Like Peas & Carrots: Data Cleansing & Product Standardization

In supply chain management (SCM), unreliable, inconsistent and out of date “dirty data” is the bane of the profession. With healthcare spending approaching 20% of the U.S. GNP and medical supplies accounting for more than 40% of that number, the fact that the industry is only now moving toward a Uniform Device Identification (UDI) system is mind blowing. We’re almost glib in our recognition that healthcare has lagged behind other industries in SCM and procurement sophistication, yet the FDA just announced its guidance on UDI adoption less than 30 days ago?

Most industry sources suggest that “dirty data” and its ripple effects cost the industry as much as $11 billion each year –and we’re just talking about supply chain waste (i.e. medical products and services). In fact, dirty patient data (Master Patient Indices) are yet another long-standing problem that isn’t counted in that estimate.

Taking “dirty data” and turning it into “cleansed data” is a service that most all of the GPOs now offer. And despite all the talk of automation –everyone who offers these services claims to have a “secret sauce”—the last mile of the process is the most important and it remains very labor intensive. But there is an obvious reason for that: Most providers are hell bent on re-building and documenting actual histories when the opportunity to do a little revisionist thinking is staring them right in the face.

Supply chain data can be cleansed incrementally –category by category. Instead of digging through the archives and trying to find, organize and digitize a history that most SCM teams would prefer to forget, these projects should be executed in the context of product standardization initiatives. In other words, instead of “getting the data right” on 174 different types of orthopedic drill bits from 6 manufacturers, why not make the decision to standardize on 1-2 product lines (if possible) and collaborate with the selected manufacturer(s) to help you get your demand pegged and your item master current and accurate? Not a bad thought.

“By utilizing a combination of clean source data and advanced analytics, savings opportunities can be cultivated and maximized,” said Tim Vandermolen, product manager, information and data services, Novation. “Industry research has shown that through the use of data cleansing and advanced spend analytics, health care organizations can realize savings of 0.5 to 1.5 percent of their annual supply chain spend. These savings can make a big difference for a hospital’s bottom line.” Naturally, I fully agree, except that I believe the savings opportunities are even greater.

The point of this post is to encourage providers who are overwhelmed with the dirty data problem to think about it a little differently. I know from personal experience how daunting and expensive the problem can be –from a provider’s perspective and from the perspective of a company that sold data cleansing services. Frankly, these projects can be torturous.

As I was reminded this weekend: Combining data cleansing with standardization initiatives just makes sense. When I’m cleaning out my garage, I throw a lot of things away –that’s the whole point.

—Tom Finn


  • Mark Lainchbury:

    Sorry Tom

    The garage cleaning analogy just does not work.

    How quickly would you garage cleaned, if every time you found two similar spanners, you stopped & formed a work group to decided which one to throw away.

    Far better to get your kids to sort your whole garage into some buckets first, then step in and make some bigger decisions (Do you still need the whitworth spanners now you no longer have that old English sports car etc)

    • Tom:


      The analogy was fresh, because as I said, I just got through cleaning out my garage. I was aware of its limits and debated using it, but I opted on the side of provoking at least one point it makes that I think is valid. Too much unnecessary and expensive data prep work is being done and/or the lack of “cleansed data” is too often used as an excuse to not do anything. I have seen it here in the US and I spent a year in the UK on a NHS project where frankly, it may have been the biggest impediment to logical progress. Procurement professionals can be pack rats. When I’m cleaning out my garage, and encounter a bunch of junk in the corner (stuff I just couldn’t bring myself to throw out the last time), I just need to “get over it” and let it go. If someone showed up and said, here’s a check for your junk and we’re going to provide you the perfect garage set-up based on your cars, garden, etc., and we’re going to get rid of these fire hazards and leave you fresh cans of paint that actually match your interior and exterior colors…. something like that Mark. Benchmarks are underutilized.

      Procurement professionals could perhaps get more comfortable recognizing that numbers of beds, patient load, patient mix, service lines, seasonality, etc., are also valid indicators of demand in numerous categories of spend; that we don’t always need to bury analysts in the data archives to figure it out. And that when faced with an opportunity to standardize while we’re figuring it out, that a little back-solving can go a long way…

      Thanks for your comment Mark.

Join the Conversation

* Required fields  [email address will not be published]