AI is triggering the most significant evolution in enterprise data strategy since the rise of cloud computing. It’s reshaping how data teams work, how businesses invest, and how value is created from data. New insights from dbt Labs’ 2025 State of Analytics Engineering report highlight how these shifts are playing out, from data teams’ budget boosts and expanded headcount to AI-powered tools transforming daily workflows.
But despite this momentum, AI is revealing critical gaps in how enterprises manage and govern their data. This represents a pivotal moment for organisations to act now to future-proof their organisations – or risk being left behind, or worse, using their data incorrectly and suffering the consequences.
AI is accelerating, not replacing, data teams
AI adoption has surged over the past year, with 80% of analytics professionals now using AI tools in their workflows, up from just 30% a year ago. Use cases range from code development (70%) to documentation (50%), signalling a fundamental shift in how work gets done.
Crucially, though, these tools are not eliminating roles. They’re redefining them. AI investment is leading budget priorities across the board, but data team sizes are growing, not shrinking. Forty percent of organisations increased headcount over the past year. Rather than replacing talent, AI is amplifying the capabilities of data professionals. Overall, AI is going to help data teams build data products at the speed and quality that they and the rest of their organizations have always desired.
Appetite for (natural language) AI
Human oversight is more important than ever. AI can move fast, but without context or checks, it can make mistakes just as quickly.
To use AI safely and effectively, more organisations are turning to semantic layers and analytics-specific AI tools that help systems better understand how data is structured and used. These aren’t just nice-to-haves. They’re becoming critical for keeping AI-powered analytics accurate, trustworthy, and scalable.
Human oversight is more important than ever. AI can move fast, but without context or checks, it can make mistakes just as quickly.
To use AI safely and effectively, more organisations are turning to semantic layers and analytics-specific AI tools that help systems better understand how data is structured and used. These aren’t just nice-to-haves. They’re becoming critical for keeping AI-powered analytics accurate, trustworthy, and scalable.
The compounding cost of poor data quality
Data quality remains the single biggest challenge facing data teams, cited by 56% of respondents. This isn’t just a technical headache; it’s a strategic risk. In an AI-first environment, poor-quality data leads to flawed models, faulty decisions, and a loss of trust in analytics outputs.
And errors don’t stay contained. One bad dataset can cascade through pipelines, misinform dashboards, and mislead stakeholders. Worse still, when AI systems are trained on that data, those flaws become embedded – and amplified.
It’s crucial for organisations to understand that AI doesn’t make bad data better. It simply makes its consequences harder to detect and more expensive to correct.
Four ways enterprises can future proof their data strategy
To thrive in an AI-powered future, organisations need to go beyond short-term tools and be proactive in defining their data strategy. Here are four strategies senior leaders and heads of data should prioritise:
- Focus investments on measurable AI-driven value
Start using AI for automation where the risk is low and the efficiency gain is high, such as documentation, code suggestions, or pipeline debugging. Don’t chase shiny tools; invest in AI where it delivers measurable impact.
- Design for AI-human collaboration
The future of data operations lies in AI and human expertise working in tandem, not as a replacement. Structure data teams and workflows to combine AI’s speed and capabilities with human judgment. This collaborative approach ensures both efficiency gains and maintained data quality.
- Make data reliability non-negotiable
Data accuracy is a business imperative. Establish data reliability as a foundational requirement across all operations, implementing systems and processes that maintain consistency regardless of increases in volume or complexity. Build in robust observability, enforce documentation, and invest in semantic tooling to keep your data stack accurate, transparent, and scalable.
- Break down silos to drive data impact
Data transformation isn’t just a job for engineers anymore. Analysts and data executives are now actively getting involved, but are often stuck waiting on engineering teams or working with incomplete or messy data. To keep pace, invest in platforms that lower the barrier to entry, foster collaboration across roles, and make high-quality, trusted data accessible to everyone – not just the most technical users.
The opportunity is now
The 2025 report paints a clear picture: investment is growing, AI use is rising, and data engineering’s impact is expanding. But the organisations that thrive in this new era won’t be the ones with the most tools – they’ll be the ones building resilient systems where AI and human expertise work together to deliver trusted data, better decisions, and long-term business value.
For companies willing to act now, the opportunity is enormous. AI is already reshaping how data is used across the business, and those who build a solid, accessible data foundation today will be the ones who move faster and go further tomorrow.
Mark Porter
Mark Porter is CTO at dbt Labs, the pioneer in analytics engineering. Mark leads dbt Labs’ engineering organisation including the development, research and infrastructure teams, supporting mission-critical customers around the world. He also drives the future technical direction of the company – ensuring dbt remains at the forefront of innovation within the modern data stack while customer adoption surges.
