In my team, we spend a lot of time thinking about the nebulously defined concept of "community health". It seems to be one of those things that everyone knows is important, but nobody can really define it.

That made me curious: from your perspective (whatever it is, moderator, community member...) how do you define a successful community? Are there any obstacles that prevent your community from being successful? (I'm particularly curious if you have insights on specific Stack Exchange communities you are a part of). Do you think these obstacles can be solved either by tooling and/or by help from the Community Management Team? What would it take to accomplish that?

As always, I'll be monitoring for the next several days to read/respond.

  • 9
    How did you go from "community health" to a community being "successful"? And if they are, seemingly unlike your premise, not the same, are you asking us to define what "success" for a (SE) community looks like? This seems way too broad a question to be able to properly answer. Like your former question this almost feels like another invitation to point out where things have gone wrong recently, but I don't believe that's the goal.
    – Joachim
    Commented Nov 18, 2023 at 22:30
  • 6
    This should definitely be at least three questions: “What's community health?” “What's a successful community?” “Are there any obstacles, do you think we can solve them, and what would it take?” I recommend editing this question to focus on whichever one of these the existing answers most address; or, if that's not possible without invalidating answers, letting it stay closed and asking the separate questions separately.
    – wizzwizz4
    Commented Nov 19, 2023 at 0:12
  • 7
    What does "community" mean in this context? Does it mean "site" as of the things listed on stackexchechange.com/sites, or is it referring to the social group sharing a purpose and a system of beliefs and values gathering around one of the things listed in the referred URL? Does it refers to the operators of the things listed on the "Users" page each thing listed in the referred URL?
    – Rubén
    Commented Nov 19, 2023 at 1:42
  • 5
    " It seems to be one of those things that everyone knows is important, but nobody can really define it." - are you expecting us to try? Or else how can we answer the other questions? Commented Nov 19, 2023 at 2:11
  • 10
    Seems like an XY question.
    – philipxy
    Commented Nov 19, 2023 at 2:21
  • 2
    Asking broadly what success means is already a pretty broad question, but asking “what is success”, “why isn’t your community successful”, “what tools do you need to be successful”, and “what would it take to get you the tools you need” is far too many prompts in one to expect useful, focused answers. Commented Nov 19, 2023 at 3:04
  • 26
    "As always, I'll be monitoring for the next several days to read/respond." Just saying, you didn't respond to a single answer to your previous question. When I asked if we should expect a response or not, I got no response to that either. Commented Nov 19, 2023 at 6:59
  • 3
    Would we be expecting responses to What we really need to know to understand your community @HolyBlackCat? There was no promise to respond there and, frankly, I don't see a need for a response there. I mean, that question is just asking for info, not opening a discussion so I don't mind having no responses and I can't see what a response could be other than "noted, thank you".
    – terdon
    Commented Nov 19, 2023 at 11:52
  • 2
    @terdon Shrug. Those "no noise" rules are primarily for posts that constitute the "library of knowledge", which those special posts aren't. And rule lawyering aside, SE has reputation of ignoring feedback, and those posts look like an attempt to remedy that (which is respectable), so it would make sense, to, you know, not ignore the feedback received via those posts. "we cannot on the one hand vote to close this" Yeah. VTCing this is outright passive-agressive. Commented Nov 19, 2023 at 13:55
  • 6
    I feel like you’ve asked this before: What makes a community healthy?. Commented Nov 19, 2023 at 14:13
  • 10
    Does this answer your question? What makes a community healthy?
    – Mast
    Commented Nov 19, 2023 at 14:15
  • 11
    We get a question that's similar to the one posted 2 years ago without any indication why the answers provided there do not satisfy OP's curiosity, so proper procedure would be OP explaining why they are not a duplicate. I think they even get a neat tooltip-like header about it after the first duplicate vote is cast.
    – Mast
    Commented Nov 19, 2023 at 17:13
  • 2
    I'd rather close the old question as 'no longer r̶e̶p̶r̶o̶d̶u̶c̶i̶b̶l̶e̶ relevant'; a lot of things have changed since then, and staff is no longer actively monitoring it, though Philippe would still get pinged of course.
    – Glorfindel Mod
    Commented Nov 20, 2023 at 15:14
  • 18
    Please don't say that you'll be monitoring the post to answer things if you won't be monitoring the post to answer things.
    – T. Sar
    Commented Nov 22, 2023 at 16:07
  • 5
    I think the comments here describe/show some things that make a Community unhealthy.
    – dan1st
    Commented Nov 26, 2023 at 1:18

16 Answers 16


Here's what I think makes a community unhealthy

I haven't even mentioned anything that happened further back than a year ago.

