Spend Analytics – Understand Who Is Doing What!

Hot Topic

We’ve been talking to a few Spend Analytics firms recently, and one point that hit me during those meetings seemed to be worth a quick discussion.

Back when I was a CPO and then a consultant, my general understanding was that if I engaged a firm to carry out a spend analysis for me, that firm was doing the work themselves. In the olden days, that meant a team of consultants slaving away translating AP data into Excel spreadsheets, coming up with the good old “spend cube” we used to love so much.

Then things started to get automated, with the extraction of spend data from ERP and other systems and at least some basic classification done without too much human intervention. With the development of AI and machine-learning, the systems have got better and better at doing much of that hard work.

But procurement executives who are buying spend analysis should make sure they understand just who is doing what. In some cases, you may well find that there are two, three or even more firms involved in producing your final product. Now there is nothing intrinsically wrong with that, but any good procurement professional likes to understand the supply chain, particularly where sub-contractors might be key to the delivery of the final product or service.

So in the case of spend analytics, you might want to talk through these elements of the process with your provider or potential provider, and be clear about who will provide and do what in terms of the following;

  1. The underlying “engine” that extracts, cleans and classifies data.
  2. The manual checking of that output, verification, potentially interface back to the client for difficult questions etc.
  3. The presentation software that supports the “front end” and enables users to look at different views of the data.
  4. Any further external data sources that might feed into the analytics e.g. external company data, news reports. (This is still fairly unusual but is starting to happen).
  5. The “service” around interpreting the data, suggesting actions to the user, etc.

You could see a situation whereby a consulting firm might not have their own “engine” or indeed presentation front end, but might major on point 5; they would be offering a managed service in effect for analytics – and as we say, nothing wrong with that. Indeed, analytics programmes often fail because the users get the data and then have no clue or no time to work out what to do with it! That service wrapper might ensure that real actions follow the analysis.

But as we say, it is good to know how the service is being delivered. If your provider is dependent on others, that is worth knowing. Another interesting angle on this came up in a discussion the other day. A customer changed supplier and found that the previous supplier refused to “release” all the customer’s spend data that they held! That provision needs to be covered off in your contract, obviously, and that also means being clear in terms of which organisations might actually be working with and holding your data through the process.

Transparency is the word really. Just bear that in mind, and you should avoid too many problems.

Discuss this:

Your email address will not be published. Required fields are marked *