Gorilla Data Quality

My previous blog post was titled “Guerrilla Data Quality”. In that post – and the excellent comments – we came around that while we should have a 100% vision for data (or rather information) quality most actual (and realistic) activity is minor steps compromising on:

  • Business unit versus enterprise wide scope
  • Single purpose versus multiple purpose capabilities
  • Reactive versus proactive approach

gorillaI think the reason why it is so is the widely used metaphor saying “Pick the low-hanging fruit first”. Such a metaphor is appealing to mankind since it relates to core activities made by our ancestors when gathering food – and still practiced by our cousins the gorillas.

Steve Sarsfield explained the logic of picking low hanging fruits in his blog post Data Quality Project Selection by presenting the Project Selection Quadrant.

So what we are looking for now is the missing link between Gorilla / Guerrilla Data Quality and the teaching in available literature on how to get data (information) quality right.

Bookmark and Share

2 thoughts on “Gorilla Data Quality

  1. Steve Sarsfield 24th October 2009 / 23:51

    I think you have to invoke the Infinite Monkey Theorem for data quality on this one. It states that a monkey hitting keys at random on a typewriter keyboard for an infinite amount of time (no insult intended for call center workers) will eventually give you good data quality.
    http://en.wikipedia.org/wiki/Infinite_monkey_theorem
    Until then, we can continue to work with the hearts and minds of people, the business processes and tools to help us get the data right.
    Thanks for mentioning my blog, Henrik.

  2. Henrik Liliendahl Sørensen 28th October 2009 / 08:34

    Thanks Steve.

    In the LinkedIn Data Matching group Dean Groves says:

    If you must be a guerilla, then you’ll have no overt support from the establishment; rather, you’ll likely get pushback for anything you propose publicly. You’ll need to plan & conduct covert operations, venturing into the heart of darkness to locate and correct data quality defects at their source, e.g. points of data entry guarded by the minions of those who would preserve the status quo. (“My staff are out straight right now; we really can’t spare anyone to help you.”) Perhaps aided by a sympathetic insider, you’ll have to make changes that won’t show on the local radar screen but will yield measurable improvements in data quality. You’ll need to make a quick exit without arousing any suspicion, and bide your time until the DQ improvements are a matter of record.

    If, on the other hand, you do have backing from above, you may be able to be a gorilla. You’ll be able to look around for low-hanging fruit (the easiest, fastest changes that have least impact and will yield most convincing results) and do your work in the open. You’ll be able to throw your weight around: to push back when you encounter resistance from the local authorities, and implement feedback loops that cycle defects back to their source.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s