You only get a Return from something you actually Invest in
Jim Harris in Books, Data Quality, Debates, Social Media tagged Best of 2011, Business Benefits, Data Governance, ROI Thursday, December 8, 2011 at 3:00AM In my previous post, I took a slightly controversial stance on a popular three-word phrase — Root Cause Analysis. In this post, it’s another popular three-word phrase — Return on Investment (most commonly abbreviated as the acronym ROI).
What is the ROI of purchasing a data quality tool or launching a data governance program?
Zero. Zip. Zilch. Intet. Ingenting. Rien. Nada. Nothing. Nichts. Niets. Null. Niente. Bupkis.
There is No Such Thing as the ROI of purchasing a data quality tool or launching a data governance program.
Before you hire “The Butcher” to eliminate me for being The Man Who Knew Too Little about ROI, please allow me to explain.
Returns only come from Investments
Although the reason that you likely purchased a data quality tool is because you have business-critical data quality problems, simply purchasing a tool is not an investment (unless you believe in Magic Beans) since the tool itself is not a solution.
You use tools to build, test, implement, and maintain solutions. For example, I spent several hundred dollars on new power tools last year for a home improvement project. However, I haven’t received any return on my home improvement investment for a simple reason — I still haven’t even taken most of the tools out of their packaging yet. In other words, I barely even started my home improvement project. It is precisely because I haven’t invested any time and effort that I haven’t seen any returns. And it certainly isn’t going to help me (although it would help Home Depot) if I believed buying even more new tools was the answer.
Although the reason that you likely launched a data governance program is because you have complex issues involving the intersection of data, business processes, technology, and people, simply launching a data governance program is not an investment (unless you believe in the Hedgehog’s framework) since it does not conjure the three most important letters.
Data is only an Asset if Data is a Currency
In his book UnMarketing, Scott Stratten discusses this within the context of the ROI of social media (a commonly misunderstood aspect of social media strategy), but his insight is just as applicable to any discussion of ROI. “Think of it this way: You wouldn’t open a business bank account and ask to withdraw $5,000 before depositing anything. The banker would think you are a loony.”
Yet, as Stratten explained, people do this all the time in social media by failing to build up what is known as social currency. “You’ve got to invest in something before withdrawing. Investing your social currency means giving your time, your knowledge, and your efforts to that channel before trying to withdraw monetary currency.”
The same logic applies perfectly to data quality and data governance, where we could say it’s the failure to build up what I will call data currency. You’ve got to invest in data before you could ever consider data an asset to your organization. Investing your data currency means giving your time, your knowledge, and your efforts to data quality and data governance before trying to withdraw monetary currency (i.e., before trying to calculate the ROI of a data quality tool or a data governance program).
If you actually want to get a return on your investment, then actually invest in your data. Invest in doing the hard daily work of continuously improving your data quality and putting into practice your data governance principles, policies, and procedures.
Data is only an asset if data is a currency. Invest in your data currency, and you will eventually get a return on your investment.
You only get a return from something you actually invest in.
Related Posts
Can Enterprise-Class Solutions Ever Deliver ROI?
Do you believe in Magic (Quadrants)?
Which came first, the Data Quality Tool or the Business Need?
What Data Quality Technology Wants
A Farscape Analogy for Data Quality
“Some is not a number and soon is not a time”
“What is is the was of what shall be”
Reader Comments (6)
Excellent issue. While it is very true that ROI on DQ requires investment in several different areas, I think you might have been a bit disingenuous when you say that investing in a DQ tool and starting a DG program won't create a positive return. (And I say this as a set of exercise weights still stands accusingly in the corner.) The DQ tool and the DG program decision could both be considered very positive signs that your executive is becoming aware of DQ and is willing to support it, and you certainly would not get very far with a data quality improvement initiative if your executive does not support it! I guess my quibble is that data quality is often couched in negative terms that could be off-putting, and possibly a more positive perspective might exclaim, "My Gawd, that is so great you finally have that DQ tool available or the DG program beginning to frame the DQ issue. Here is the best way to prove the value of ...".
I hear you saying that a DQ tool and DG program are insufficient to produce a return on investment, but do other people hear you say "Don't bother purchasing a DQ tool because it won't create a solve your DQ problem". To paraphrase that Russian writer who said something like "There are 24 reasons that a marriage succeeds, and it just takes one of them going awry to skew the marriage". The DQ tool and DG program are maybe just 2 out of 24, but they ARE 2 out of 24, which is better than the zero dozenth that many organizations start from.
I guess you are saying that the act of purchasing a car will not move me from New York to Los Angeles, but purchasing the car, committing to drive for 3-4 days, planning a good route, not drinking alcohol and committing to not stopping at every mall along the route will move me towards the sunshine.
