The Quest for the Golden Copy (Part 1)

The Quest for the Golden Copy (Part 1)

May 05, 2010 by in Customer Data Integration, Data Governance, Data Quality, Master Data Management

Yesterday in his excellent and highly recommended blog post, the veritable Homer of both Data Quality and MDM, known to us mere mortals as David Loshin, explained The Myth of the Golden Record.

Today, I will continue this theme by beginning a four-part story on a somewhat related topic.  I will require your assistance in order to tell this tale properly, but more about that in just a few moments.

 

The Quest for the Golden Copy

Our story begins a long time ago (like sometime last century), where in the strange corporate land known as Customer Incognita, an ambitious man named Jason was attempting to usurp the throne of the beautiful and cunning CIO named Veritas.

Fearing the loss of her high station, Veritas decides to appoint Jason as the leader of the bold (but doomed) Project Argo, which has the business objective of resolving the organization’s long standing fatal flaw—namely, its inability to achieve a single view of its own customers—by creating the “golden copy” of each unique customer. 

Veritas was certain that not only was attempting to achieve this goal fraught with vast complexities, but it would also prove to be both a fool’s errand and Jason’s ruin. 

For after many decades of impressive business growth, including numerous mergers and acquisitions, Veritas had seen the organization become quite literally littered with innumerable data silos and embittered by ruthless political fiefdoms. 

The decaying ruins of previous projects still uselessly executed on production servers, incurring hardware and software maintenance costs from dozens of different vendors.  Most of the overburdened IT staff guarded these relics as if the justification of their continued employment depended on it—since in many cases, this was very true. 

Most of the overwhelmed business units maintained their own private data, fiercely resisted collaboration with any other business units, thus leaving each one of them to persist by relying on their own self-serving version of the truth.

Veritas knew well that throughout the organization’s discordant offices and hallways was echoing countless conflicting definitions of the term “customer.” 

Vertias knew this meant that within the organization’s dark and dysfunctional labyrinth, roamed countless digital clones of the same unique customer, all distorted and twisted into a mangled mash-up of meaningless information, upon which critical daily business decisions were being made—decisions based on little more than blind luck.

Veritas knew that Jason’s pride and arrogance would never allow him to turn down the prestigious appointment to lead Project Argo. 

Surely, the project would fail, Jason would become burdened with most of the blame, and then Veritas would remain the ridiculously overcompensated CIO now left to to rule unchallenged over the organization’s vast enterprise information wasteland.

Just as Veritas expected, Jason accepts the challenging appointment and so begins:

The Quest for the Golden Copy

 

Social Mythology

Long before either written language or reality television, it was the listening to and retelling of stories that served as the means of both education and entertainment.

Some stories “recorded” the history of actual events, while other stories were myths intended as an allegory to either convey meaningful ideas or provide a cautionary tale.

Stories were a social activity where groups of people gathered, not only to listen, but also to contribute.  Therefore, the story was very much a communal experience. 

Although today we conveniently think of Homer as an individual poet who wrote the epics the Iliad and the Odyssey, most scholars believe that these classic myths were the result of a community collaboration of multiple storytellers across centuries.

This Social Mythology was essentially the “social media” of classical antiquity.

By now you are probably thinking that I just went off on my most mythic tangent ever. 

No, not exactly—because this time I actually do have a point . . .

 

You have also been appointed to Project Argo

In an attempted revival of Social Mythology, I would like your storytelling assistance. 

As indicated above, this blog post begins a four-part story.  Over the next three weeks, I would like us to write this story together—as well as collectively decide if the story will provide a cautionary tale, or convey some recommended best practices.

Next week in Part 2 of our story, we will meet the Argonauts—the project team that Jason must assemble to help him complete The Quest for the Golden Copy. 

For Project Argo to be successful, what roles and responsibilities need to be added to our team?  And what roles should technology and methodology play in our story?

Please contribute your story ideas by posting a comment below.

Tags: , , ,

8 Responses to “The Quest for the Golden Copy (Part 1)”

  1. Phil Simon

    May 05, 2010

    Depending on how much pain you want to inflict on Jason, we could have multiple countries, currencies, and languages. How about a massive dose of internal politics?

    Reply to this comment

  2. Henrik Liliendahl Sørensen

    May 05, 2010

    Remarkable work (and only started) from you here Jim. And very appealing for someone like me who has two obsessions: History and Data Quality.

