, what the current situation is right now, what our options are, what our best course of action is and what our contingencies should be. In other words, we would have to become proactively present and focus on responding to the circumstances as they are.

The point here is that if intentions/goals are clear, and we are focused on the reality of the circumstances and adjusting accordingly, and we aren’t being distracted by the way things should be (ambiguous data), we can still make our way in the world. By relying too much on data, the danger is that we are not practicing the skills required to determine and adapt to current circumstances without data. The danger is that the blind start leading the blind.

For a smaller company without the budget or resources to go and collect lots of historical data and crunch it accordingly, perhaps focusing on being proactively present, and developing the skills required to compete without relying too heavily on data, is not a bad bet.

How does My Employee Life support a more proactive workplace? 

An important part of the My Employee Life framework is to proactively define the wildly important things that really matter, in advance, so that everyone is clear about intentions, directions and rules of the game. Another important part of the performance drumbeat process is the weekly worklog, which has everyone proactively reflecting on their week, adjusting as required and setting intentions for the following week.   

My Employee Life still collects plenty of data and we are currently looking at ways to improve our predictive capabilities, but we are also happy to recommend to our customers to not lose sight of the importance of staying proactively present. Whilst your competitors are drowning themselves in too much data which is causing too much unnecessary chaos, your workplace can be focused on adapting to the current circumstances as they are.