By jPablo Caballero | December 14, 2017
“SPIN selling” is a sales method from the eighties which is explained in a book of the same name. From the eighties! Why on earth did it trigger this reflection on xAPI?
I recently reread an old classic: “SPIN Selling”, by Neil Rackham. The edition I have is from 1988, that is from almost 30 years ago, some 12 or 13 years before SCORM came into being. Old. That is why I was surprised when, toward the end of the book, it started to make me think about xAPI.
First, let me provide just a bit of background on the book. The author and his organization, through lots of research, came up with a sales model that focuses on large sales. Supposedly, this method improves sales performance significantly. The book explains the method, how they came up with it working with many companies, their research and so forth. Part of their work involved observing the salespersons’ behavior, training them on the new method, and evaluating the characteristics and results of sales calls after the training. These processes took quite a lot of work in terms of data collection and analysis.
But let's leave the method itself aside for a moment, assuming that it works, and let's focus on the training part. Sales techniques is an area where training can potentially shine in terms of ROI: if the training is effective and causes a change in behavior (the sales agents actually apply the new method), there will be a direct impact on the bottom line (again, assuming that whatever was taught was indeed a better sales technique).
Of course, L&D is interested in proving the ROI of any kind of training, or at least proving that training has a positive impact on some aspect of the business, that it improves something. Having lots of data can be extremely useful. With it, we can draw up correlations, find patterns, etc. And this is what xAPI can give us: lots of data from different sources.
Back to the book, concretely to Appendix A which is the part that I found very interesting and made me think of xAPI. After training sales agents on his new technique and seeing positive results time after time, the author was still not comfortable with just the feeling that both the method and the training were working. He wanted to prove that the positive results were due to his method and the training on it. He dug into several apparently successful projects and found information that disproved that the positive results were due to his method and training. In some cases, the success could be traced to sector and market conditions, in others cases to organizational changes that had taken place just before the training, in another one the problem was the wrong selection of the control group versus the training group, etc. What I find interesting is that after finding positive results by correlating the training data and the performance data, it took deeper analysis and more data to find the truth. Factors such as the selection of control groups, the Hawthorne effect and others had a significant impact on the final results.
When people talk about the potential of xAPI to help measure the ROI of training, it is usually in relation to training data and performance data, and the correlation between them. Yes, we need both types of data, and xAPI can definitely help us get it in an efficient way. But what I gather after reading appendix A of this book is that still, we should always be on the lookout for other factors that might be skewing our correlations and results. So, not only must we keep in mind that correlation is not causation, but also we should be careful to find the right correlations.