A quick search for “self-service analytics” on Google yields more than 9 million results. That’s quite a lot. And for a good reason: just about everyone in the business intelligence (BI) and analytics space has been talking about this subject for more than a decade. Anyone who’s walked the exhibition hall of a big data conference has undoubtedly seen those words plastered on booth backdrops, zoom across screens, and appear in countless keynotes. It’s safe to say self-service analytics has captured the hearts and minds of the entire industry.
Some experts claim self-service analytics is just a myth, while several companies claim to provide it as a service—others believe it’s finally coming with the next wave of artificial intelligence and natural language processing set to hit software. For a subject so often written and talked about, it’s incredible that our industry can’t make up its mind. Not to sound like a broken record, but we think it’s time to take another serious look at the state of self-service analytics and initiate the next chapter of the story.
What Exactly Does Self-Service Mean?
It helps to define what exactly we’re talking about here. Gartner defines self-service analytics as “a form of business intelligence in which line-of-business professionals are enabled and encouraged to perform queries and generate reports on their own, with nominal IT support.”
A couple of things to keep in mind here (that we’ll unpack later) as we explore the subject:
- The emphasis on enablement and encouragement of business professionals to ask their questions, get answers, and generate reports and valuable insights.
- The proclamation that it should require little to no IT support throughout the process.
With this in mind, let’s take a look at the state of self-service today.
Self-Service or Lip Service?
As data lovers, we love to set goals and measure success. It’s in our DNA. If we want to measure the success of something like self-service analytics solutions, one key metric immediately comes to mind: the rate of adoption.
The promise of self-service has always been a software environment where every employee in an organization can participate in the data conversation—without a background in data science or extensive product training—by asking questions relevant to their domain and get the right answers to help steer better decisions and overall business success. Unfortunately, that promise has not come to fruition—yet.
While many BI vendors have built powerful software that makes it possible to deliver rich business insights with extensive training or expertise, nobody has conquered self-service. How can we be so sure? Today the rate of BI adoption hovers around a paltry 30% of all employees.
Anyone claiming they have delivered a real self-service solution is paying lip service to the idea without admitting they have not yet delivered on the greater promise of accessible analytics software that makes it easy for all employees to answer meaningful questions. While many of the most popular BI solutions make it technically possible for business experts to ask questions, visualize data, and build dashboards or reports, they fall short in achieving this goal without months of setup, weeks of training, or daily assists from the data team. And even then they usually end up with a broken dashboard designed to answer a single question, or resort to pulling the report into excel.
Is “Self-Service” the BI Unicorn?
With all the money dumped into the BI and Analytics space over the years, you’d have thought the challenge would have been solved by now. If billions of dollars invested in software hasn’t delivered total BI adoption, is it even possible? Perhaps, at this point, are we merely chasing some self-service unicorn that can’t be caught? We don’t think that is the case. We believe the next wave of cloud BI and analytics tools will finally deliver on the age-old promise that everyone has been seeking all these years.
We know you’ve heard this claim before, but this time it’s different. Think about it for a second; the modern cloud analytics stack—made of data pipeline tools (like Fivetran), cloud data warehouses (such as Snowflake and Redshift), and cloud-native analytics tools (like Sigma)—presents an opportunity we’ve never seen before. Today, it’s easier than ever to source data from every corner of the business, store that data in managed, elastic cloud data warehouses—where even the toughest queries can be processed in seconds—and interface with billions of rows of data without ever having to write a single line of code or learn a complicated BI tool. If you’d have even suggested this future was on the horizon even ten years ago, people would have written you off as crazy.
It’s Time to Tear Down the Wall
Technology is only a part (albeit usually a large part) of any game-changing business solution. But what many often forget is that for any transformational technology to reach its true potential, we must also change the way we think about the challenge standing in front of us. And that’s why it’s time for a definitive paradigm shift in the way organizations approach data access, modeling, and governance.
To reach total BI adoption, it will require connecting the two realms that have remained separate through the entirety of BI history: the data team and the business experts. BI vendors have historically held business experts back by failing to provide untethered access to data or sought to make them think and act like a programmer to get the answers they need. Unsurprisingly, this hasn’t worked.
“ It’s not the line manager’s job to be a programmer or analyst any more than it is the programmer’s or analyst’s job to deal with managing the business. What’s needed is to assist the manager. ” — David Teich, Forbes
Sure, many BI solutions that provide business teams access to extracts and subsets of data, or let them “explore” data in a sandbox curated and limited by pre-defined data models. But the reality is that no data team can ever anticipate every question a business team will ask. It’s a fool’s errand to think otherwise. Instead, we should trust the business experts to know their domain, while also providing an analytical interface that enables unbridled data exploration—and the ability to ask the toughest questions without the need to learn or write SQL, or some proprietary version of it.
That’s why we’ve built Sigma, a tool designed to unlock deep data exploration via a visual analytics experience inspired by the most adopted interface of all time: the spreadsheet. We believe that the Sigma interface, coupled with visual data modeling tools that allow business teams to join the data curation process, is the first step in breaking through the adoption barrier and realizing a future where every employee can not only access insights, but also navigate their analytics journey without constant help from the BI team.
So, Where Do We Go From Here?
While the technology is certainly available to close the BI adoption gap, there’s still more work to be done before we can bring business intelligence to the other 70% of employees. From here, companies building the next generation of cloud-native software, and data-first organizations adopting these technologies, need to work together to further bridge the gap between business and data teams. We need to keep looking for ways to streamline data flows into the warehouse and provide data access to entire organizations with easy-to-use but powerful data exploration tools. And that's exactly what Sigma will continue to do moving forward.
The road ahead is long, but we all know where we need to go. It’s time to stand up and solve this challenge once and for all.