Bob Violino
Contributing writer

Top 6 roadblocks derailing data-driven projects

Feature
Jan 30, 20239 mins
Data ManagementData Science

From budgeting issues to buy-in challenges, data initiatives all too often fall short from the very start. Here’s how to overcome those hurdles and set your org up for success.

data
Credit: Claudio Schwarz / Unsplash

Data is what drives digital business. Consider how strategically important it has become for companies to leverage advanced analytics to uncover trends that can help them gain decisive insights they might not otherwise possess.

But data-driven projects are not always easy to launch, let alone complete. In fact, enterprises face several challenges as they look to leverage their information resources to gain a competitive advantage.

Foundry’s recent Data & Analytics Study looked into why organizations have difficulty making good on the promise of data-driven projects, and revealed several key roadblocks to success. Here are the top six reasons data initiatives fail to materialize and deliver, as revealed by the research, along with tips from IT leaders and data experts on how to overcome them.

1. Lack of funding for data initiatives

Funding can be hard to come by for any technology initiatives, particularly in an uncertain economy. This certainly applies to data projects. These undertakings might be competing with a host of other initiatives in need of financing, so it’s important for IT leaders and their data teams to present a strong business case for each project, and to not make them overly complex.

“While budget is always tricky, this is a question of priorities and right-sizing the body of work,” says Craig Susen, CTO and technology enablement lead at management consulting firm Unify Consulting. “Looking for obvious outcomes [does not] always require reworking the entire infrastructure.”

Being data-driven is as much a cultural pursuit as it is anything else, Susen says. “It requires designing/rethinking key performance indicators, capturing data in a smart timely manner, landing it in common areas quickly,” he says. “Then it can be evaluated and aggregated, either applying advanced visualization technologies or working it against machine learning algorithms. It’s all a complicated bit of science. Having said that, many companies overcomplicate this process by trying to do too much all at once or over-indexing in places that don’t drive true value to their businesses and customers.”

CIOs and other technology leaders need to develop strong working relationships with fellow C-suite members, particularly CFOs. In many cases it’s the finance executive who makes the decision on budget approvals, so to improve the likelihood of getting the needing funding technology chiefs need to be able to demonstrate why data-driven projects are important to the bottom line.

2. Lack of a clearly articulated data strategy

Lacking a complete data strategy to guide data-driven projects “is like not having an outline to guide a thesis,” says Charles Link, senior director of data and analytics at Covanta, a provider of sustainable materials management and environmental solutions.

“Every project should contribute some paving stones to the road leading to the desired destination,” Link says. “A data strategy identifies how to align information and technology to help you get there. Your business should be able to travel down the road as you deliver value.”

To be successful, a data strategy should have both a data management component — generally IT tools, technologies, and methods — and a data use strategy, Link says.

Oftentimes there isn’t a clear understanding within enterprises of what data is available, how the data is defined, how frequently it changes, and how it is being used, says Mike Clifton, executive vice president and chief information and digital officer at Alorica, a global customer service outsourcing firm.

Companies need to create a common language among stakeholders in advance of establishing any data-driven projects, Clifton says. “If you don’t have a solid foundation, budget and funding are too unpredictable and often get cut first due to a lack of clear scope and achievable outcome,” he says.

3. Technology to implement data projects is too costly

Making the challenge of getting sufficient funding for data projects even more daunting is the fact that they can be expensive endeavors. Data-driven projects require a substantial investment of resources and budget from inception, Clifton says.

“They are generally long-term projects that can’t be applied as a quick fix to address urgent priorities,” Clifton says. “Many decision makers don’t fully understand how they work or deliver for the business. The complex nature of gathering data to use it efficiently to deliver clear [return on investment] is often intimidating to businesses because one mistake can exponentially drive costs.”

When done correctly, however, these projects can streamline and save the organization time and money over the long haul, Clifton says. “That’s why it is essential to have a clear strategy for maximizing data and then ensuring that key stakeholders understand the plan and execution,” he says.

In addition to investing in the tools needed to support data-driven projects, organizations need to recruit and retain professionals such as data scientists. These in-demand positions typically command high levels of compensation.

4. Other digital transformation initiatives took priority

Digital transformations are under way at organizations in virtually every industry, and it’s easy to see how projects related to these efforts could be given a high priority. That doesn’t mean data-driven projects should be put on the back burner.

“If digital transformation efforts are taking priority over data initiatives, then you need to re-evaluate,” Link says. “All digital transformation initiatives should envelope data initiatives. You cannot have one without the other.”

Ignoring the data aspects of transformation could invite failure of other initiatives. “I would be concerned to pursue digital transformation without a solid data strategy, as the results, iterations, and pivots needed to be successful should all be data-driven decisions,” says David Smith, vice president and CIO at moving and logistics company Atlas Van Lines.

“If this is an organizational roadblock, I would recommend using the digital transformation initiative as the genesis of a data strategy execution,” Smith says.

5. Lack of executive buy-in or advocacy for data initiatives

If senior executives are not sold on data-driven projects, their chance of success will likely diminish because of lack of adequate funding and resources.

“Lack of buy-in from the top can kill a data-driven project before it starts,” says Scott duFour, global CIO at Fleetcor, a provider of business payments services. “I am fortunate that isn’t a problem at Fleetcor, as I get buy-in for projects from our CEO by partnering with leadership running lines of business to validate the importance of big data for company growth and success.”

To get executive buy-in, technology leaders must be able to articulate from the beginning what the outcomes of data projects will be and align them to business priorities or pain points, Clifton says. Ironically, all digital-related deployments depend heavily on data to achieve benefits, “so whether or not the executives realize it, they are funding data initiatives,” he says.

The organization’s data strategy should inform executives about how data projects can support the goals of the business. “The data initiatives should focus on the accomplishment of those objectives through actionable intelligence and automation,” Link says.

In some cases, the lack of support might stem from the fact that business leaders do not really know what they want from data projects, and therefore do not understand the value, Smith says. “If they cannot see the value, then they won’t support it,” he says.

It’s a good practice to use small proof of concept opportunities to show the value through operational dashboards or the automation of manual tasks, Smith says. “This will create interest from the executive team,” he says.

6. Lack of appropriate skill sets

The technology skills shortage is affecting nearly every area of IT, including data-driven projects.

“Without enough IT talent and people with the right skill sets, it’s tough to get data-driven projects done,” duFour says. “And the IT employee shortage is real in several areas of IT.” To try to draw technology workers, Fleercor offers flexible working arrangements and provides training so employees can improve their skills.

“We have also cast a wider net in the talent search,” duFour says. “Although a four-year degree or more is ideal, companies should look for potential employees with associate degrees, IT-type certifications, and other pertinent skills that can help move data-driven projects forward.”

Hiring talent with the specific technical experience needed to lead and manage data-driven projects “is a challenge in this competitive job market, but it’s key in ensuring you have the right skills in place to successfully implement the projects,” Clifton says. “Without the right skills and expertise up front, companies can start a project and then run into issues where the team is unable to quickly and effectively identify and resolve the problem.”

Data scientists, data stewards, and data forensics experts are becoming mainstay roles, Clifton says, whereas data architects were the higher-end skills most needed in prior years.

“Affordable talent has been my biggest challenge,” Link says. “There is no one right answer. I have brought in fresh talent from recent graduates and invested time, only to have them poached at crazy salaries. In my experience, there is a lot of value in having people co-located for faster learning and collaboration. My latest approach is to work with organizations like Workforce Opportunity Services to build my own team from high-caliber workers. It will take time to get there but we are focused on the long-term results.”