5 Reasons Data Strategies Fail

5 Reasons Data Strategies Fail

A holistic strategy is critical for data and AI maturity to take hold in a business. The large consulting companies are evangelizing this, and businesses are listening. Fortune 500 companies to SMEs are defining and implementing data strategies.

The results are mixed, with only about a quarter of businesses succeeding in implementing a holistic strategy. What goes wrong? There are 5 main root causes for strategies that don’t take hold:

The key to holistic data strategy comes down to alignment and implications. Data strategy implies a limitation that doesn’t exist. It isn’t confined to the data organization or specific technologies. Holistic means the enterprise. Businesses that treat data strategy broadly are far more successful.

Data maturity and value creation journeys start with small data teams. Leaders are hired based on a manager’s or director’s profile. That approach makes sense based on where the team is and the near-term expansion plans.

A director is rarely a strategic leader. They meet the need to implement a data strategy but are not ready to participate in its creation. At that career stage, most leaders are unaware they need to be part of the process and do not understand it.

The result is a data leader who tries to pull executive leaders into managing the technology instead of value creation. They say “strategy” but immediately describe tactics like hiring, investing in infrastructure, and new projects.

Businesses usually look at the data team’s leadership as something to fill in as the team grows instead of a critical success factor from the start. No matter what stage of maturity, the data team needs a strategic executive leader.

Digital transformation is just the beginning. Data, analytics, and models come right after traditional software. Viewing transformation as one-time or having a finish line is common, and the mindset hobbles the data strategy. Why?

The business must change to adopt traditional software for automation and products to succeed. The company must continue its transformation journey to thrive with each subsequent technology wave. Maturity is an iterative process in which the last step sets up for the next.

A holistic strategy requires the business to adopt a unified transformation strategy, roadmap, and timeline. For companies that struggle to implement data strategy, the data team and individual business units transform in isolation from each other.

As a business unit adopts data products, it matures to leverage them. In isolation, each business unit matures at a different rate and uses different processes. Duplicate infrastructure and tools are another symptom.

Processes, best practices, and lessons learned are not shared across the business. After a year or two of operating this way, the business’s transformation journey is disjointed with leading and lagging business units. The duplication and inefficiency must be rolled back before any more progress can be made.

A holistic strategy reveals the need for a training strategy and plan. Data, analytics, and models will touch every part of the business. When people are unprepared, adoption stalls and value is slowly realized.

In businesses struggling with data strategy, training happens after data products are delivered. The reactive process leads to data products sitting on the shelf unused. Requirements are not provided or are poorly defined, and products don’t fit business needs.

Transformation is continuous, and so is literacy training. Before each stage of transformation is rolled out, people at every level must be trained. Executive levels must be trained prior to the strategy planning process. Mid-leadership must be trained before they are asked to implement. Front line employees must be trained before they adopt and execute.

The strategy should be evangelized and personalized to each team. People must know more than just the technology and game plan. They should understand how this impacts them.

Uncertainty is a massive barrier to data strategy success. Teaching the plan and implications reduces one side of that uncertainty. Data and model literacy training addresses the other side.

During the strategy planning process, the business decides what opportunities to pursue based on all available options. Goals are set, and strategy is implemented. A holistic data strategy is an extension of that process.

Data strategy planning begins with those goals and asks, “How should data, analytics, and models support the business?” Answering that question allows the business to select initiatives based on all available technology options to achieve the business goals. Data strategy and implementation align with the business so data products can be used to produce value (automation and decision support) and deliver value (features and products) to customers.

Those struggling with data strategy evaluate initiatives in isolation from the business’s strategy planning process. Initiatives are selected based on the highest value opportunities that the technology enables. That approach sounds great, but those initiatives can conflict with the business’s overall strategy.

Initiatives that aren’t aligned with how the operating model is built to produce value result in data products that cannot be integrated into existing workflows. Products and features do not align with how the business delivers value to customers. They cannot be integrated into existing product lines, don’t fit current customer segments’ needs, and customers are unprepared to adopt them.

I have outlined a lot of change and investment required to support data, analytics, and models. The value proposition must be compelling, or hype is the only thing driving data strategy. For transformation to be sustainable, the business needs small short-term returns and much larger long-term growth.

Data strategy must strike a balance between innovation and incremental initiatives. When the data team dictates strategy alone, nearly all initiatives are long-term innovation projects. It’s more risk than most businesses are comfortable with.

When the business dictates strategy alone, nearly all initiatives are short-term quick wins. There is insufficient revenue or cost savings to support increased investment and the pain of transformation. It isn’t enough risk to generate significant returns.

A 70% short-term and 30% innovation mix is appropriate for most industries. Technical leaders and high-growth companies must adjust that mix to fit their risk/return profile.

As I’ve mentioned, continuous transformation means today’s change must set up for tomorrow’s. A short-term mindset implements the best solution for current needs. If those don’t align with the next technology wave, the cost and level of effort go up significantly. The best solution also becomes a massive barrier to progress.

The term holistic is the perfect framing to think about data and AI strategies. There was time to respond during past technology waves, but that’s not the case now. Transformation must be optimized for cost, effort, and time to value. Alignment comes from a holistic strategy and speeds up the business’s transformation rate. Technology is a competitive advantage across industries, so the transformation rate metrics have become significantly more critical to business survival.

Images Powered by Shutterstock