TL;DR Making valuable people feel like they're fighting an uphill battle against those who should really value them is unhealthy. I feel like it's a weird time to dream about what healthy looks like. Maybe things should stop being so unhealthy first.

But I do have one ask right now

I'd like there to be more staff working on testing and approving AI-detection heuristics.

Right now, it's just Slate. Don't get me wrong- I really appreciate Slate and the work they're doing, but a lot of the work in moving things along in the approval pipeline is stuck on having to go through staff, and that's a significant bottleneck.

There are several very useful heuristic proposals that are stuck in backlog hell that'd I'd like to see being tested.

I have one particular heuristic proposal of my own which I put a significant amount of effort into acquiring statistical data about, and which I think would be very useful if approved. It's sitting there. It has been sitting there for three months now. It's not that Slate doesn't want to get to it. She gives periodic updates to the heuristics group about her todo-list, and it's on there. But things just keep coming up and she has so much on her plate. All the while (and I am not trying to cast any negative light on Slate- this isn't her fault- this is why I'm asking for more staff to be allocated to this), more and more posts which match this heuristic pop up every day. It's frustrating.

  • 9
    Woah, this is just brutal. Why post this? It seems like an effective way to get Stack Exchange to not want to genuinely engage with us. Commented Nov 18, 2023 at 23:48
  • 33
    @RebeccaJ.Stones Well, Philippe did explicitly ask about obstacles to community health, and how they could be addressed. The eighth bullet point (about feature requests) gets said repeatedly in the comments to feature announcements, and it's inappropriate there – so where else is it supposed to be said but somewhere like this? (Besides, sometimes we say nice things: if we held criticism back when asked for it, how could anyone trust us when we say "this is good, we are satisfied with it"?)
    – wizzwizz4
    Commented Nov 19, 2023 at 0:20
  • 33
    @RebeccaJ.Stones A community unhealthy if members feel they can't provide honest feedback. Commented Nov 19, 2023 at 0:21
  • 23
    @RebeccaJ.Stones Sometimes honesty is the best policy. Especially in the face of what could be impending failure. Its hard to be positive when honestly, we're constantly told we're valuble, and things are done 'for the community' while things are not getting better. Commented Nov 19, 2023 at 0:27
  • 14
    Add to the list: Writing blog posts that vaguely insinuate that the userbase is full of bigots. Firing moderators following questionable processes, such that they are even hampered from responding due to religious holidays, in the name of "inclusion". Constantly seeking demographic and other data from the community to which the company is absolutely not entitled, and the clear sole use for which is to fill out further reports about "diversity" and blame the community for bigotry again. Commented Nov 19, 2023 at 2:14
  • 2
    @RebeccaJ.Stones The quote that begins your own answer: "Are there any obstacles that prevent your community from being successful?"
    – philipxy
    Commented Nov 19, 2023 at 2:40
  • 6
    @RebeccaJ.Stones in case you haven't noticed, I am not shy about making "brutal" observations about the company's conduct toward its community and asking it "tough" questions about transparency. exhibit A: this and this. and for the record, I do want the company to engage with us. I just don't expect facing this presentation of events to discourage someone if they're willing to own up to their failings and try to do better. do I expect that to happen? .... ehhh.... do I wish it would happen? definitely.
    – starball
    Commented Nov 19, 2023 at 9:12

You are asking the wrong question here.

In order for something to be "successful" that implies that there's a goal to achieve, a finishing line to cross. Individuals have goals. Communities just… are. And they'll continue to be until they die.