    What I often have witnessed is that data quality improvement projects (or programs) need a super hero to take the lead on the journey. And if this Jason can have a team of 40 -50 super heroes, well he might reach Georgia (no, not that Georgia, the other Georgia) and collect the golden thing.

    For now, I will stay on listeners side and wait for the next episode. Next Wednesday prime time?

    Reply to this comment

  3. Natasha Gabriel

    May 05, 2010

    This should shape up to be a fascinating and enlightening tale…

    @Phil you mention inflicting pain on Jason, and we may chuckle at that, yet that’s our reality today. Our business environs are so transparent that all those factors do indeed play a part. If we’re inflicting pain we may as well add in external repositories and subscriptions.

    Technology should automate, streamline and empower Jason. The team should have a “Golden Copy” champion as well.

    When we think of “Golden Copy” relevance should play a role in the methodology. Depending on who is looking for the “Golden Copy” different factors make it “golden” so the methodology and technology should facilitate this.

    Reply to this comment

  4. Jim Harris

    May 05, 2010

    Thanks Phil and Henrik for the comments, your feedback is always appreciated.

    @Phil — Both international data and internal politics will most definitely be in Jason’s future. After all, I don’t remembering reading about any easy “quests” — since those usually make for boring stories :-)

    @Henrik — Many of the Argonauts from the classic myth were heroes in their own right, such as Heracles to name one, so Jason will need to recruit as many other heroes as possible. I am not sure if either Georgia is where they will find their “golden copy” (if it even exists).

    As for the schedule, the story will run every in prime time for the rest of the month — meaning, every Wednesday in May at 12:00:00 UTC.

    Reply to this comment

  5. David Loshin

    May 05, 2010

    D’oh!

    Reply to this comment

  6. Jim Harris

    May 05, 2010

    @Natasha — Thanks for providing several great ideas with your comment. External data in some form will definitely have a role to play and I really like the idea of a “Golden Copy” champion. Relevance (i.e., point of view regarding the definition of “golden”) will be crucial to the methodology, without which it would be like encouraging the business stakeholders to argue over it as if it were the Golden Apple — which, for the those unfamiliar with the myth, that’s how a little thing called the Trojan War got started. :-)

    @David — Just to avoid any possible confusion, between the two of us, you are definitely more like the legendary poet Homer of Greek Mythology, and I am definitely more like the doughnut eating buffoon Homer Jay Simpson of The Simpsons. :-)

    Reply to this comment

  7. Ken O'Connor

    May 06, 2010

    Hi Jim,

    Great idea… here’s my tuppence worth (sorry two euro cent, since Euro Changeover).

    For some reason, I’m not a big fan of “role titles” (perhaps too difficult for me, or too open to misinterpretation).

    I tend to think in terms of what needs to be done:

    Begin with the end in mind…
    Big picture stuff first… then break it down…

    - Define what data is required (the “ideal” Golden Record)
    (So easy to write the above in a single sentence – not easy to deliver !)

    - Define WHY it is required – to satisfy what business requirement(s)… there should be many.

    - Define the quality requirements of the required data.
    All fields may not need to be perfect (never will be anyway) – what is the “Target data quality” to make the Golden Record useful to the business, and worth the effort.

    - Identify the source(s) of the required data.
    In this case, they are many, scattered worldwide, in multiple languages, on multiple platforms.

    - Specify the data quality metrics to be captured.

    - Research the business rules associated with the source data (defining what should be in the data – in thoery).
    This is one of the biggest challenges in my experience).

    - Measure the quality of the available source data.

    - Understand the implications of the quality of available source data to satisfy the target data requirements.

    - If necessary (and it will be), and if feasible (to be determined), implement data quality improvement measures to raise the quality to the required level.

    - Worst case – if the facts tell you data quality is too low and cannot be improved – call it as it is, cancel the project and save yourself a ton of money!

    - Build in real-time data quality monitoring and alerts to enforce the data quality rules at all points of data entry

    Looking forward to learning how Jason gets on !

    Rgds Ken

    Reply to this comment

  8. Jim Harris

    May 06, 2010

    Thanks for the detailed comment Ken!

    I really appreciate you taking the time to provide all of that excellent feedback. I too am not a big fan of “role titles” and like to think in terms of specific tasks without simply resorting to saying well, we will need a DBA, an analyst, a programmer, a SME, etc.

    When you think in terms of titles and not tasks, it can be easy to build a team that, at least on paper, seems like they can get the work done.

    I am looking forward to incorporating your feedback into the storytelling process.

    Best Regards,

    Jim

    Reply to this comment

Leave a Reply