Data Professional Introspective: Do This to Be the Toast of the Town

COL04x - feature image - already 300x300You are a data management professional, so we know you are a hero already.   But if you’d like to be crowned the Prom King or Queen, lavished with praise and recognition (perhaps bonuses, free vegan lunches, a trip to Hawaii, etc.), you need to ‘Show them the Money.’ That is, demonstrate either how your organization can raise revenue, or how it can lower costs.

Data management projects or programs can be hard to sell internally, as most organizations lack a strategic view of their data assets, and therefore tend to not value the data highly.  Indirect benefits, including faster time to market insights, improved data sharing, or shorter development cycles, just don’t kindle that gotta-do-it-now fire.

However, if your leadership were to believe that they can productize and sell the data, their attention would focus most satisfactorily.  For instance, an online retailer might explore selling customer multi-year purchase histories to a non-competing company.  Executives suddenly get serious, asking questions like: “how do we document the data for this product, who is our market, how do we assure quality data for our customers?” and of course, “can you get this out the door by the end of the month?” If the value proposition is there, your organization will enthusiastically do what is necessary.

Although data management professionals eventually become good at convincing business leaders about the value of efforts such as formalizing a Business Glossary for the Patient Master Index, or developing a metadata strategy and schedule for the repository, we’re still often faced with the “Next Quarter’s Results” attitude (or a similar response “I can buy a new self-service BI platform for that”).  What sounds flashier to the Board: shiny new technology or multi-year achievements that pay long-term dividends?  You know the answer.

Let’s turn to the other side of the equation, Net Profit (aka Net Income, Net Earnings, or the bottom line).  Net Profit is calculated by subtracting the organization’s total expenses from the total revenue for a given time period, typically annually.  The operative words for today’s column are ‘subtracting expenses,’ that is, lowering costs.

Most of the data management efforts that we advocate indirectly affect revenue or save costs over time. As stated above, these indirect benefits, though real, don’t blow the Board’s hair back.  But what if you were to show them that, for example:

  • They’re paying too much for purchased data – in some organizations, buying the same data twice
  • Their contracts have ‘gotcha’ clauses that impose steep termination fees or other undesirable conditions
  • Their service level agreements don’t specify metrics or controls
  • They don’t charge penalties for late, missing, or incomplete records
  • Procurement doesn’t reach out to the business units to solicit data quality rules or detailed requirements
  • The business units, in turn, often don’t partner well with procurement
  • The business unit that originally purchased the data isn’t using it any longer, but because no one is responsible for monitoring usage or canceling the service, the money keeps going out the door.

These issues directly affect costs.  If you can demonstrate that your proposed project, targeting Data Provider Management[1], will save costs, the executives are going to be very interested. After all, they’ve heard complaints about acquired data before, and they probably have some idea what they’re spending for it. [Full disclosure, there are even greater costs internally – duplicate interfaces, same data sets from conflicting sources, no authoritative data sources, ad hoc interfaces, etc. – but they’re hidden, so we’ll leave that for Phase 2 of your project.]  In addition, focusing internally impacts more people within your organization, whereas the external providers are de facto mercenaries, and your organization doesn’t have to care about them.

Why is it challenging?  Well, usually Procurement doesn’t have a well-designed relational database handy that can easily provide the answer to a question such as: “List the data feeds that contain customer demographic data, sorted by data owner, contract start date, renewal date, and an annual fee over $25,000.” I have yet to see a procurement department that has detailed knowledge about everything they negotiate, purchase, and put under contract.  Also, Procurement’s attorneys are often overworked and don’t have time to read the fine print.

Here’s a true story.  A financial firm purchased credit risk data from a leading data provider.  The business unit which originally requested the data determined that they no longer needed it, after several years of use and provisioning the data to other business units. When Procurement informed the data provider that they did not intend to renew the contract, the vendor informed them that all data received since contract initiation needed to be purged from all data stores. (!)

And another true story.  A Federal agency had purchased stock market data for decades. During an effort to streamline their procurement process, it was discovered that one expensive data feed, with an enterprise license, had been purchased three times by three different business units.  They had been paying triple for years, and the vendor didn’t bring this fact to their attention (caveat emptor).

And a success.  When still another organization finally decided to address its purchased data, it was able to eliminate feeds that weren’t needed, consolidate licenses, improve performance through service level agreements, and save $2.3 million per year.

Now that’s worth the effort, right? Low risk, high reward. How can you pitch this to your executives as a valuable project for you to lead?

It is recommended first to arm yourself with some juicy complaints or stories from business units purchasing data, highlighting data quality issues, refresh timeliness, sharp price increases, provisioning after a data gap, etc.

You can then flesh out the following steps in a project proposal to fully inform the organization what shape its purchased data is currently in, discover cost saving and data quality improvement opportunities, and standardize an approach for future purchases.

  1. Get the support of Procurement. This shouldn’t be too difficult, because they would probably like some help, and they will also be rewarded if they lower costs.
  2. Convene a working group consisting of data owners who purchase data and Procurement representatives.
  3. Request that the owners survey their business unit and describe what data they are purchasing, who uses it, and for what [Note: your team will need to follow up and offer to help them].
  4. Work with Procurement to develop a list of purchased data and current pricing (this may require some manual research).
  5. Create a spreadsheet / simple database, sort and query to identify duplicate purchases, data with few users, and feeds with overlapping content, etc.
  6. Interim Results – Present any low-hanging fruit at this point – e.g., “Hey, this feed is $30,000 per year and the user retired.”
  7. Work with Procurement to explore if data contracts and service level agreements include notification of data changes, escalation clauses, rectification of data gaps, etc. – and identify any discrepancies.
  8. Work with the business units to explore their level of satisfaction with the purchased data, if they can suggest data quality requirements, or performance metrics that would make the data better or more usable.
  9. Find a well-crafted service level agreement and make it into a template with instructions for business units.
  10. Look for opportunities to consolidate services; for instance, if your organization purchases multiple data products from the same provider, work with Procurement to negotiate a volume discount.
  11. Present your results and deliverables jointly with Procurement, with actionable recommendations.
  12. Make sure that one recommendation is to create a simple asset library, allowing business users to know what data is available – this alone will save both time and costs.

Voila!  You have accomplished God’s Work, helped streamline data purchasing, and undoubtedly have discovered cost savings.  You’ve increased Net Profit and your team’s bronze statue is about to be commissioned. Congratulations!


[1] This is the name of a Data Management Maturity Model process area in the Data Operations category.

Share

submit to reddit

About Melanie Mecca

Melanie Mecca, Director of Data Management Products and Services, CMMI Institute, led development of the Data Management Maturity (DMM) SM Model. Her team created a highly interactive method for assessing an organization’s capabilities against the DMM, and she has led numerous assessments for organizations in the financial, Federal, and technology industries. She directed creation of the Building EDM Capabilities, Mastering EDM Capabilities, and Enterprise Data Management Expert (EDME) courses leading to DMM certification. In 30+ years solving enterprise data challenges, Ms. Mecca has architected and implemented data management programs and projects, data strategies and architectures, and designed enterprise data services. She is an active presenter of classes, seminars, webinars and case studies, and is a strong advocate for data management education, with a passion for assisting organizations to realize business value from their data management programs.

Top