The acronym “GIGO,” (standing for “Garbage In, Garbage Out”) is a well-worn computer science principle. Simply put, the best system in the world will produce nothing but garbage if you only feed garbage input into it in the first place. If you want meaningful results, you need to ensure your base data, itself, is capable of producing meaning.

We Procurement pros should get comfortable with this principle as well – it applies to us every time we perform a spend analysis. As we go about identifying key strategic sourcing projects, we rely on a foundation of spend data to inform us on the categories to touch and the strategies to enact.

If we allow garbage data in, the garbage we get out will result in lost opportunity costs down the road. Yet, as simple as this concept is, we routinely get it wrong. Pick your data source of choice; I’d bet good money that I could find at least a handful of mucked up spend data in desperate need of cleansing.

The Danger of Dirty Data

We’re in the second half of December as I write this, which is prime time for New Year’s resolution writing. I’ll forego the traditional weigh loss resolution, as valiant as it is.
Moving into 2018, I propose we all make a New Year’s resolution – we get our data cleansed and in-shape so that our spend analyses can do their part in supporting strategic sourcing initiatives!
In case you need some reasons to do this, here are three examples of how easy it can be to allow dirty data to slip into your analysis:
  • Naming Convention Differences. If your organization has multiple offices, departments, or business units that control their own spend, there’s a good chance they all have a unique name for their vendors – including the vendors they have in common. Consider how many ways AT&T can be written (“AT&T,” “ATT,” “AT and T,” and the list goes on) – no one naming convention is logically more sound than another.  Allowing alternative spellings to run rampant could lead to losing out on consolidation opportunities or, worse, may lead to ignoring a large aggregated spend that was cut into a handful of shrimpy spend amounts.
  • Doubled-up spend. On the other end, there’s also a chance you may inadvertently conflate spend as well. Multiple cross-source records may really be describing the same charge. For example, it isn’t uncommon to collect reports from their Concur systems or Amex accounts to incorporate T&E spend into an analysis. However, some companies will double down by summing up these reports as a single line item in a separate report elsewhere, simply marked “Concur,” or “American Express.” It is important to understand how different data sources work together to avoid this issue.
  • Categorization differences. Even if our data is clean, free of duplicates, and confirmed to not be doubled up, we can still run into issues. Take the supplier Airgas as an example. This is a common enough supplier or gas-related MRO spend. It is also a common safety supplies vendor as well. It is critical to ensure we understand which of these two types of products we are purchasing, or the mix if we’re purchasing both so two different sourcing initiatives don’t earmark the same spend for their market baskets.
These are all common occurrences. How pervasive these problems can become depends on a few factors – the number of data sources, amount of siloing of spend ownership, and the absence or presence of a strong data governance policy all play a part. Throw in some mergers and acquisitions, and you’re guaranteed to run into headaches.

Time to Clean House

For all you Procurement Pros who plan on taking Christmas week off, I vote you spend a little bit of that time thinking through how we might all go about ensuring we start 2018 on the right foot.

So how do we do it?

If you need a break from ripping through gift wrapping and drinking eggnog, check back next week for quick hit list on steps you can take to clean up your data in the new year.
Share To:

Brian Seipel

Post A Comment:

0 comments so far,add yours