Misdirected Analytics: The Cost of Not Prioritizing

One of the most common reasons analytics teams fail to drive impact is very simple: too many projects, too many ideas, and too little prioritization.

When leaders treat every request as urgent and teams are spread thin across several efforts, the result is predictable: slow delivery, limited follow-through, and underwhelming results. Divided resources and focus reduce impact and delay progress on high-value opportunities.

Without a way to assess priorities, your analytics team becomes a reactive support desk instead of a strategic force. To avoid this, it’s important to ask: Do we prioritize analytics projects according to the business value they deliver?

This is part five of our series on eight common gaps quietly draining ROI from data, analytics, and AI initiatives—and how to move from investment to impact. Here, we will explore key challenges blocking effective prioritization and how to be more strategic with the projects you select.

Consider a supply chain manager at a consumer goods company pushing to forecast demand for a fast-moving product using historical sales data. On the surface, it seems straightforward, but the data turns out to be incomplete and inconsistent, with key promotional events and regional trends missing or misclassified.

The result? A model that performs worse than, or on par with, the current process. The valuable time a team took to produce the model has garnered results that stakeholders don’t trust or use. This could push organizational adoption back by months—or even years.

Stakeholders must have a thorough understanding of their available data to properly assess the feasibility of their desired projects. Without an awareness of the data’s edge cases and the sources at their disposal, there will be unrealistic expectations and wasted resources. If the data needed to answer a specific question isn’t available, one needs to reframe the question or seek more fitting data.

Without thoroughly understanding one’s data resources, it’s hard to distinguish high-value opportunities from low-probability projects. When teams have clarity on their data’s characteristics and uses, it’s easier for them to set the right priorities.

To accurately prioritize analytics projects, teams need a clear understanding of the business value each project is meant to deliver.  If analysts and stakeholders aren’t aligned, even a prioritized project will end up unused.

Consider a project that estimates when a food delivery will arrive to a consumer. What is the current decision-making process? How will the company communicate the new framework to customers? What are the business guardrails? Is it worse to overpromise or underpromise on delivery times, and how can we quantify the cost of each type of error?

This alignment must happen before work begins. Once we have defined the true purpose of the project, we can begin determining the teams, data, and tooling the project requires.

Analytics teams need to be aligned with stakeholders at the very beginning of a project. They need to agree on the overarching and subtle project themes. Otherwise, they’ll find it hard to achieve priorities. Technical teams must understand the nuances of how stakeholders currently make decisions, while stakeholders need to grasp how their business context translates into technical constraints and possibilities. When teams understand the full business context of a project, they can better prioritize all possible projects.

Consider a marketing team pushing to buy a new AI-powered customer intelligence platform after seeing multiple demos. This platform has AI-generated audience segments for targeted marketing and automated campaign optimization. But when the analytics team reviews the current company setup, they find that there are existing tools to support customer segmentation, A/B testing, and campaign reporting.

Despite already having tooling for targeted campaigns, most campaign decisions are still made manually because teams don’t trust or use what is available. Is the new tool the best solution, or is working on improving current tools and building stakeholder buy-in more productive?

At times, the newest iterations of data tools can look like the best fit for overcoming gaps that have been plaguing the business. But before committing to a new tool and prioritizing its implementation, teams need to fully understand their current capabilities. Evaluating existing tools against questions like these can be helpful:

• What are the features of our existing tools, and have we used all of them?
• What high-value results have we seen from existing tools?
• What parts of our existing tools take the most time to use?
• What parts of our existing tools frustrate our team the most?

Overestimating what’s feasible with new tooling or underestimating what is currently available can lead to projects that don’t meet productive business goals.

When teams focus more on the tool and less on the business need, it can lead to prioritizing projects with less demonstrable impact over projects that can truly propel the business forward. Ultimately, the best tools don’t just have better features; they drive better business outcomes.

1. Meet early and often throughout a project lifecycle.

Data limitations emerge as work progresses, and data anomalies require business context to interpret properly. Technical teams must communicate data constraints clearly, while stakeholders need to explain what unusual patterns might reflect in real-world operations. Ensuring meetings throughout a project lifecycle matches the iterative pacing of analytical projects. Each experiment that occurs throughout a project allows the team to learn something new and make key decisions on how they should adapt. Having a unified view on the direction of a project leads to higher rates of adoption.

2. Translate business significance into statistical significance.

Consider a wholesaler using sales forecasts to guide purchasing decisions. What’s worse for their business: under-forecasting (leading to stockouts and frustrated customers) or over-forecasting (resulting in excess inventory and storage costs)? Do they have warehouse flexibility? How does customer frustration with missing products compare to the bottom-line impact of excess inventory? These business trade-offs directly inform technical modeling choices and allows us to translate ideas into high-value projects.

3. Maintain flexibility in tooling and methodology.

Analytics rarely offers one perfect solution. Instead, there are multiple viable approaches—each with distinct trade-offs. Success comes from choosing the approach that best aligns with business priorities and constraints. Should the team use Tableau Dashboards? Should their tech stack involve Databricks or Apache Airflow? Where will the team store the data? Understanding what we have access to as a company and how these different tools fit our current project needs is more critical than utilizing a specific product. Knowing what is available and how it aligns to our use case helps us better prioritize our projects.

4. Assess and rank project value.

If there are too many ideas on the table, implement a prioritization session. Capture all ideas currently in the team’s backlog. Find a way to quantify their value. One simple way to do so is to plot them across an effort versus impact matrix. Once the team properly assesses effort and impact, have discussions on how best to staff projects in an order that prioritizes the projects that best meet the company’s current priorities. This allows for a fruitful pipeline as well as a robust backlog.

Every decision in the analytics process serves a specific business goal. When stakeholder vision aligns with the technical modeling approach, projects face fewer costly rewrites and deliver outcomes that are more likely to be implemented. If stakeholders understand key framework decisions, they will trust and use the framework more readily.

Prioritization can be strategic rather than reactive. If there are twenty potential projects but the team only has capacity to address five, a robust picture of available data, tooling, and business value enables teams to confidently choose which five should receive resources. Completed projects are the ones that deliver value, so it’s often better to focus on delivering one project well than making vague progress on multiple issues.

In our work with clients, we’ve been able to help companies succeed. We helped:

• A global quick-service restaurant chain streamline their data sources and align cross-functional teams to be able to properly prioritize their analytics projects.
• A leading company develop a strategic roadmap and Analytics Center of Excellence, empowering them to prioritize projects based on ROI.
• A consumer goods company unify their work under a shared strategy, ensuring all projects were evaluated based on ROI and feasibility.

When technical and business teams share a common language around feasibility, constraints, and value, prioritization provides strategic clarity. At Elder Research, a MANTECH company, we help organizations align on priorities that drive real value. We’re always glad to connect about ways we can support your data and analytics efforts.