A few weeks ago, we stopped talking and started listening.
We put together a community feedback survey and asked Meltano users to tell us what’s working, what isn’t, and what we should prioritise next. These are the people running pipelines in production, building connectors, and debugging failures at inconvenient hours. Fifty-five of you responded with honest, considered answers that have directly shaped how we’re thinking about the road ahead.
This post closes the loop. Here’s what you told us, what we took from it, and what we’re committing to do.
Who Responded
The community that showed up reflects the community we’ve always known: mostly data engineers and team leads, with a healthy mix of developers and DBAs. The majority of respondents are running Meltano in production, and most have been doing so for one to three years or longer. A few were new to Meltano. A handful told us they had tried it and moved on.
That last group matters just as much as the rest. We want to hear from people who left, not only from people who stayed.
What You Told Us
Connectors are still the biggest friction point
If there was one signal louder than everything else, it was this: you want more connectors, and you want the ones that exist to be properly maintained.
Eleven respondents named “more connectors” as their top request. But the open-text responses pointed to something more specific than just quantity. The frustration is with quality: buggy implementations, outdated APIs, conflicting Python versions, and no clear signal about which connectors are actively maintained and which ones have been effectively abandoned. One respondent put it plainly: “Getting the connector ecosystem stronger could make this a serious contender.”
That is exactly the kind of feedback that shapes priorities.
You want production confidence, and you want it to be the default
Eight respondents named enterprise-ready features as a top priority. Reading that alongside the open-text responses, the real message underneath it is about trust. Teams want to know they can depend on Meltano at scale, that it handles governance and auditability, that it is stable enough to be a long-term platform choice. They want it to be dependable by default, not something that requires heroics to keep running.
One respondent was direct about a concern that surfaced after a recent announcement: “Acquisitions of open-source projects can make users anxious if intentions are not clear.” That is a fair observation. We will address it not just in words but in how we build and communicate going forward.
Documentation is good. Good is not the standard we’re aiming for.
Most of you rated the docs at three or four out of five. That is not failing, but it is not the experience of someone who feels fully equipped to run Meltano in production either. The open-text responses made clear what’s actually missing: real-world production examples, architecture explanations, best practices guides, and troubleshooting walkthroughs that reflect how teams actually use Meltano. Not reference pages. Not “hello world.” Practical guidance for the problems you actually run into.
Debugging takes too long
A consistent thread across responses was the time spent figuring out why something failed. Stack traces that point to the problem without helping you solve it. Trial and error as the main debugging strategy. Six respondents specifically called out developer tooling and better observability as a priority. Getting from failure to fix should not require that much effort.
The community is one of Meltano’s real strengths
This one stood out. When asked where they go when stuck, most respondents named Slack and GitHub. Many described feeling very supported by the community. That is something the community built, not something we handed you. Our job is to hold up our end of that relationship.
What We Took From It
Reading 55 open-text responses gives you a clearer picture than any bar chart.
A few things came through that the numbers alone would not show. People care about the direction of this project, not just the features. One respondent wrote: “Please don’t try to overcomplicate it. Go back to what made this project so great in the first place.” Another asked Meltano to “do one thing really excellently.” That message appeared more than once, worded differently each time. The desire is for Meltano to be outstanding at data movement, not spread thin trying to be everything.
We also heard that the barrier to contributing is higher than it should be: “Lowering that barrier could bring in many contributors.” That is fixable. We should fix it.
And then there was this, from a solo data engineer in France: “Keep building, you’re the best.” Sometimes a single sentence in a survey reminds you what this work is for.
What We’re Doing About It
On connectors: We’re establishing a connector maintenance programme with clear ownership, a regular update cadence, and a transparent deprecation policy. We’re improving connector discoverability so you can tell at a glance what’s production-ready and what’s experimental. And we’re doing a focused reliability push on the most widely used connectors: better stability, better documentation, and better examples for each.
On production confidence: We’re overhauling how error messages work. The goal is that when something fails, the first thing you see gives you a useful next step, not just a wall of output. We’re also working on standardising logging and pipeline state visibility so you spend less time guessing and more time resolving.
On documentation: We’re shifting our documentation philosophy from reference material to operational playbooks. That means publishing real end-to-end production patterns, adding architecture explanations that show how the pieces fit together, and building out a best practices guide for real deployment scenarios. The bar we’re holding ourselves to: if a new Analytics Engineer reads the docs, can they get something meaningful running in production without hitting a wall?
On enterprise readiness: We’re defining what “enterprise-ready” actually means for Meltano. Not a vague aspiration, but a concrete list covering security, governance, auditability, and support. We’ll publish a roadmap. And we’ll be following up directly with respondents who flagged this as a priority to understand what they specifically need.
On contributing: We want more of you involved. We’re looking at how to lower the barrier to contributing, whether that’s better first-issue tagging, clearer contribution guides, or dedicated time for new contributors to get oriented.
A Note on Transparency
Several responses touched on a desire for more openness about the project’s direction and what it’s trying to become. Running this survey is one step. Publishing what we heard and what we’re doing about it is another. But it’s an ongoing commitment, not a one-off post.
If you responded to the survey: thank you. If you didn’t but have opinions: you know where to find us on Slack and GitHub. And if something you wrote in that survey is something you’d like to expand on, we’d like to talk.
The Meltano community is one of the most candid and technically sharp groups in open source. That’s worth taking seriously.
Now, let’s get to work.