Gartner Blog Network

Does Information Utility suffer a Half Life?

by Andrew White  |  January 29, 2014  |  3 Comments

I heard a funny story on the way to the coffee machine this morning (actually yesterday).  Yet another end user organization said to me: “It was just 6 months after our project (ERP migration) that we discovered we had lost control of our data and it started to materially impact our ability to perform”.  This phrase, and many like it, is very common.  What isn’t as common is the 6 month number.  I get the feeling the average is actually 9 months, though there are a few around or even less than 6 months.  Some limp along, after 12 months, and continue towards oblivious without realizing how useful consistent information is to critical business performance.

So is this figure, 3, 6 or 9 months, a kind of “information utility half life”?  What factors would dictate the time it takes before something hits the fan?  Would a manufacture’s so called information usefulness half life differ from an insurance organization?  What about a firm that actually sells information – would that be even more dependent on information usefulness or utility?  What about the size and complexity of the application landscape – I would imagine this has a factor also.  I am sure there are others.  What do you think?

I had another idea some time ago related to information fitness.  Fitness is a term used very specifically in evolutionary theory and I am intrigued with the idea of applying the concept to information.  Too often the term “data quality” is thrown out there as a broad umbrella to mean a lot of different things.  Some others use the term fidelity, and so they “wrap” quality under it.  For those that use data quality, the term “fidelity” might be an element of it.  But fitness seems to be a good, new word to cover all known and even known unknown elements of quality and fidelity and overall persistence of value over time.  Some questions:

  • How long does value in information persist?
  • How can we account for its value if its own utility, usefullness, quality, or fitness falls? 
  • Does the value curve fall at the same rate as the fitness curve?
  • How does informatin fitness change, and how is it changed, once information assets are combined (think bills of material)

All good questions – right?  Even better to discuss over a beer – perhaps we can explore at our upcoming Enterprise Information and Master Data Management Summits:

Category: chief-digital-officer  information-as-an-asset  information-innovation-yield-curve  information-leadership  information-management  information-theory  information-value  master-data-management  mdm  mdm-summit  mdm-summit-na  

Andrew White
Research VP
8 years at Gartner
22 years IT industry

Andrew White is a Distinguished Analyst and VP. His roles include Chief of Research and Content Lead for Data and Analytics. His main research focus is data and analytics strategy, platforms, and governance. Read Full Bio

Thoughts on Does Information Utility suffer a Half Life?

  1. Dennis Moore says:

    Andrew –

    Great insight, as usual. As Ford Goodman often says, “data degrades.” However, not all data degrades at the same rate. For example, many people now keep the same personal e-mail address and personal phone number for their entire lives, but their work e-mail address and work phone number may change much more frequently. Last names of American men change less frequently than last names of European men, and both change less frequently than last names of American women. Home addresses change as well, as do the members of your household.

    Data may not have a strict utility=e^(time) half life function, and the utility may vary significantly by source. Your field sales force may know that an important customer has changed address long before your purchased third party data source is aware. Your purchased third party data source, however, may be aware much sooner of a negative change in a key customer’s (or supplier’s) credit rating than does your field sales force.

    Clearly, the utility of data declines with time, but that time factor is different for different data elements, and it varies by source. A more recent update in a less trusted source may be more trustworthy than a very old update in a normally trustworthy source.

    Master data management systems should be able to factor data element, data source, and time into an easily configured fitness “calculator” that produces the “best version of the truth” for the business.

    Thanks for another thought-provoking blog!

    – Dennis Moore, Informatica

  2. Andrew White says:

    Hi Dennis, thanks for the response and post.

    I wanted to call out an item you mentioned, “Clearly, the utility of data declines with time”. I suspect some analysts would believe that information utility surely does NOT decline with time. Or more precisely, some information utility may decline over time. Some information utility may actually increase with time. If I knew a secret (the winner of a horse race next week), the utility of that information should fluctuate but once the race has finished, the utility of that information drops to zero. It might even go negative if I bother to store and keep that data in short term memory – due to opportunity cost.

    So this whole discussion about “what value information” is a fun one to explore. It never is black and white I think.

    But your response is spot on – and I agree with your points, overall.

Comments are closed

Comments or opinions expressed on this blog are those of the individual contributors only, and do not necessarily represent the views of Gartner, Inc. or its management. Readers may copy and redistribute blog postings on other blogs, or otherwise for private, non-commercial or journalistic purposes, with attribution to Gartner. This content may not be used for any other purposes in any other formats or media. The content on this blog is provided on an "as-is" basis. Gartner shall not be liable for any damages whatsoever arising out of the content or use of this blog.