5 Strategies to Overcome Common Pitfalls and Succeed in AI Projects

According to research firm Gartner, nearly two-thirds (63%) of organizations are uncertain about their data management practices for artificial intelligence (AI), leading to a predicted 60% abandonment rate of AI projects by 2026. To combat this, industry experts have identified five key strategies to ensure data efforts succeed and foster meaningful AI adoption. Expose People to New Tools: The evolution of AI has been gradual but accelerating, especially since the release of ChatGPT in November 2022. This accessible technology has opened up AI to a broader audience within organizations. For instance, an airline industry leader, Masters, noted that non-analytical teams are now considering how AI can enhance their productivity and operational effectiveness. His team leverages bots on their data platform to disseminate insights across various departments, making AI relevant to everyone. Answer Key Business Questions: Effective use of AI requires identifying clear business objectives. Masters highlighted that some AI applications, such as dynamic pricing tools and predictive maintenance, offer tangible financial returns. These are easier to justify and implement. However, for less quantifiable benefits, such as improving decision-making processes, his team focuses on pre-empting and addressing key business questions. By leveraging the Databricks data platform, they can combine diverse data sets, like customer surveys and in-flight experiences, to provide insights that drive better decisions. Establish a Unified Approach: The pressure to adopt AI often comes from multiple sources, including tech vendors pushing their latest solutions. Masters and his organization have created a systematic process to evaluate AI tools and proposals. Each week, a multidisciplinary group of specialists—including architects, business analysts, and product managers—reviews technology suggestions. Higher-level assessments by senior executives, such as the VP of customer journeys and engineering, help align AI initiatives with broader business goals and allocate resources effectively. Exploit Your Platform: Traditional data silos pose significant challenges for AI adoption. Virgin Atlantic, for example, has embraced a unified data platform using Databricks' Unity Catalog. This consolidation allows the company to run simulations and predictive models with ease, providing real-time operational insights. During disruptions, the airline can quickly query and analyze data to manage passenger connections efficiently, enhancing customer service and operational agility. Keep Fostering Curiosity: While the technical aspects of AI are crucial, data leaders must prioritize understanding the business and fostering curiosity among their teams. Masters emphasizes that the increasing integration of data platforms means less time is spent on tooling and more on strategic thinking. Data teams should focus on articulating business priorities and encouraging curiosity to identify innovative uses of AI. This approach helps reduce noise and ensures that efforts are aligned with organizational goals. Industry insiders and business leaders agree that these strategies are essential for successful AI implementation. They highlight the importance of a holistic approach that combines technical excellence with a deep understanding of business needs. Companies like Virgin Atlantic, which have embraced a unified and curious data culture, are better positioned to leverage AI's full potential and overcome the common pitfalls of project abandonment. Virgin Atlantic, known for its innovative use of technology, has a strong focus on data-driven decision-making. The airline’s strategic partnership with Databricks and its proactive approach to evaluating and integrating new AI tools demonstrate a commitment to staying ahead in the rapidly evolving tech landscape. By fostering a culture of curiosity and ensuring that AI solutions are aligned with business objectives, Virgin Atlantic is setting a benchmark for successful AI adoption in the industry.