Self-service analytics, a term that has come and gone—and come back again—in the analytics space for the past decade plus, was pushed to the forefront once again at the Gartner Data & Analytics Summit 2023. At a high level, self-service analytics is enabling and encouraging business users to run queries and create reports on their own with limited help from IT. However, what self-service analytics meant 10 years ago versus today has evolved quite a bit. The concept has become so popular again that it was featured in over 50 abstracts at the conference.
As data analytics adoption grows across organizations, self-service analytics has become even more important today. According to a recent Gartner survey, the number of employees leveraging analytics and business intelligence has increased 87% in the organizations surveyed. The need for business users and analysts to do more with data is growing faster than ever. With the growth in analytics, the concept of what self-service analytics is has changed over the course of the past decade, while the future of self-service analytics is undergoing radical changes with the introduction of new business-user-friendly analytics tools. Below we’ll go in depth on our findings from the Gartner Data & Analytics Summit and recommendations for the future, as well as challenges to be aware of in your efforts to further self-service analytics within your organization.
Written by: Michael Marolda,
Senior Product Marketing Manager at Tellius
In This Post
The Changing Definition of Self-Service Analytics
As the definition of self-service analytics has changed over the last 10 years, so too have the capabilities touted by business intelligence and analytics tools. At one point, the idea of being able to click and change variables in a dashboard to do some analysis was revolutionary. Then came drag-and-drop visualizations with a smoother user experience, designed to lower the barrier of entry to business intelligence. However, the platforms still created friction for the user base who needed to understand how to use these more traditional business intelligence tools to run analysis and get answers to critical business questions. The advent of AI-powered analytics has changed the game and opened new possibilities for self-service analytics at the business user level.
At Gartner’s Data and Analytics Summit, Tellius hosted a packed roundtable discussion with over 20 data and analytics leaders around how self-service analytics can be taken to the next level with AI/ML. The first question at the roundtable, “What is the definition of self-service analytics?,” had a quite wide range of answers, from providing access to dashboards for a broad community of business users to giving business users the ability to create more advanced ML models for predictive analytics. As the discussion continued, one of the participants mentioned that in the ideal end state, she would like a Google-like interface to surface data based on questions, while another mentioned the idea of providing business users with an experience like Siri or Iron Man’s J.A.R.V.I.S.: i.e., a natural language interface for an advanced AI system, specifically for analytics. As you can see from these examples, the definition of self-service analytics has indeed evolved greatly from where we were when the concept was initially introduced.
The Future of Self-Service Analytics
Enabling self-service analytics is not just a technology problem, but also an organizational and architectural problem. There is a never-ending struggle in many organizations between centrally controlling data and analytics from a data team and providing freedom to make reports at the line of business team level. Each of these paradigms has its benefits and drawbacks. On the one hand, the centralized model provides the ability for data teams to maintain consistency over the data and share best practices while allowing for consensus building within the organization. A decentralized model, providing self-service analytics to individual domains within an organization, allows for domain experts to supercharge their decision-making and provides a greater degree of business agility by enabling these domain experts to iterate on their analysis at greater speed. These two paradigms are the inverse of each other, meaning a centralized model will be slower and less innovative, while a decentralized model will have less governance at the expense of getting the most value out of the data.
Gartner’s proposed solution to the problem, a best-of-both-worlds model, is to open analytics franchises in each department. This new hub-and-spoke model calls for a data and analytics franchise in each department with a data and analytics center of excellence at the center. The analytics franchise model brings the best practices of the centralized model to the individual domains by incorporating more of the quantitative and technical skill of a data team in marketing, sales, HR, and the other domain-specific areas that would benefit from being more data-driven. According to Gartner’s recommendation, for each domain opting into becoming an analytics franchise, there would be an analytics leader responsible for bringing the best practices of the data and analytics center of excellence to the specific domain. This new paradigm would help to ensure consistency while allowing for greater innovation and, ultimately, better business results.
In addition to the need for organizational change to advance a self-service analytics philosophy, Gartner suggests a few critical tools to help your organization push forward:
Data Preparation
First, through what is known as the “last mile of analytics consumption,” self-service tools for data preparation will help individual teams reduce time to insight by allowing for an iterative and agile process of finding, joining, and transforming data at the user level. It is especially important to incorporate low- or no-code tools to open data access to new lines of business users. This self-service approach to data preparation allows data teams to focus on more crucial issues while giving analysts and business users direct access to the data needed for important decision-making.
Data Catalog
Another key feature of modern self-service analytics is a data catalog. During the Tellius roundtable on self-service analytics at the Gartner summit, one thing was abundantly clear: Finding and sharing data can be a fraught proposition in many organizations. The trouble lies with the process of finding and understanding the underlying data prior to analysis. A data catalog enables users to be more efficient with their data usage by providing a unified repository of metadata from all data sources available to an organization. Data catalogs help organizations understand where the data is located, how to access the data, and what the data is. These products should be a shared responsibility between domain owners and centralized data teams. This collaboration will improve accountability and governance across an organization.
Cloud Analytics
Finally, cloud data warehouses and data lakes can help centralized data and analytics teams focus on more important tasks. Cloud-based analytics tools like cloud data warehouses can help reduce tedious administrative activities while automating work traditionally performed by the central data team. These tasks include provisioning, managing, securing, optimizing, and scaling. This provides central data teams with more flexibility to handle incoming requests, improve data governance, and train domain teams on important data literacy issues.
Challenges of Self-Service Analytics
There are still many challenges to the successful adoption of a modern self-service analytics approach at organizations. One challenge of particular importance to the adoption of the analytics franchise model is staff shortages due to hiring difficulties. Gartner’s model calls for individuals with quantitative and technical skill sets within each analytics franchise domain. According to a recent Gartner survey, 45% of data and analytics leaders reported skill and staff shortages as a critical roadblock to success. With the U.S. unemployment rate still near a 50-year low, this challenge is not going away anytime soon.
A second challenge across most organizations is data literacy. According to Gartner, 47% of data and analytics leaders rank data literacy as one of the top three challenges in their organization. Poor data literacy leads to poor business decisions due to the inability to interpret data correctly. It can also cast doubt on the ability of organizations to get business value out of data and analytics initiatives over the long term. Meanwhile, a failure to document business outcomes from data and analytics efforts can lead to lower analytics investment in the future, thus potentially impacting decision-making down the line.
Conclusion
From legacy business intelligence to augmented decision intelligence, self-service analytics today is a far cry from its initial conception. Yet challenges still remain to the successful adoption and implementation of a self-service strategy across an organization. How best to assemble your tool set and organize your business is still up for debate, and while Gartner has proposed some solutions, there’s no “one size fits all” model for success.
To learn more about how Tellius enables self-service analytics, watch our webinar, Self-Service Analytics: The Next Era of Data Democratization with AI and Machine Learning.