And to follow up on your second point, every organization is running a system that creates the information product they operate from. And while it is often a sub-optimal information product, it is still producing a value, in that it is keeping the organization operating, however inefficiently. If I might be permitted another analogy; having a random pile of wood behind your cabin is valuable in that some of it can be used to keep your cabin warm this winter, but if it is piled randomly it won't season well, some will probably rot before it can be used and it will take up a lot more space than necessary. So it is sub-optimal, but it still has value, it will still keep your cabin warm and maybe support toasting marshmallows.
So the $64,000 question is: What will make me realize that I should plan, handle and store my wood better? Since the woodpile is sufficient for my needs right now, does an executive order have to come down to start me on the path to a better woodpile? The popular choice would be to just rest on my laurels.
Thanks for throwing some logs on the cozy fire with yet another thought-provoking comment, Gordon.
Although my perspective could be considered a negative spin, which, as you explained, could prevent executive management from purchasing data quality tools and launching data governance programs. And I certainly agree that it can indeed be a tremendously positive sign that a data quality tool and/or a data governance program has received enough of that essential, but rare element (aka Executive Unobtainium) of executive sponsorship, without which no lasting success could be possible with either data quality or data governance.
However, my underlying concern is the “Mission Accomplished!” sign prematurely adorning the Executive Board Room after the funding for purchasing a data quality tool and/or launching a data governance program has been approved.
Toasty Marshmallow Regards,
Jim
Totally agree Jim. One of the biggest breaks we had on one Data Warehouse (DW) development project with integrated data quality was when the budget fell through for the purchase of the Fancy/Dancy front end business intelligence (BI) tool, which totally quashed executive expectations and left us "free" for next 6 months to profile and conform the data sources and build our EQTL (i.e., ETL with integrated Qualification of data as it surged through the pipe to the DW).
The purchase of the BI tool up front would have been the proverbial "lipstick on a pig" and would have de-focused the team. After those 6 months of dedicated data quality (DQ) work, they put up a simple SharePoint to display the data initially and it got rave reviews, because the DQ was high.
And that healthcare organization now understands their data quality well enough that they won't be putting up a "Mission Accomplished" sign any time soon.
Hoping for S'More great blog entries from you soon.
Cheers,
Gordon
Jim,
What you are getting at is two different ways to measure an effort. The act of buying and implementing the tool (or Gordon getting the car) is referred to as installation. ROI is an installation measure. Meaning I invested $X and received $Y return.
The second way to measure is called realization. It takes much more time and is way harder to measure and get folks on board. Here's my example to throw in the mix.
When I bought my first iPod, the act of getting it implemented into my day and life didn't financially (ROI) make sense to do so. But, after getting to know what it can do, talking with others, and getting to the point where I couldn't be without it, I didn't think about the cost again. That's realization.
For data quality and governance, realization occurs when the users of the applications aren't thinking twice about the data. It's there, can't live without it.
However, along the way to realization, you can celebrate installation wins. For example, a good customer data governance program after being implemented, socialized, re-socialized, presented, socialized some more and so forth will finally get a win like a reduction in developer time for onboarding a CRM system. So you are correct, without the people behind it informing and making it happen, just setting up DQ/DG doesn’t provide anything.
Thanks for installing your insight-realizing comment, Kevin :-)
You raise an excellent point about the distinction between installation and realization, which reminds me of the distinction between potential energy and kinetic energy.
Data quality tools and data governance programs unquestionably possess potential energy (i.e., the energy stored within something), but requires doing the hard daily work to unleash their kinetic energy (i.e., the energy created by accelerating something from its state of rest) and especially to maintain their data quality inertia and data governance retrograde motion.
Best Regards,
Jim
From the LinkedIn Group for Data Governance & Data Quality, Yekaterina Sheremetyev commented:
“And the beauty of investing in data quality and governance is that unlike other assets, data assets are not depleted with use and continue to give back.”
And Joseph Cillo commented:
“The problem with showing ROI for Data Quality is that Data Assets are not typically accounted for like other assets. There usually is not a balance sheet or income statement transaction associated with increasing the value of Data Assets, so the ROI is usually indirect. A lot of hard work on data governance shows up as a lot of expense, and the benefits are only indirect. Finding hard numbers on ROI is a bit tricky. You have to find where better data has increased revenue or decreased expenses, and find a way to quantify it. It's really not likely that you will ever be able to say, 'We increased the value of our data assets by 20%'”
And Stijn Christiaens commented:
“A big driver for data governance is operational efficiency: improving on broken steps business processes. As such, it the ROI calculation will frequently depend on the specific step that you are improving. If the driver is regulatory, it might be more direct: not being able to bring a product to market (e.g., as in pharma), or paying large fines (e.g., finance).”