In the U.S. there’s a push for more opportunity and diversity in the tech industry (for good reason, judging from recent statistics). Diversity is an important social goal. Where I live in Oakland, California, good work is being done to foster inclusion in tech. But I see another, related problem: We need more diversity of thought to stop producing the same types of data for the same types of audiences. Here’s my evidence.
Data isn't enough. It seems that business intelligence technology, big data startups, and analytics are everywhere. Nothing wrong with people becoming more productive and making more evidence-based decisions. Of course, many technologies seek to replace people with algorithms: Nothing wrong with that either, in some cases (I explore this in my recent report, Data is Easy, Deciding is Hard.)
But while we're collecting data, why don't we do more for the human decision maker? Tech vendors are producing lots of impressive dashboard and visualization functionality, but not enough tools for synthesizing complex evidence, evaluating difficult situations, and overcoming our bad decision-making habits. Tech is producing too many nicely displayed facts without explanation: Lots of 'what' and not enough 'why'.
Data viz isn't enough. With more diverse thinking, we could build practical tools that visualize decisions, show causal mappings, and capture a whole story from numerous sources of evidence. Consider the new 8.2 release from Tableau, a very successful maker of data visualization tools. The company says it's "obsessed with data. Connecting to data, analyzing data, and communicating with data." Their new Story Point feature is nice. But, as you can see in their Austin Teacher Turnover example, the 'story' is long on facts and short on real story: We don't see the specifics of the Reach program, we don't know which Austin groups supported it and which ones didn't, or why it failed. And we don't see what decisions were made by school officials implementing the program, or which actions are connected to which outcomes. Rather than yet another data viz, why don't the smart, capable people at Tableau think differently and produce something more comprehensive and innovative?
BI getting bigger, not better. I'm not alone in questioning the value in some of the new data tools. Business intelligence usage is flat. A popular 2014 survey by TDWI reported a 6% decline in those finding significant impact, down to only 28%.
We're not connecting action to outcome. One of the best critiques / analyses I've seen is Accenture’s extensive study Analytics in Action: Breakthroughs and Barriers on the Journey to ROI. Their research shows that “most organizations measure too many things that don’t matter, and don’t put sufficient focus on those things that do, establishing a large set of metrics, but often lacking a causal mapping of the key drivers of their business.” Accenture underscores the “need to industrialize the insight-action-outcome sequence”. Highlighting the absence of tools designed for decision-making, they conclude that most companies “fail to embed analytical insights in key decision processes so that analytics capabilities are linked to business outcomes.”
Frank Bien of Looker tells the hard truth: “The common view of the past five years is that users are stupid and that data needs to be spoon-fed to them via pretty pictures…. It’s time to strike a new balance: to join ‘big data’ to business data in such a way that it serves the business – and doesn’t just grow a big data repository.”
What can be done? Hiring people with diverse experience, and engaging a diverse set of customers, is a good first step toward finding better problems to solve. Diversity of investment – in the public, private, and third sectors – is another needed step, and that's being recognized. Christopher Mims wrote recently that "The entire Bay Area appears to have given up on solving anything but its own problems: those afflicting the same 20-somethings who are building these startups." Of course they don't do this all by themselves: Venture capitalists are being accused of "focusing exclusively on the first-world segment of twentysomething yuppies".
Yes, we need more hiring diversity, but please don't take away the 20-somethings. As a startup founder in the Bay Area, I benefit from several of their clever, disruptive, well-executed solutions, particularly Lyft, Munchery, Caviar, and Instacart.
#divtech #dataviz #diversity #siliconvalley #decisionmaking