It's important to remember that a community is an organic thing that is larger than any individual that makes it up. It grows. It changes. It adapts. It evolves. Often not because of any individual goal, but in spite of it.

Gardening analogies have a long history when talking about Stack Exchange sites, for good reason. I will continue that tradition here:

One thing about growing a garden is, a lot of people will look at a meticulously maintained garden full of bright green herbs and red tomatoes in perfect lines and think, "That is a healthy garden." They are unlikely to think the same when they look at an abandoned plot full of weeds and wildflowers, of thistles and thorns, of detritus and decay.

But the truth is, both of these are perfectly healthy. If anything, the weeds and wildflowers are healthier because they continue to thrive in the absence of maintenance, whereas the former requires constant tending.

Are they successful? That depends entirely on what your goal is. Weeds and wildflowers thrive in the prairies, where they provide a diverse and vital ecosystem for countless species. If your goal is the overall health of the environment and the planet, it's incredibly successful.

But if your only goal is to have bright red tomatoes to feed your family in the fall, weeds and wildflowers make a terrible garden.

So asking us what makes a community "successful" is missing the point entirely; success and failure is defined entirely by whoever is tasked to tend the garden, not by the garden itself. The garden is just gonna keep growing and adapting to fit whatever environment it finds itself in. Either that or it will find itself deprived of essential nutrients and just die.

So I'm gonna have to throw the question back at you: How do you define a "successful community"? We used to have a strong transparent relationship with Stack Exchange the Company and a good idea what our shared goals were, but that hasn't been the case for many years now. Many of the goals that we took for granted have apparently been abandoned and replaced with arbitrary decisions we don't understand. We no longer have any idea what your actual goals are, what you're trying to accomplish here, so how can you expect us to know what counts as "successful"?

  • 5
    Hear! Hear! Commented Nov 19, 2023 at 5:07
  • 2
    I challenge the notion that the Stack Exchange "core community" doesn't have a goal. What happened to "make the internet a better place"? or "build a high-quality knowledge-base in the form of Q&As"? or "help each other out when software docs are bad and maintainers are unresponsive" (or some analogous situation for non-software things)? In fact, that was what I was initially going to write an answer post about. But my thoughts took me down a different road.
    – starball
    Commented Nov 19, 2023 at 8:33
  • 2
    @starball Individuals have goals. The company has goals. Even sites can have goals. But at the risk of mixing metaphors, even if everyone's goals are aligned and the garden is pointing in the right direction, the community itself is mostly just along for the ride.
    – goldPseudo
    Commented Nov 19, 2023 at 12:16
  • 2
    I also should probably point out, it's not like there's just one "community" involved. If anything, our communities are a community of communities, all carving their own way in the environment.
    – goldPseudo
    Commented Nov 19, 2023 at 12:18
  • 4
    Right now it feels like they are pulling up our tomatos and planting fancy tulips on the hope of making mad money Commented Nov 19, 2023 at 14:24

I'd say real engagement and a 'sense' of community.

I'd also say it depends on what sort of community you want to build. I get the impression the use of community as used by company leadership is very focused on 'anyone who could be using the network'

I'd say that to best serve the 'broader' communities who might use our sites - we need to foster healthy communities of experts, communities of purpose and communities of people.

I'd probably point at my answer here - and point out that well, we don't have healthy communities because the community structure and what anchors people to the network has consistently been undermined by attempts to meet short term goals. Communities don't respect fiscal quarters and damage to trust takes time to heal.

I'd have three soft metrics for what a healthy community should look like

  • As a community of purpose, folks find value in being here. They are here to acquire or share knowledge. The Q&A sites are our core here, not teams, or whatever genAI project or other fad. Unhealthy main sites would likely affect teams take up, but that's out of the scope here.

    With the public Q&A sites marginalised, people end up going elsewhere, and sites lose their sense of purpose.

  • As a community of experts, there's value in networking here with others in the same field. It would be nice if my Super User experience or Meta.SE experience could help me land a job for example. If not, at the very least picking up useful bits of knowledge from subject matter experts. Many of the experts who used to be on say Server Fault are no longer on the network or peripherally active.

