The Early Talent Industry is a tricky obstacle course with new challenges constantly arising. You need to ensure your programmes stick to budget, while still creating an impact. You need to predict the numbers of new joiners your company will need over the next year, usually a year in advance. And on top of that, you need to demonstrate that your programmes are providing ROI to your company. However, proving the benefits your programme provides can often be the most difficult of the lot. You have mountains of data, but little time to review it all. Sometimes, you need to be able to process key pieces of evidence quickly. You need to know how to distinguish, in the words of Nate Silver, the signal from the noise.

Silver’s ‘The Signal and the Noise’ looks at the art of prediction. Why can we predict the weather for tomorrow, but can’t predict an earthquake? How can we predict stock market fluctuations? For the Early Talent Industry, the following ideas could make those mountains of data manageable. Ready? Here we go (as predicted…):

1) Data Trends and Patterns

We are forever attempting to recognise patterns within data to save time. Unfortunately, we can create patterns where none exist or try to fit data to patterns we think we’ve spotted. Being able to take that all-important step back and objectively identify the trends that are occurring in your recruitment and development processes, instead of fitting data to meet the trends you want to see is a crucial skill. At the end of the day, a square peg will never fit a round hole. However, before you can objectively identify trends, you need to recognise what your data is telling you.

2) Tools are only as good as their users

There are many types of data analysis software on offer to help analyse and process any information we may have. However, software is often a tool created to help humans make better decisions. It is limited by the data it is given and by the way it has been programmed. For example, there are often “unknown unknowns” that haven’t been considered when inputting data, but that also have an impact. One way to mitigate the risk of “unknown unknowns” affecting your data analysis is to think probabilistically. By creating a range of possible scenarios using information from previous years, you can not only predict potential new joiner scenarios, but also hopefully cover any “unknown unknowns” that may not have been previously considered.

3) Do your predictions work?

Time and resources can be wasted if the scenarios you predicted to occur, didn’t happen. If for example you predicted that the business would require fifty interns across a business year and in actuality they required only 20, you will have wasted a lot of time and effort for little return. One way to ensure you are saving your resources and time, is to regularly review how your predictions matched the actual scenarios you saw. This will allow you to improve how you predict future business demand as well as improving how you spot true patterns and trends in your data. What review processes do you use to improve your prediction techniques?

4) Don’t just focus on the trees

Statistical data can sometimes be misleading, so remembering the broader context can be helpful. If your data tells you your Early Talent are unhappy with the experience they have had with your company, it doesn’t necessarily mean you need to give your entire programme a dramatic makeover. Was there a reason why your Early Talent may have been unhappy? Was there a problem with the technology used or a delay in communications that might have affected their experience? Don’t lose sight of the forest, by only focusing on the trees. Instead of focusing only on the detail, place your data in the bigger picture.

Data can help make our jobs easier. How can you ensure you are using your data and time effectively to make certain you create the biggest impact possible in your recruitment and development processes?

By Saj Jetha

Tweet us your thoughts @thesmartytrain

Image Designed by Freepik