There’s no denying we live in a new world when it comes to data and analytics. A frequently-quoted statistic from IDC is that humanity is expected to have generated 44 zettabytes of data by 2020, just two short years from now. It’s hard to even comprehend the magnitude of such a titanic avalanche of data. But that’s just the beginning. By 2025, IDC predicts data volumes will more than quadruple. That’s right - in less than a next decade, we’ll have created over 180 zettabytes of data.
It’s no surprise, then, the data analytics market is expected to grow from $130B in 2016 to $203B by 2020, and continue this upward trajectory as we approach 2025. The advent of consumerized, self-service tools are making particular headway, putting data in the hands of business users and decision makers without needing a swarm of analysts.
The idea of self-service analytics is ruffling quite a few feathers in IT departments, however. Faced with years of fending off shadow IT and BI companies seeking to circumvent IT by selling directly to lines of business, many IT leaders are concerned about the impact of these analytics solutions on data governance. It’s felt like a zero sum game - either IT must control everything centrally and be beholden to slow, inflexible reporting cadences, or turn over data to departments and individuals to use at their own peril, risking misalignment or duplication of metric definition. Or worse.
But it doesn’t have to be this way. Self-service analytics and enterprise grade data governance are very compatible. Some practices need to change, but organizations shouldn’t rethink everything in order to get data into the hands of the users who need it most.
New Process, Same Owner
While data itself should be accessible to end users the official definitions of metrics must still come from a centrally governed and managed process. Analysts must still ensure data is correct, meaningful, and timely. The big shift, however, comes in how enterprise definitions are discovered and enforced..
Instead of trying to ensure alignment on a per-report basis, which is typically done in a traditional analytics organizations, analytics leaders and their teams should shift to enforcing data standards per metric or per attribute. By offloading the report creation while separating metric creation, they can much more effectively scale without sacrificing data quality.
In fact, pushing the process of getting specific answers out to individual users can help with one of the biggest data governance challenges—giving everyone visibility into what exists and closing the gap between data stewards and data users. This is at the heart of self-service analytics and data discovery.
Get Clarity on Meaning
Central IT teams previously not only controlled who had access to data, they were responsible for discovering the meaning of insights gleaned from data. In the new world of self-service analytics, this changes. IT teams still need to ensure that people don’t have access to data they shouldn’t and ensure that metrics and attributes exposed to end users are correctly defined, but they only need to do that a single time, instead of acting as data interpreters every time someone asks a question.
They can do this by following the same governance that’s worked in the world of traditional business intelligence—then defining terms once in modern self-service technologies and trusting the end users to use the data the right way, all while knowing they’ll have visibility if data is misused in anyway
This has two benefits. First, it dramatically reduces the burden on analysts, who can spend more time on meaningful data projects, like big data applications or AI initiatives, instead of spending hours simply interpreting data. Second, it allows business users, who are the most intimately familiar with the context of the data, to interpret it through this lens, without fear they are misunderstanding the data, or worse, that they cannot trust it.
Security and Privacy Are Non-Negotiable.
Data discovery and self-service analytics tools make data directly accessible to non-technical users, but it’s critical these tools don’t prevent IT from enforcing security and privacy. If these two core requirements aren’t prioritized, the potential risk outweighs any benefit accessible data provides to a business.
Many people have argued the advent of self-service analytics means that new governance policies and processes are required, but in fact, this is more about the technology employed to democratize access to data. Historically, enterprises have had to live at one of the ends of the spectrum: either control the data tightly in the name of privacy and security but seriously reduce the ability for business users to get answers, or provide access to data and forego the ability to secure it. That was the only choice afforded by legacy analytics.
Today, everything is different. In the same way that someone searching the internet using Google feels like they have all the power in their hands as they privately surf the web, in reality everything they do is visible to someone at Google with access. The same ability to empower end users without giving up visibility, and thus risking privacy or security, is finally possible for enterprises. Self-service analytics doesn’t create an entirely new way of getting answers; it extends a workflow that’s already been proven successful in the consumer space.
As we continue to see an explosion of data, the ability to provide data-driven insights to business users to inform decision making at every level will be a critical factor in determining which businesses succeed—and which fail. IT teams, however, don’t need to throw out the data governance playbook entirely. Rather, by evolving the process by which they enforce data definitions, creating a common understanding for how metrics are defined, and choosing the right solution that ensures centralized control over security and privacy, IT teams can reap the benefits of self-service without endangering their business.