Building a Data-Driven Culture

Success with AI requires having good data, which requires that organizations create a comprehensive data strategy to support initiatives. That strategy should include key elements, such as a unified data architecture, AI readiness, governance, compliance — and culture.

When it comes to successfully leveraging data, strategy and culture are twin necessities. According to a recent guide from Google Cloud:

"Combining a strong data strategy with a thriving data culture ensures that every team member understands, uses, and contributes to the value of your data — and leads to better utilization of AI. An effective culture creates a shared understanding of using, owning, and creating value from data."

Building a solid data strategy and culture includes focusing on how to enable data consumption aligned with business value to develop outcome-driven AI investments, self-service analytics, and ROI monitoring. It also means promoting data-driven decision-making, incentivising data literacy, and upskilling within the organization.

Without the integration of strategy and culture, leaders might find their initiatives stalled, stuck in the pilot phase, and unable to deliver measurable business results. In fact, a recent study by a leading industry analyst predicted that, by the end of 2025, more than 30% of generative AI (GenAI) projects will be abandoned after proof of concept due to poor data quality, inadequate risk controls, escalating costs, or unclear business value.

How does an organization avoid this fate? A successful AI initiative thrives in an environment that fosters data-driven decision-making, boasts a culture of data fluency, and enables broad organizational access to a wide range of both structured and unstructured data, balancing data quality and data quantity. And it relies on partners and third parties for specialized expertise.

What’s the Deal With Data?

If the AI engine — its models and algorithms — are the brain, then data is the oxygen that keeps it alive. Data, especially the proprietary kind, is vital to the success of any AI initiative. It’s the essential ingredient that transforms a common AI implementation into one that’s tailored for a specific business, its customers, and its ambitions. Lacking proprietary data, most AI initiatives will deliver no more than generic, commonplace responses devoid of depth and insight.

But teams shouldn’t only focus on structured data. A treasure trove of unstructured data lies untapped, ready to deliver new and novel insights. Found in social media, digital assets, customer support chat sessions, and customer meetings, unstructured data is estimated to make up as much as 90% of the proprietary data a company collects and stores. Today’s AI opportunities depend heavily on this unstructured data.

Adding this vast amount of data to an existing structured data set can unlock opportunities to build a competitive advantage. For instance, an organization might combine data from product reviews, customer support chats, or social media posts and integrate that with purchase history to enable GenAI agents to create highly personalized offers. This type of customized experience fosters loyalty and can ignite innovation.

To make the most of GenAI, you need access to ALL your data. As Google Cloud notes:

“More specifically, you need all your data available as input — not locked away in silos or inaccessible. In particular, unstructured, multimodal data, such as videos, images, and text, is often scattered across different platforms and stored in diverse formats, making it difficult to consolidate and implement comprehensive security and governance policies.”

As Google Cloud notes, “copying or migrating data to one place is often not practical nor is it cost-effective.” Instead, they recommend an open approach to data, supported by a unified platform that combines all types and formats of data, including open formats, into a single, reliable source of truth that everyone can use to extract insights and make decisions. With this strategic foundation in place, fostering a data-first culture becomes much easier.

Data Fluency: Beyond Pivot Tables and Price Lists

With this abundance of data, it’s essential that organizations build data fluency. Employees must become adept at finding, accessing, curating, analyzing, and packaging data. In a data-driven organizational culture where AI initiatives succeed, this type of data literacy is nearly ubiquitous.

To begin the journey to data fluency, organizations need to first get their data house in order. Employees need ready, convenient access to a wide variety of data types and sources. In short, organizations must move from a model of exclusive access to one that democratizes data access.

Organizations can foster data literacy by ensuring data is easy to:

• Use: Create standards and systems for users to easily access the right data. Provide ample training opportunities and introduce self-service tools and templates to make data analysis less mysterious and more ubiquitous.
• Track: Provide model transparency, so users can check answers and validate automated outcomes.
• Trust: Protect data with advanced privacy and security measures; ensure all regulatory compliance standards are met.
• Trace: Support “data provenance” — that is, providing clear lines of sight into how the data was collected (from what campaign, product, transaction, or customer interaction).
• Understand: Build understanding through mentorship, examples and training. If an organization lacks the ability to decipher the data, it will experience data-induced procrastination — commonly known as “analysis paralysis.” In these cases, new-found data and data skills will slow down, not speed up, decision-making.

Culture Sets the Stage

Finally: Culture. An organizational culture is generally defined as the set of behavioral norms and unwritten rules that shape how work gets done and how executives make decisions. If an organization’s current environment routinely relies on personal preference or opinions for decisions, it’s time to shift to one that incorporates data into decision-making most of the time.

Tackle this challenge with a two-pronged approach: Tops-down and bottoms-up. Leaders should demonstrate what a data-based decision looks like and set expectations accordingly. Leadership should expect and reward evidence-based strategies and decisions. However, as an organization transitions, it’s important to not penalize those who miss the mark, but instead to coach them to success.

• Provide training and self-service tools and make data available to a broad swath of employees.
• Enable and encourage experimentation and mentor employees to demonstrate how to transform a strategy fueled by opinions into one that’s backed by evidence.
• Prepare for a sustained journey.

Without ongoing, high-profile executive appreciation for data-driven insights, a culture change will be viewed as a temporary fad and is likely to be short-lived. When data is woven into the core of everything an organization does, from daily operations to executive mindset, and when it becomes the common language to describe strategic choices, only then will organizations be able to reap the benefits of a data-driven culture.

Embrace Expertise

With a data-driven culture in place, organizations are ready to take full advantage of the benefits promised by AI. The best results are realized when there’s a strong foundation of relevant, current data, coupled with a curious and well-trained workforce. Achieving that is no small feat and certainly not a project to tackle alone: A recent study by McKinsey revealed that more than 70% of executives point to data management as one of the top barriers to AI success.

There’s support available for organizations hoping to tame their data and implement repeatable processes and AI tools, such as the expertise offered by Tech Mahindra, a Premier Google Cloud Partner, Google Cloud, and OracleDB. This team of industry leaders helps enterprises leverage, manage, and implement scalable GenAI solutions that deliver measurable impact.

Tech Mahindra’s design thinking-led approach and frameworks help find GenAI opportunities, while experts identify the right GenAI strategy and roadmap for each unique organization. Organizations can rely on Tech Mahindra and Google Cloud’s OracleDB-based solutions to help foster data-driven decision-making, cultivate a culture of data analytics, and unlock the full potential of GenAI.