Skip to content

Common Obstacles to Analytics Success: Execution

What are the common hurdles encountered when putting analytics to work in a business, both in developing analytical models and applications and in building enterprise analytical capability?

This question is central to IIA’s mission of helping organizations navigate the many challenges to achieving analytics maturity.

Succeeding with analytics, and sustaining that success, is complex. It requires both a multi-pronged approach and an awareness of the pitfalls that analytics leaders and teams often face along the path toward increased analytics maturity.

In the second installment on our series on some of the biggest obstacles data and analytics organizations face today and ways to overcome them, we focus our attention on execution challenges.

1. Projects: Not Embracing Agile Analytics

IIA has observed the growth of agile methodologies tailored to the unique demands of analytics project execution, a trend referred to as “agile analytics.” Many organizations have eagerly incorporated this approach into their execution plans, reaping its benefits.

Early attempts to apply agile methods directly to analytics sometimes faltered due to rigid implementation. Unlike software development, certain analytics tasks don't neatly fit into short, independent work threads. Occasionally, the reluctance to embrace agile within analytics is wrongly attributed to analysts rather than recognizing the need for a tailored approach.

To leverage agile analytics effectively, align it with specific goals like multifunctional teams, active business expert involvement, improved project prioritization, iterative analysis, and quicker insights. It's crucial, though, to customize agile techniques to suit analytics needs. Identifying which aspects apply directly and which require adaptation enhances team execution and engagement.

Mastering agile execution takes practice, but adopting agile elements can add much-needed structure to complex or open-ended projects. Transitioning from the traditional "waterfall" approach can modernize analytics organizations, making them more efficient and better aligned with business needs and expectations.

Warning Signs:

Some of the common warning signs in this area include:

  • Rigid application of classic agile methodologies leads to confusion and/or lack of success
  • Lack of experience with agile
  • Team members who declare that agile methods aren’t needed

Overcoming Agile Obstacles:

A thoughtful, flexible approach to agile analytics project management and execution will benefit from involving teams in the design process.

  • Use flexible project templates, tap agile-experienced staff for advice on best practices, and adopt elements that set analytics teams up for success.
  • Hire people with agile experience if internal resources are thin or non-existent.
  • Read up on the well-documented ways to adapt traditional agile techniques to enable agile analytics and then adjust to fit your organization’s culture.

Developing a Product Orientation for Analytics and AI

Experts at IIA put together this eBook that provides a look at the steps and strategies to develop a robust product orientation for analytics and AI projects. The eBook covers:

  • Why a product orientation is necessary
  • The tasks involved in analytics and AI product management
  • A glimpse into a case study on product orientation at Regions Bank
  • The key prerequisites and capabilities to establishing a successful product orientation

2. Models: Slow Adoptions of Analytic/AI/MLOps

One of the frequent challenges of analytics execution is the ongoing management of the multitudes of processes that are deployed. As the number of deployed analytical processes has exploded in recent years, the need for better protocols to manage them has increased. This need has led to a rise in what is known as AIOps/MLOps/AnalyticOps. We’ll call the related family AnalyticOps here. They are all tied to the underlying theme of being much more formal in the management and tuning of analytical processes. It is no longer sufficient or scalable to have those who coded an analytical process then manage it once deployed.

Leading technology companies have been rapid adopters of AnalyticOps. Such organizations already had a heavy need for the better-known DevOps function to support scalable systems in general. Much as agile analytics methods are simply general agile methods tuned for analytics, AnalyticOps is about efficient management of the underlying systems, processing, and execution of analytical processes by borrowing from DevOps practices.

There is no doubt that making AnalyticOps a part of their strategy is critical for organizations who aspire to become analytically mature. Unfortunately, many organizations have been slow to implement AnalyticOps.

Warning Signs:

Some of the common warning signs in this area include:

  • Analysts complain that they spend as much time managing production processes as they do building new ones
  • Analytical processes are too slow and using many computing resources to reach the scale required
  • There is a lot of focus on making sure an analytical process is producing accurate results, but little time spent optimizing it

Overcoming AnalyticOps Obstacles:

In order to effectively make AnalyticOps a part of your strategy and speed implementation:

  • Begin thinking about how an analytical process will be deployed as the initial scoping is being done. Don’t wait for it to be an afterthought.
  • Document the expected plan and have someone with expertise in optimizing system performance – an AnalyticOps specialist – drive the implementation.
  • Once the process is deployed, be prepared for the need to make some changes until the process can be optimized effectively.
  • Set the expectation that additional time will be required for tuning the code itself, as well as the mix of computing resources the code is executed against, before the process will work as efficiently and cost effectively as possible.

Revisiting Common Obstacles eBook

In this eBook, we focus on some of the biggest obstacles faced today. We grouped these obstacles into four categories and tackled two obstacles per category:

  1. Business

  2. Execution

  3. Data and Technical

  4. People

3. Innovation: A Lack of Focus and Funding

As companies invest in enhancing their analytical capabilities, the focus on innovation, a key driver behind these investments, can sometimes wane. Analytics has driven innovation across various industries, but as the demand for everyday analytics grows, attention and funding can shift from pursuing the next breakthrough. With the current emphasis on AI in business, analytics teams have a significant opportunity to explore and adapt novel methods for both traditional and new data and use cases. However, harnessing these innovations requires resources, including time, effort, and tooling. Without a deliberate focus on enabling this process, valuable innovation opportunities may be missed.

Warning Signs:

Some of the common warning signs in this area include:

  • There is no line item in the budget, nor time in the team’s annual plan, tied specifically to innovation
  • Analytics teams’ incentives discourage innovation by being solely focused on operational deliverables
  • The company culture doesn’t value any effort that isn’t a success, and so people are afraid to innovate

Overcoming Focus and Funding Obstacles:

To avoid losing the drive for innovation:

  • Make time during annual planning to discuss potential innovation projects for the analytics team.
  • Arrange for innovation discussions with business partners and leaders to generate fresh ideas for projects and exploratory analytics work. An analytics leadership council can also be an enabler for innovative project ideas as it brings a broad audience together to discuss ideas.
  • Fight to have at least a portion of the annual budget targeted at innovation.
  • Also be sure to allocate innovation time formally into employees’ annual plans.
  • Last, carefully manage the expectations around innovation initiatives so that stakeholders understand the risk.