Preview releases are among the clearest ways an open-source community can shape the future of a database before it becomes a production reality. They give users early access to new features, a chance to validate upgrade paths, and an opportunity to catch issues while the change is still inexpensive.
In our recent survey, we asked a simple question:
“Have you used a MariaDB preview release?”
The responses tell an interesting story about awareness, experimentation, and where the MariaDB community still sees friction.
Compared with the last polls, we had a slightly smaller number of participants.
The results show something that will probably feel familiar to many people in the MariaDB ecosystem: most users are aware that preview releases exist, some are curious enough to try them, but only a smaller group uses them in a more structured way.
That is not really surprising.
When you work with databases, especially in production, you do not test new software just because it is there. You test it when you have a reason, when you have time, or when you believe the value is worth the effort.
The topline result: most respondents have not used a preview release
The biggest single response was straightforward:
- 39% said “Yes — out of curiosity”
- 52% said “No”
More than half of the respondents have never used a MariaDB preview release. At the same time, a very large group already tried one…
Curiosity is the dominant driver
Among those people who have used a preview release, the strongest reason by far was:
- “Out of curiosity” — 39%
That matters because it shows something healthy about the MariaDB community: users are engaged enough to look ahead. They want to see what is coming, experiment, and stay informed.
At the same time, curiosity-led adoption usually means the product is being explored informally rather than integrated into the normal validation process. In other words, users are interested, but many are still treating preview releases as something to inspect rather than something to depend on.
That distinction matters for the MariaDB Foundation and the broader MariaDB ecosystem. If preview releases are mostly attracting curious users, then they may be underused as a source of systematic validation and feedback.
Functional use cases exist, but they are relatively small
Beyond curiosity, several practical motivations appeared in the survey:
- 3% used preview releases to evaluate new features
- 3% used them to prepare for a future upgrade
- 1% used them in production
- 1% used them to provide feedback
- 1% used them to validate in staging or a lab environment
- 1% used them to benchmark performance
- 1% used them to catch bugs or regressions early
- 1% used them for a customer or internal project need
These are small numbers individually, but together they show that preview releases are already serving multiple roles across the community (12%).
Some respondents are clearly using them exactly as preview programs are intended: for evaluation, testing, benchmarking, and early issue detection. That is encouraging because these behaviors lead to a better release before general availability.
Still, the percentages remain modest. The survey suggests that structured pre-release usage is present, but limited.
Production use is rare — which is probably appropriate
One notable response is:
- 1% said they had used a preview release in production
Preview releases are not meant for broad production deployment. Database professionals are usually cautious, and for good reason. Stability, predictability, and supportability matter.
Still, it is interesting that some respondents selected this answer.
Maybe they run controlled workloads. Maybe they needed to validate a very specific feature. We do not know from this survey alone, but I know some users who are famous for deploying previews in production environments
But it is a useful reminder: as soon as software is public, some users will push it further than expected. That makes messaging around preview releases very important. We need to be very clear about what they are for, how they should be tested, and what users should expect.
A few practical implications
If the goal is to increase the value of preview releases, the survey suggests a few directions:
Make “what’s worth testing” obvious.
Users are more likely to try a preview when they know exactly which features, fixes, or changes deserve attention.
Lower the setup cost.
Quick-start containers, test guides, and upgrade validation checklists can turn passive interest into real usage.
Invite targeted feedback.
Instead of asking for generic impressions, ask users to test specific workloads, edge cases, or upgrade scenarios.
Close the loop publicly.
When preview-release feedback leads to improvements, openly acknowledging it reinforces that participation matters. Could this lead to a MariaDB Foundation Sea Lion Champion Award?
What the MariaDB community can take from this
This survey result is not bad news. In many ways, it is promising.
It shows:
- there is already substantial awareness of preview releases,
- curiosity is high,
- and there is a real, if still small, segment of users engaging in meaningful early testing.
The opportunity now is to convert that curiosity into participation.
That could mean making preview releases easier to evaluate with a coordinated plan, easier to compare against current stable versions, easier to collaborate with recognized benchmarkers, and improving the evangelization to a broader community.
Thank you again to all participants, and don’t hesitate to take part in the new poll: What is the main source of the MariaDB Server you use?