BI Trends

The Price of Success: Managing Unsustainable Feature Growth in Analytics

By Brian Brinkmann
Share on LinkedIn Tweet about this on Twitter Share on Facebook

Eighty-five percent of application teams are currently offering some form of embedded dashboards and reporting in their applications. Unfortunately, the price of successfully offering analytics in your application is an endless flow of requests for new features. The challenge to satisfy users with more analytics features and increasingly sophisticated capabilities is staggering—and software development teams can no longer keep up on their own.

>> Related: Why “Build or Buy?” Is the Wrong Question for Analytics <<

The result? A bloated development queue and burnt-out developers. Fortunately, there is a way to handle the never-ending requests for more analytics features.

Partnering with an analytics development platform allows application teams to customize exactly what they need while reducing time to market by providing a litany of pre-built elements (including charts, dashboards, and customization themes) that serve as a ready-made infrastructure for embedded analytics. Developers avoid having to write and check every line of code as they would if they were building the application from the ground up.

Analytics development platforms also support more sophisticated capabilities than other business intelligence solutions (including those built in house). “Modern embedded analytics platforms don’t deliver a set of monolithic tools,” writes Gartner in a recent report, 5 Best Practices for Choosing an Embedded Analytics Platform Provider. “Instead, they support a full stack of integrated analytic functions—from reporting and dashboards to self-service analytics, alerts, collaboration, data preparation and machine learning on a unified, scalable architecture with common administrative and management functions.”

What challenges are application teams facing with embedded analytics? And how can they leverage a development platform to keep up?

Analytics Today

For years now, we’ve seen software teams updating their embedded analytics once a year or every few years to add features that service the core value of the application. However, those features have evolved from simple reports and data visualizations to self-service features, predictive analytics, write-back, and many more advanced capabilities.

Requests for more analytics features have continued to pile up, resulting in an inflection point of unsustainable feature growth for embedded analytics platforms. Unless you’re one of the top five global software companies (with seemingly endless resources and talent at your disposal), the ability to offer your customers every analytics feature they want has become nearly impossible. Complicating matters is the fact that these aren’t capabilities you can just set and forget. They need to be maintained and kept up to date.

Analytics Tomorrow

According to our 2018 State of Embedded Analytics Report, by 2019 the average application will support 11 of the 13 analytics capabilities we surveyed, which include data visualizations all the way up to predictive analytics. The number one capability is data visualizations, with nearly 70 percent of applications supporting it.

But across the board, almost every capability is supported by more than 40 percent of applications. This represents incremental growth in the availability of these features. Still, even the most advanced capabilities are becoming more prevalent. In fact, our survey shows that more than 30 percent of applications already support Artificial Intelligence (AI).

To remain competitive, software companies are having to iterate faster than ever before. At the same time, analytics features are taking longer to deploy because of their growing complexity. Eighteen percent of our survey respondents say it took them longer to get to market with their analytics solution than expected, and more than 25 percent say the people and resources required were greater than they expected.

The Future of Software Development

As we’ve established, the price of successfully embedding analytics grows as application teams are increasingly overwhelmed by users demanding new capabilities. Typically, companies facing this challenge have chosen to build their own analytics solution or buy one from a third party because of cost or resources. However, as we outlined earlier, a third option is to take a combined approach that blends the best of both build and buy by leveraging an analytics development platform.

According to our State of Embedded Analytics survey, a combined approach helps application teams stay ahead of the embedded analytics feature demand:

  • Compared to building, it allows companies to get to market faster and reduce the long-term resources needed for maintaining and updating the features in their embedded dashboards.
  • Compared to buying, teams are able to completely integrate the look and feel of the analytics with their existing application and deploy sophisticated capabilities.

Analytics development platforms are not only the best way to get around long development queues and burnt-out developers, but also a viable path to sustainable software development.

To learn more about choosing an analytics development platform, read our 2018 BI Buyer’s Guide >


Originally published July 12, 2018; updated on July 31st, 2018

About the Author

Brian Brinkmann is the VP of Product Management at Logi Analytics. Brian has over 15 years of analytics and BI software experience. Prior to joining Logi Analytics, he held senior product strategy, management, and marketing positions with MicroStrategy, creating BI applications for marquee customers such as Nike and Franklin Templeton. Brian holds a MBA and a MEM from Northwestern University, as well as a Bachelor of Electrical Engineering from the University of Dayton.