  • As a community of people. I've known of couples who met on SE. One of my local former SE users invited another one to be his best man. Healthy communities are communities of people even online. I think that's one of the main things lost with the focus on product/revenue/profit at all costs.

An unhealthy community also tends to have 'leaders' who are, well, hubristic. I've literally seen someone drive two communities to the ground because he was so convinced 'his' way was right, and in the second, basically replaced a (admittedly somewhat flakey) hub/directory for a community with a 'basic' forum, refusing any help... driving users away.

Leadership means listening and doing. Not enough is done to keep communities healthy, and many actions done in the last decade have actively sickened communities.

We don't have healthy communities because a lot of core users were driven away, people who provided leadership, guidance and empathy were removed and a good many attempts to make things better were, well essentially sabotaged cause "We know better". I'd say the blind focus on products--previously careers, now genAI--is part of the source of alienation, but only because its been at the 'cost' of the sites when it fails. I'm expecting the genAI push to fail - and hoping the fallout isn't too bad. We've lost too much already. I'd also argue that the relationship between the company and community has been somewhat abusive, and while there's bright spots, the company has been sinking back to the exact same behaviours that ruined relations and communities in the past.

While to an extent the folks who lead within a community can do our best, after a while without the support we keep asking for, we get stretched thin. Most of the folks I consider 'my' community are on Matrix, Slack or Discord - and sometimes conversations of value end up staying there cause my folks are burnt out by the actions of SE's management.

I'd love to see people actually choosing to hang out on SE, be it chat, meta or main sites but over time the desire for it for many has been eroded.


This is one factor of many but I think it is important.


The Community must be able to trust wherever they are meeting/whatever (platform) they are using. Similarly, both Community members and the people running the place should be able to put some trust in the Community overall (not necessarily in every Community member but the Community overall)

With SE (When saying SE here, I am referring to the SE network), that means the Community would need to be able to trust you (In this post, "you" refers to SE Inc., not Phillipe or any specific person) and the site-moderators for it to be healthy.

Unfortunately, there have been some things where this trust has been lost (in the context of the SE network - I don't think I need to give examples, there should be many obvious ones). It's easy to lose trust but it's very hard to build it. There are a lot of issues regarding that on SE (not just recently but also previously) and this hurts the Community.

Please, work on rebuilding at least some of the trust you lost. While I (and probably also many other Community members) understand that you want to make money, please don't let SE/the Community suffer for it. For example, you could put some effort into improving the site using things like the Staging Ground (As one of the beta testers, I am biased in that regard) or act on things the Community demands (for a long time). You will probably not get to the same level of trust with that but you might be able to regain some trust or it might be easier for the Community to accept your efforts into GenAI if we get improvements to the public site as well.

Do you (again, this refers to the company and management, not people like Phillipe who I am sure care about the Community) care about the Community? If so, show it. We want to see how much you care about the Community.


Activity levels

While an active community may not be healthy, some communities are unhealthy because they have extremely low participation.

For many small sites, I can identify a combination of factors that probably contribute. For example, some sites are limited by their subject matter (the death blow to Windows Phone SE) and others are limited by the rules they enforce (which can be either arbitrary or important). However, even when none of those factors seem to be in play, there are some sites that just aren't thriving. (Which would indicate that removing all those arbitrary rules from a site that has them may not cause that site to gain all that much activity.)

Look at Writing SE, where I'm a moderator. It has an almost universal subject matter, rules that (I think) can easily be explained, and I can't really identify anything in particular that would be holding it back currently. It's just that there's hardly anyone there, which causes problems (e.g., lack of upvotes and comments). After experiencing an extinction-level event in 2019 that wiped out all the moderators and many of the active users (metaphorically of course; they're alive but no longer on SE), the site just hasn't bounced back. Activity may even be atrophying a little, despite the fact that what happened in 2019 has been resolved and shouldn't be causing new problems. And when you look at the activity levels on much of SE, even the most popular of sites are also decreasing in activity.

