How Fujitsu SSL Used Analytics to Prove Learning & Development Success
Fujitsu SSL needed to prove learning and development initiatives success and that brings us to their use of analytics. Read their story.
Good analytics stories should always start with an important business issue. In the case of Fujitsu SSL, the issue was transforming the business to achieve seamless cohesion between the solutions and systems integration parts of their organization.
An important enabler of this transformation was a type of learning intervention called peer coaching. This technique brings small groups of managers together to learn from each other and apply what they learned to the transformation project.
Fujitsu SSL needed evidence as to whether or not peer coaching was working and that brings us to their use of analytics.
Determining Learning & Development Success: Four Areas to Analyze
Analytics in learning and development can be notoriously difficult, especially if we think analytics means calculating ROI.
Did Fujitsu SSL need a specific ROI? No. They only needed to decide whether or not to continue the peer coaching program. A specific number is unnecessary because if the program was working to drive transformation, then the ROI would be through the roof, and if it wasn’t working, then this meant a different approach to supporting transformation was needed.
Can analytics prove a learning intervention is working? Not definitively, but it can provide the evidence necessary to making an informed business decision.
In Fujitsu SSL’s case, their leadership needed to know whether, all things considered, the data suggested the peer coaching program was working. To answer this realistic question, they analyzed four areas:
How did the business units that did a lot of peer coaching perform compare to those that didn’t do as much?
How did managers involved in the process rate the peer coaching sessions?
How did individuals who took part perform?
How did the learning intervention affect employee satisfaction?
To address the first question, they needed to be able to combine Finance data on unit performance with the learning department’s data on the peer coaching intervention. The actual analysis wasn’t intimidating, just a plot of unit profits against the number of coaching meetings in each unit. The results showed a strong correlation between peer coaching and business results. It’s not proof, but this data strongly supported the view that the learning intervention was helping with the transformation and should be continued.
The second question involved gathering survey data from participants for purposes. One finding was that managers rated the sessions highly. This data added support to the hypothesis that the program was creating value. The second finding was that some learning modules were more effective than others and this allowed the company to improve the program.
The third question involved looking at individual performance ratings versus participation in the learning program—again a matter of bringing in data from different systems. The results, though not positive for every individual, showed that more often than not those involved in learning improved their performance.
Finally, they looked at employee satisfaction and again saw a positive correlation with involvement in the program.
These findings lead me to ask two questions; Based on this collection of evidence would you continue the program? (Let’s not play around here, the answer is yes). Secondly, what do you feel about a training department that tests the value of an important program against the evidence in this way? (My answer is that it builds credibility, which will be valuable even if they can’t do analytics on every learning program).
Lessons Learned
There are several important lessons from their approach. The first one is what Peter Navin, who leads HR at Grand Rounds, describes as a “mosaic of measures.” No one piece of analysis provides a definitive answer, but when you piece together a mosaic of measures that creates a clear enough picture for management to make an informed decision.
Secondly, you need to be willing to invest in a leadership development initiative for a period of time before you can tell if it is working. It is inherent in the nature of leadership development that it takes time for the impact,–or lack of impact,–to show up. Fujitsu SSL had nine years of data on their program. If you are serious about improving learning, then you must be patient. This doesn’t mean blind faith, but it does mean that analysis is done right when we have enough data to draw reasonable conclusions.
Finally, just to repeat a point made earlier, this data doesn’t definitively prove this particular learning intervention had a positive outcome, but it does provide enough evidence that leadership can feel confident that the initiative makes sense–which is exactly what the training department needs to deliver with learning analytics. And as long as they make a modest investment in the relevant tools and skillsets, it’s possible to deliver this kind of detailed analysis faster and even more accurately.
