If actionable business insight, the stuff of BI legend, is the Mt. Everest summit, know this: There's little point to building or buying a better ice pick if the climbing team has no arms.
When I think about the business of technology, I'm reminded of a joke by the late, great comedian Mitch Hedberg. "I didn't go to college," he used to say, "but if I did, I would've taken all my tests at a restaurant. Because the customer is always right."
Although most technologists are loath to admit it, we're in the restaurant business. And we're doing a disservice to our clients by starting from the position that the customer is always right.
To put it bluntly, business clients aren't always right. Not even close. Take, for example, business intelligence, what I consider to be the most ironically named sector of software.
Potential customers of BI are rarely as data-literate or software-savvy as their increasingly complex problems demand them to be. Few if any aspire to a data-driven culture, and fewer still have a practical roadmap to it or hold technology's feet to that fire. So the idea that customers shopping for BI know what they need -- or have a vision of what's possible -- is overstated.
It's our job to make sure that tech investments are more than expensive flirtations with vanity. To that end, here are five seemingly intelligent, "always right" customer requests around BI that IT organizations must challenge.
1. Let's deliver what's in the RFP/BRD.
It was, after all, put together with everyone's input. And what could possibly go wrong when that much consensus is involved, right?
Before you read your next RFP or BRD, do yourself a favor and queue up Richard Strauss's Thus Spoke Zarathustra.. If you're a devout nerdling, you'll recognize it as the ape-playing-skeletal-drums song in "2001: A Space Odyssey." That Neanderthal's gleeful use of the giant bone as a tool represents the moment we became human. And it's entirely appropriate to have that music playing in the background as you read your next RFP or BRD because while we as a species outgrew the body hair, we never outgrew the bone.
What time still hasn't taught us is that while tools are meant to enhance human capabilities, they usually have to build on something. And the more sophisticated the tool -- i.e., software that automates a complex function or process -- the more likely that there needs to be some substantive skill present before the tool.
In other words, if actionable business insight, the stuff of BI legend, is the Mt. Everest summit, then even the best climbing boots aren't going to tone your organizational calves or increase your lung capacity. There's little point to building or buying a better ice pick if the climbing team has no arms.
Oh, and I challenge you to find an RFP or BRD that isn't focused on boots and picks.
Our hard-wired tool-centricity needs a syringe to the neck. And the dialog between tech and business, the one captured in RFPs and BRDs, needs to focus exclusively on building competencies and skills to populate and consume BI's dazzling dashboards.
Speaking of which …
2. Let's focus on making our dashboards awesome.
The point here is a simple one: The value of BI is the investment you're making in data. Those pretty dashboards aren't even "the last mile." More like the last inch. And everything that comes before them is more important. Your data mudroom is more important. Your ETLs, metadata, taxonomies, exception reports. Everything.
Which makes it difficult when I hear …
3. Let's work with the data we have.
A quick data science joke. A woman walks into a cavernous, pitch-black room. The only light in the place is a tiny lamp in the corner stuck to a wall, shining on the floor. A man is huddled on the floor beneath the lamp, face close to the ground, looking for something. The woman asks him what he's looking for and he replies: "business insight."
The woman gets down on her knees and starts searching too. After awhile, she looks up and asks: "Did you lose it here?" And he replies: "No, but there's light shining here."
That man is your typical BI customer. And that light is shining on the data that he thinks he has.
Your goal should be to either light up the entire room or save yourself the expense of a BI implementation. Because in my experience, it's never that subset of "lit" data that's interesting. It's the relationship between that data and the rest of your unlit data that holds real business value.
For instance, getting all of your sales contacts into a CRM system is mildly useful. But the real value of that data can shine through when you delve into its relationships with your phone, email and travel logs, letting you answer questions such as: Are sales commissions really being earned?
By limiting themselves to "the data we have," most BI dashboards are aspirationally challenged. They barely answer the whats and leave the more interesting whys to the imagination. The unforgivable part is that they give your org the false comfort that you're making data-driven decisions when you're still operating on gut.
InformationWeek Must Reads Oct. 21, 2014InformationWeek's new Must Reads is a compendium of our best recent coverage of digital strategy. Learn why you should learn to embrace DevOps, how to avoid roadblocks for digital projects, what the five steps to API management are, and more.