8-minute read

Gleaning actionable insights from datasets is a core function of modern business. It’s essential in every workflow from designing useful dashboards and other visualizations for internal stakeholders and users, to optimizing a marketing channel based on insights about customer behaviors and preferences.

The value of highly interconnected and easily accessible data is unmistakable. More than a decade ago, a financial services company in Delaware began an analytics initiative that eventually yielded more than $7 million in ROI, thanks to lower operating expenses and increased loan volumes, according to CIO Magazine. The effort zeroed in on how specific data fields affected critical tasks like loan processing, and helped decision-makers optimize form design, standardize field definitions, and conduct regular data cleaning.

On the flipside, poorly designed data analytics projects take a major toll in the form of costly delays and failures; Gartner estimated in late 2017 that more than 4 in 5 big data projects don’t deliver any value. The causes of their failures run the gamut and include common issues with project baselining, data governance and system/source interconnection.

Why many data analytics projects don’t live up to the hype

At the beginning of the decade, big data was one of the most hyped concepts in tech, and for good reason. The concept of aggregating and analyzing information at a vast scale had clear potential to transform IT as well as line-of-business operations, and many organizations were able to capitalize through the transition to cloud computing infrastructures and DevOps-focused cultures.

Others have struggled, and understandably so. Overseeing an analytics project is a complex process, with many moving parts prone to failure, such as:

Project purpose

Imagine looking for something that you wouldn’t even recognize upon sight. It sounds crazy, but it’s the jumping off point for many analytics initiatives, which begin as quests to simply “find” insights within a dataset without knowing what question is being answered.


To avoid this pitfall, agile management emphasizes rapid iteration that continuously ensures your projects address real business problems. In other words, the Agile approach ensures your analytics strategy evolving in tandem with the targets you’re seeking to hit.

Baselining and methodology

Similarly, it’s hard to know if an analytics push succeeded if there’s no definition of successful performance to begin with. In a comprehensive 2016 survey, more than one-third of PMs said project schedules weren’t baselined. One-quarter also confirmed that no consistent methodologies or risk management practices were applied to their projects. When methodologies are in place, they may be outmoded ones like waterfall that don’t emphasize collaboration and continuous improvement.

Data governance

Becoming a data-driven organization is not a ticket to success in and of itself. There must also be ample access to reliable and clean information from relevant sources and systems. Unfortunately, many companies don’t have the wherewithal to make this happen. A 2018 survey of C-suite executives and IT personnel revealed 46 percent didn’t have a data governance strategy at all, while 40 percent had no dedicated budget to support it.

Stakeholder engagement

An analytics project should help its stakeholders reach well-defined goals. A project without actively engaged stakeholders is like a plane without a pilot – it won’t go anywhere. Poor engagement leads to uninformed decisions and lackluster execution, which is why it’s important to have a formalized process in place for getting stakeholder buy-in early on.

Lifecycle management

How long is too long to hold onto data? While there’s no universal answer to this question, inexpensive storage has made it easy for organizations to retain information for longer than ever, leading to situations in which stale data can corrupt analytics initiatives.

The Agile analytics difference

Agile methodology helps keep analytics efforts focused and on-schedule. By making the switch to Agile, organizations can ensure better responsiveness to change, superior collaboration, and overall higher success rates for projects. In fact, Agile projects have a 28-percent higher success rate than traditional ones.

In practice, an Agile project can bring previously siloed teams together. That helps with stakeholder engagement, baselining, project scoping, and all of the other frequent problem areas we described above. An organization with Agile analytics can better connect its data sources, consolidates its process, and accelerate overall time-to-market for strategic products and services.

Logic20/20 will help your team connect all of the dots during even your most complex analytics projects, so that you can solve big problems without getting bogged down by unclear objectives or outdated data. We believe in the power of visualizations. Learn more on our business intelligence and analytics page.

Like what you see?

Paul Lee

Ilya Tsapin is a Director at Logic20/20, and lead of the Architecture Practice.