Where are the people and why aren't they coming to the site? I'm not sure. We could probably find users on other SE sites who would want to participate here, but I'm not sure how to effectively do that. And I'm much less sure of how to get brand new users to SE.

  • 2
    Communities are tricky that way... it takes a while for them to reach critical mass. If something bad happens and a good portion of the most engaged members move on, it's not just a matter of building up again as if it were a brand new community; there's a hole that has to be climbed out of. For every community member you're aware of that was negatively impacted in 2019, there were many lurkers you aren't aware of that were affected. The effects of 2019 aren't causing new problems... they are still the problem. The SE culture isn't what it was when these communities were founded.
    – ColleenV
    Commented Nov 20, 2023 at 14:22
  • 2
    We went to Discord/Reddit. A lot of folks that enjoyed Stack SE are moving towards other places to get knowledge, help, or community. SE used to be the best place to get info, but now it isn't anymore. I get far more out of a random C# discord group than I ever did out of Stack Overflow. Same goes for the other stuff - there is a Discord Community for everything under the sun, and they usually have a way lower friction to interact with than Stack has.
    – T. Sar
    Commented Nov 21, 2023 at 12:30
  • More so, while discord sucked at building knowledge, with the new features like Forums and Threads, things are way better now. One can easily replicate everything stack does on Discord with very little effort.
    – T. Sar
    Commented Nov 21, 2023 at 12:32
  • @T.Sar "Give up and go somewhere else" obviously isn't the answer I'm looking for. Is it really just friction, and why don't I see more one-off users on Writing then? (I know that many of the old Writing users ended up forming Codidact, but that was to get away from SE, and they're also struggling with extremely low activity levels.)
    – Laurel
    Commented Nov 21, 2023 at 14:26
  • 1
    @Laurel that's not what I mean. I was answering that last question you posed, "Where are the people and why aren't they coming to the site?". We moved on to Discord, and other places. Stack was the best way to find experts, but when I can be on a discord group with 300 writers/coders/RPG players ready to chat and talk about issues, Stack loses a bit of it usefulness. It's more or less the same reason of why people moved away from Orkut and went to Facebook, and now are migrating to Whatsapp/Telegram groups.
    – T. Sar
    Commented Nov 21, 2023 at 14:45
  • @Laurel In a sense, internet communities tend to fracture in small, fast-paced groups instead of remaining as one big, slow-moving group. Unless you're very good at what you are doing and can provide value folks can't find everywhere else, folks will move on to places that they feel more welcome.
    – T. Sar
    Commented Nov 21, 2023 at 14:49
  • @Laurel This happens to everyone, from facebook and Wikipedia (with some wikis being branched out to other places, like Fandom or Wikidot) to World of Warcraft guilds and book clubs. When there are too many people together, they splinter up and regroup according to their interests and affinities, often somewhere else, leaving the original group behind.
    – T. Sar
    Commented Nov 21, 2023 at 14:55
  • 1
    @Laurel To give a more personal example - It's been years since I touched stack overflow for anything code-wise, because I don't need to anymore. When I have an issue with something, reaching for that project's github or the official documentation is usually enough.
    – T. Sar
    Commented Nov 21, 2023 at 14:58
  • @T.Sar That's my own frustration coming through, sorry. Anyway, I think that the problems on SO are very different from those on smaller sites. Except, arguably, these small sites are the splinter groups that were formed off SO (and continue to draw traffic through HNQ), so perhaps we do feel some of the pains as a ripple effect. I guess that also hints at a solution, maybe, which is to advertise more on SO, though that can only effectively be done by SE developers. (Advice for writing documentation on Writing SE, anyone?)
    – Laurel
    Commented Nov 21, 2023 at 15:32
  • 2
    In 2021 when I mention Windows phone.SE had zero activity, a few users jumped at my throat accusing me of deliberately wanting to close the site, and why would I do such a mean thing. One user was offended when I suggested that people should be honest to admit the truth: the site had no activity and there was no justifiable reason for running it. I was suspended for one month. There's one site which was popular and had a lot of activity but after 2019 mods often closed new questions for being off-topic. Answers were/are deleted because unsupported by facts. Result: It's a ghost town. Commented Nov 22, 2023 at 13:06
  • 2
    My point being that sometimes the moderators themselves do not help a site to thrive because they put up so many restrictions and limitations. This, fortunately does not happen on ELL (where you are also a mod) but it is happening on EL&U. There's a huge drop of participation/activity also due to the questions continual decline of quality. There must be another way to welcome newcomers without closing their Qs as off-topic or dupes on first sight. Too many Qs on EL&U are getting closed, too easily and too quickly. Commented Nov 22, 2023 at 13:12
  • 1
    @Mari-LouAСлаваУкраїні Point heard about ELU, but I'm working towards fixing it, which has already had a noticeable effect on close rate. (Arguably the problem on ELU, IMO, isn't the moderators but site culture, which might be harder to change.) I do recognize that some of the rules we enforce (not just on ELU) are arbitrary and we wouldn't be sacrificing much by dropping them, and are even likely to benefit. If you have more ideas, I'll see you on ELU Meta ;)
    – Laurel
    Commented Nov 22, 2023 at 13:51

