4 Ways to Sabotage a Data Analytics Project

by | Jul 25, 2019 | General BI

Reading Time: 5 minutes

If we were to treat data analytics projects in the same way a physician would treat his or her patient, we would come to the realization that customized care is key. Every project is its own living being, with its own unique challenges and needs. It is for this reason why it is only natural that a solution should be custom as well.

However, despite the differences among various projects, there are a few symptoms of declining health that are more universal. If you find your data analytics project showing the following symptoms, allow yourself to take a step back and reevaluate the prognosis.

1. Not having a clear end goal

With every project you embark on, you must establish clear starting and ending points. At some point in your data journey, you should ask yourself, “Where is our organization now and where do we want it to be?” For some organizations, an end goal could be to have a tighter grasp on inventory management and sales. While for other organizations, it can be to improve staff performance and overall efficiency.

If your organization hasn’t defined clear start or end goals, then you may find your team wandering off and taking on smaller, less important tasks. Or they might not be engaged at all because they don’t understand where they are going. For leaders who want to make their team members shine, establish the destination and create a roadmap to get there.

In this case, a clear plan with defined start and end dates provided focus for the organization and allowed it to achieve measurable improvements in drug costs.

2. Working within a divided team

No patient with a complex diagnosis is cured by one person alone. So, in order to provide quality care, you assemble a team. But how valuable is a team if the team members aren’t in sync? If your team is tackling a project, you have to make sure that everyone understands their role, the obstacles that they’re about to face, and how they can work together to solve them. Regardless of your industry or the type of project that you’re working on, your team has to be on the same page.

What happens to an organization when its team is divided? It doesn’t function. When people have different objectives – or unclear objectives – they go off in their own siloed directions and fail to rally behind a goal.

One step towards having a unified team is to use the same definitions. If the meaning of your data doesn’t translate in the same way throughout your organization, what value does it hold when you’re working with it?

Dave Policano, a consulting CFO, advises organizations to “take time to define the terms that compose all metrics, no matter how basic.” Regardless of what the project is, he wants team members to have clear responses to what they’re working on. For some it can be what they “mean by ‘revenue’?” or “What do we mean by ‘attendance rate’?” By assigning meaning to a word, task, or idea, your team can feel confident when executing the plan of action.

3. Poorly designed dashboards

To digest data, a carefully crafted dashboard is as significant as the data itself. You may understand what your data represents, but how do you share that information with others in a way that is visually appealing and easy to understand?

A “lack of architectural consideration for production,” travels throughout your organization’s veins and impacts their ability to make data-driven decisions.  Chris Preimesberger of eWeek, states that:

“Many data science projects start without consideration for how the developed pipelines will be deployed in production. This occurs because the business pipeline is often managed by the IT team, which doesn’t have insight into the data science process, and the data science team is focused on validating its hypotheses, and doesn’t have an architectural view into production and solution integration. As a result, rather than getting integrated into the pipeline, many data science projects end up as one-time, proof-of-concept exercises that fail to deliver real business impact or causes significant cost-increases to productionalize the projects.”

Is this why some healthcare executives admit to not using their dashboards on a daily basis?

4. Using old data

Your doctor wouldn’t use old test results to diagnose a new condition. That seems ludicrous. So why are you using old data in an attempt to further your understanding of your organization? The side effects of using old, outdated data include headaches and unnecessary hiccups that take too much time to fix. While old data may hold value for some projects in which the data doesn’t change over time, it is counterproductive to use for new projects or those in which the data is not static. Data cleansing is a vital step in the right direction towards improving the likelihood of achieving actionable results.

What other “symptoms” are holding organizations back from successfully deploying an analytics project? Let us know in the comments below.

You may also like