How to Build the Taxonomy You Need to Get the Spend Insights You Want
It’s easy to be seduced by Spend Analytics and the promise of a complete and accurate view of your spend. The data! The insights! The information-based decisions that will follow – and the savings that will flow! Make no mistake: these are all legit promises. But achieving them requires building a rock-solid foundation for your Spend Analytics machine. And that means starting off with the right taxonomy.
For tips on how to make taxonomies less taxing, I sat down with our own taxonomy wizard Katie Key. Katie shared sound advice for anyone who wants to create a sound taxonomy.
Why are taxonomies so important?
A taxonomy is the categorical hierarchy of spend and sourcing groups, from general to specific. At JAGGAER, we believe a custom taxonomy is vital to our clients truly understanding their spend. Our customers build the categories they want and need to see, in order to make decisions about their organization’s spend. We can implement up to six levels of taxonomy if our clients need it.
To understand why taxonomies are so important, it helps to go back to the three main phases of spend analysis:
- Data Acquisition – gathering all of an organization’s spend data from whatever sources contain it.
- Data Processing – cleansing, classifying and enriching the data.
- Reporting & Analysis – extracting the data and insights that allow you to make decisions about your spend.
Taxonomies are critical in the Data Processing step, when line items of spend are categorized into the custom taxonomy. Without a strong taxonomy, classification becomes weak, and the resulting reporting and analysis won’t offer much insight – if any – into spend.
How should you go about building a taxonomy?
At a high level, building a taxonomy means setting up the categories and sub-categories that you’ll use to classify your spend data. For example, you might have a category of “Business Services,” with a sub-category of “Marketing.” Within the Marketing sub-category, you could have additional sub-categories for “Advertising,” “Creative” and “Market Research.” And those sub-categories might break down into even more detail – like print and online advertising for example. Taxonomies can vary greatly by industry, and it’s important that your taxonomy reflects the goods and services relevant to your organization.
As you are building your taxonomy, consider these best practices, that have worked time and again for our clients:
- Set objectives. What do you want to get out of your reporting? What answers are you searching for? This will guide how your taxonomy is shaped.
- Align with existing internal reports. The way spend is categorized within those reports is often a good starting point for your taxonomy. If there is a classification that is particularly valuable – like a report that you turn to time and again – be sure to reflect that in your taxonomy. You might also recognize structures that are working poorly. You can correct or abandon those in your new taxonomy.
- Keep upcoming initiatives in mind. Think not only about your organization’s current needs, but about its potential future needs. Is it likely that you would need to open a new office, or remodel an older office? Then build in categories for office equipment and furniture, even if you haven’t spent much there in the past.
- Find the sweet spot for detail. Sometimes we see data with very little detail. You can’t classify those line items to meaningful sub-categories if you don’t truly know what that spend was for. And inversely, if you have a lot of spend data detail – do you really need to be that granular? Are you really going to source office chairs separately from office desks, or will you source office furniture as a whole?
Are there pitfalls people should avoid?
Definitely. In a nutshell, I’d say the biggest pitfalls are either over-thinking or under-thinking the taxonomy. Here is what I mean:
- Not having enough levels means that too much spend will be classified into general categories – and won’t provide any insight above what your general ledger already provides.
- But if you have too many categories, you won’t be able to make sense of your reporting. Think about your desktop – if you have 100 folders, you have to take time to look through them to find what you want. Focus on the categories that matter. We recommend 10 (or less) top-level categories – things like Business Services, IT, Facilities – so that you can get the big picture of your spend, and then broken down from there as needed for the goods and services your organization needs.
- Categories that are too granular are problematic also – you won’t be able to see the big picture as easily, and might as well be looking at spend line item by line item.
- Avoid creating redundant categories. For example: Business Services > Information Technology Consultation, and another category for IT > Business Services Consultation. This is probably the same kind of spend, but if it’s broken into these redundant categories, it will be difficult to get a clear picture of how much you’re spending on consulting for IT.
- Unused categories might not necessarily create a problem for your reporting, but can look like a red flag to decision makers, or worse, look like a false budgeting success.
- And finally, if you have too many “other” or “miscellaneous” categories in your taxonomy, spend will be lumped there, and again, you won’t get a clear picture of what you’re spending on. If you have the data details, your data should be classified to an appropriate category.
How will our spend analysis output change once we have this taxonomy in place?
Simply put: you’ll have reporting that makes sense and that can drive strategic budgeting and operating decisions within your organization. It’s the holy grail of spend analytics, and it’s impossible to achieve without a solid taxonomy.