For years, database changes have been the black sheep of DevOps pipelines. While application deployments sailed smoothly through automated CI/CD workflows, database updates remained stuck in the stone age-weekend deployments, crossed fingers, and emergency rollbacks at 2 AM.
The latest State of Database DevOps Report from Liquibase paints a different picture for 2025. Based on survey data from organizations across 25 countries, we’re seeing a fundamental shift in how teams approach database change management. The most mature organizations aren’t just keeping pace with their application counterparts-they’re setting the standard for what modern data infrastructure looks like.
AI Workloads Separate the Wheat from the Chaff
Here’s where things get interesting. The AI/ML boom isn’t just creating opportunities-it’s exposing which organizations have their database house in order.
Early-stage DevOps teams are getting crushed. A staggering 25% of organizations in the “Initial” stage (traditional environment with Dev and Ops separate) cite AI/ML and data pipeline complexity as their biggest database management challenge. This drops to 15% for both “Managed” and “Measured” levels as DevOps maturity increases.
Meanwhile, mature DevOps organizations treat burgeoning data pipelines and AI/ML workloads like business as usual-only 10% report them as major challenges.
“With the sudden surge of AI/ML initiatives, there’s been an increase in the need for process modernization and automation for scale and repeatability,” explains Pete Pickerill, Liquibase Co-founder and Head of Developer Relations. “At Liquibase, we’re seeing our user base change as a result of these initiatives-with growing adoption of platforms like Databricks that are built for AI.”
The mature teams have moved beyond basic concerns. They’re focused on optimizing visibility into database changes (17% cite this as a challenge) and tightening security and compliance (15%). When your database change management is already automated and integrated, handling AI-scale data throughput becomes just another Tuesday.
The Visibility Revolution
Forget the speed metrics for a moment. The real game-changer in database DevOps isn’t faster deployments-it’s knowing what’s happening.
An impressive 54% of organizations report that improved visibility into database changes is the greatest benefit of implementing DevOps practices. Close behind comes reduced manual workload for DBAs and developers at 53%.
Across the board, organizations cite a lack of visibility into database changes as one of their top three challenges at 43%. The pattern is clear: visibility precedes velocity. You can’t automate what you can’t see, and you can’t trust what you can’t track.
The maturity progression tells the story:
- Less mature organizations focus on applying version control for database schema changes and benefit most from better visibility (Initial stage: 55% cite this benefit)
- Mature organizations leverage CI/CD database deployment automation and reap rewards from reduced manual workload (Measured: 54%, Defined: 63% cite this benefit)
Security Takes Center Stage
If there’s one theme dominating the 2025 database DevOps conversation, it’s security. Nearly half (48%) of respondents cite managing security and compliance as their most pressing challenge-the highest among all concerns.
This concern is reflected in emerging trends. Forty-three percent of organizations are adopting enhanced security and compliance automation as part of their development and database workflows. Meanwhile, 40% aim to improve their enhanced security and compliance posture through database DevOps initiatives.
Organizations with mature DevOps practices are running more automated checks for quality, security, and compliance (77%), resulting in improved security posture (77%).
“The emergence of specialized security scanning tools that seamlessly integrate with dev environments and CI/CD pipelines has been an enormous game changer in information security,” the report notes. “Teams can now continuously validate security throughout the development process-no more last-minute surprises and human error.”
The Complexity Conundrum
Here’s the reality check: implementing database DevOps isn’t simple. The complexity of integrating database workflows with DevOps practices stands as the single biggest barrier to adoption at 54%-even for the most mature organizations.
But there’s a deeper issue lurking beneath the surface. The skills gap is real and growing. Forty-three percent of organizations report that lack of skills or training for modern tools represents one of the top barriers to adopting, implementing, or expanding database DevOps.
When it comes to concerns about automating database change management, lack of team readiness or training for automation overwhelmingly tops the list at 30%.
“DevOps is equal parts people, process, and tooling,” Pickerill emphasizes. “Time and again, I have seen organizations try to save money by skipping training, declining services, or forgoing hiring to shore up skills gaps. In the long run, they end up spending more in opportunity costs and failed execution.”
Productivity: The Ultimate Goal
Strip away the buzzwords and technical metrics, and database DevOps boils down to one thing: making teams more productive.
Half of survey respondents aim to improve developer productivity and team efficiency through database DevOps initiatives. This ranks as the top business metric they hope to improve, followed closely by reduced downtime and improved system availability (44%).
The technical metrics they’re pursuing tell a similar story:
- Throughput of database changes (47% prioritize this)
- Deployment frequency (42%)
- Mean time to recovery (40%)
- Developer satisfaction and productivity (40%)
Organizations are backing up these goals with concrete tooling investments. Most report using database management tools like SQL Server Management Studio, DBeaver, and pgAdmin to author database schema changes (36%), while 54% leverage CI/CD database deployment automation and 53% use version control for database schema changes.
Looking at emerging trends, 47% are embracing CI/CD integration and automation, while 44% are shifting toward SaaS DevOps and Database-as-a-Service tools.
The Automation Reality Check
Many organizations hope DevOps will provide a quick solution, but the reality is more complex. As Pickerill notes, “The first problem that DevOps solves is a cultural one. Every team member must have a deep understanding of the entire delivery process, from concept to production.”
The focus on collaboration bears this out-27% of organizations prioritize improving alignment and teamwork across developers, DBAs, DevOps, and other roles, followed by safety considerations like error prevention, compliance, security, reliability, and uptime (19%).
Database DevOps by the Numbers
The scope of transformation becomes clear when looking at Liquibase’s growth as a bellwether for the industry. The platform has achieved remarkable adoption: 5,000 GitHub stars, more than 100 million lifetime downloads, and 66% average annual growth over five years. In 2024 alone, it was downloaded more than one million times per month and performed 1.4 million operations daily.
Docker downloads show particularly impressive acceleration-95% year-over-year growth for four consecutive years, with Docker becoming second only to Maven downloads, which exceeded six million monthly downloads from June 2024.
Looking Ahead: The New Database Reality
The 2025 State of Database DevOps Report reveals an industry finally maturing. Database change management is shedding its reputation as the weak link in the DevOps chain. Organizations that invest in proper training, embrace automation thoughtfully, and prioritize visibility over pure speed are seeing real payoffs.
For teams still struggling with manual database processes, the path forward is clear from the data: start with visibility and version control, build confidence through automated testing and compliance checks, then scale with CI/CD integration. The organizations that master this progression won’t just survive the AI-driven data explosion-they’ll thrive in it.
The weekend deployment era is ending. The question isn’t whether database DevOps will mature-it’s whether your organization will lead or follow the 43% already gaining better visibility, the 53% reducing manual workload, and the growing number making database changes confidently during business hours.
##
