A major problem for organisations is that analytics consultants conduct (relatively simple) analysis but don’t explain the techniques, or detail precisely how they conducted their analysis. If, and when, the analysis is needed to be repeated, only those original consultants are able to conduct the analysis. Any new analyst has to first figure out what was done before.
Some consultants believe that this “black-box” approach serves them (and maybe the client) best. It often guarantees return business.
JumpData do not believe this is always the correct approach. We believe in a collaborative approach, so that our clients can understand what we have done, why we have done it, and are able to conduct (if only simplified versions) our analysis in future. We also believe that we should be able to share some of technical knowledge with our clients, so they may improve in the areas in which they are weak – and we may also learn from them.
Recently we designed a new methodology for a client for a pseudo-conjoint study. We conducted the analysis then handed over the results. However, we also presented a detailed explanation of how the technique worked, how it could be used in future, and left examples of such in two pieces of software (including a version in Excel).
For, although the client could model future projects using the workbook we left with them, we saw it as an opportunity for us to develop the approach together. They after all, had more knowledge of the market they were going to using the technique in, and could make useful suggestions in the technique that we hadn’t thought of.
Responsible for data analysis, modelling and automation