Data is vastly more available, but the skills needed to analyze it are still far from universal.

Some years ago, Google’s chief economist Hal Varian blogged about the “democratization of data,” which he believed would make “information that once was available to only a select few…available to everyone.” This trend is one that “finally puts crucial business information in the hands of those who need it,” as Salesforce’s Robert Duffner put it recently.

But this particular democracy offers no guarantee that you’ll get to participate — suffrage is far from universal.

More People Have More Data

There are clear signs that data is becoming more widely available, and that the movement to democratize data is making real progress. Prior barriers such as infrastructure, culture, tools, and governance that once kept data out of the unwashed hands of the employee masses are quickly eroding. For example:

  • Rather than building specific analytical applications, IT departments increasingly focus on building infrastructure to empower users through self-service approaches.
  • Companies are progressively embracing the value of sharing information. Organizations like General Electric report that information silos are no longer holding them back.
  • Analytical tools to manipulate the data are widely available; familiar desktop productivity and server database software can now incorporate analytical tools.
  • Savvy organizations manage processes to avoid anarchy from “projects that tap into data sets of questionable quality and put compatibility, compliance, and security at risk.” The Coca-Cola Company, for examples, works to improve availability and decrease duplication by establishing large, shared data platforms.

As a result of these trends, many employees have access to more data. Recent research from MIT Sloan Management Review finds support for this: 77% of respondents report an increase in access to useful data since last year. Additionally, the number of organizations reporting increases is increasing — from 70% in 2012, to 75% in 2013, to 77% in 2014. There is a double increase here: More organizations are reporting access to more data. Access to data is no longer the bottleneck it once was.

Democracy? Not Quite

And yet, the result is not quite the new data democracy lauded in the blogosphere. A democracy would mean that everyone gets to participate in the move toward broader access to new troves of data. That’s not the case. Far from it.

Instead, this Potemkin data democracy is an illusion that appears when we incorrectly equate the ability to access data with ability to use data. Access to data is the first, critically important step for a data democracy, but by itself isn’t sufficient to make one happen. Data democratization also requires knowledge: The complex abilities to work with data, understand data analysis tools and techniques, and transform data into insights. Without these abilities, the data democracy is only an illusion — and most people are still unable to participate fully.

Lack of data knowledge hamstrings people in two ways: First, they are unable to use the readily accessible data well themselves, and second, they are unable to tell when others are using data poorly or disingenuously. Consequently, people with limited understanding of how to use and assess data (as well as evaluate the insights derived from data analytics) become second-class citizens in a data-ocracy.

As a result, current efforts to democratize data have more of the qualities of a meritocracy than a democracy. In a meritocracy, participation is associated with ability. Benefits from the democratization of data accrue to those with the skills to benefit from it.

Evidence of the meritocracy of data is ubiquitous. Power structures within organizations are not going away, they are changing. We see considerable evidence in disparity from key indicators like the quantity of jobs, salaries, hiring, and promotion. This disparity is an indication that, while the data plutocracy is gone, we are far from a data democracy.

Responsible Suffrage in a Meritocracy of Data

If a data meritocracy is emerging, how can we be responsible citizens? Fortunately, we have other notable examples of meritocratic approaches to draw from. For example, many open source communities embrace merit. Red Hat, for example, attributes some of their success to a merit-based approach. While everyone is encouraged to share their thoughts openly, consensus is not the intent; rather, people listen most to those who have earned that attention. A single person can make a difference at Red Hat, but only after earning trust and respect of others — earning the right to participate. Beyond offering positive guidance, these existing examples also offer guidance to avoiding pitfalls that can come with a meritocracy inside a company:

Don’t expect benefit without effort: Rewards from trends towards data democratization will be skewed towards individuals working to take advantage of them. Effort will be the distinguishing factor as data and tools commoditize.

Shun shunning: No one was born with data skills. Stay open to people with newly developed skills.

Guard against biases hostile to diversity: A related issue is that adhering to the idea of a meritocracy can disguise a desire to maintain a status quo that isn’t open to newcomers. Make sure that discussion and critique of new ideas focus on the contribution, not the contributor.

Avoid deification and elitism: Once seen as merit-worthy, talented individuals can become stars — a position laden with a lot of negative baggage. Acknowledging one person’s capabilities should not interfere with the ability of others to contribute.