Long-time user here.

I used to be more active on Stack Exchange - Specially on Worldbuilding, Workplace, Software Engineering and RPG. I'm usually a commenter/lurker most often, but I try to engage when I can to help others, if possible.

My interaction with the Stack Exchange Network has been dwindling down for quite a bit for some time, now. I use it only for personal reasons, and I almost never touch it for work-related reasons, even being a user of Teams in the past.

This "distancing" from Stack happened because of a multitude of reasons, but the most important one is that stack had become more stressful than it is worth - and not because of the community, but because of management.

I've seen beloved members driven away either by changes that were detrimental to the software, or by direct action by the staff. I've seen a couple of hires that are honestly mindboggling, which made me wonder if those in the wheel have tried any sort of background check before putting them in charge of anything. I've seen the rug being pulled from beneath the community's feet with strange policies, broken promises, ignored requests.... It's difficult to be a user of Stack Overflow (Or the exchange as a whole) when the site is this badly mismanaged.

Worse yet is having to use Teams at work when I'm extremelly pissed at the company for what they are doing to the places that I once called home. Suffice to say, we don't use Teams anymore.

What makes it even worse is that this placed used to be THE place to find what you need knowledge wise. Now, however, I have a far better chance finding what I need by joining someone's discord or asking directly on their github. Or, heck - reading the documentation, even. Modern frameworks and languages are leagues above what we had 10 years ago, documentation-wise. We now often have live samples, tools to try out code, and a bunch of easily available complete guides on how to use things.

Stack used to be a beacon of light during a very dark and very foggy era of the internet, where knowledge was hard to find. Now, the times have changed - the skies cleared and the sun rose, and we can easily find our way without relying on that single beacon of light in the dark.

Why I would be coming back, then? What makes this place better than, say, Discord, Reddit, or whatever? Honestly, very little nowadays.

Mostly the other users.

But, as the company actively makes decisions that move then away to other places, the siren song that calls us to leave gets stronger. Each good user, each good CM that leaves or is driven away makes the song louder. Why stay, when the reason of me being here left?

Why stay, if the people I looked after were driven away? Why shouldn't I follow them?

My hype with the Stack Exchange Network has been negative for the best part of a few months, now. I'm still here, hopeful to see change, but... each day that I log in and I see our requests ignored, strange changes pushed forward, wonky priorities and tone-deaf blog posts, the temptation to drop everything and give in to the siren's call gets stronger and stronger.

  • 1
    You are more than welcome to follow along wherever they go. ;) Commented Nov 22, 2023 at 21:40

I am sorry to say this, but this question feels like another time waster.

Let me go thru a small thought experiment.

First: Why are we here asking this?

If we are asking ourselves what an "healthy community" is it is safe to assume that we think that what we have now it is not an healthy community.

Second: What is our Community?

Let's start with another assumption. We are not a group of friends. Actually, many users even hate to be called "friends" by others. Our community is just a bunch of strangers gathered on a network for some reason. There are no other ties between us.

Third: What keeps us here?

If we exclude collateral thing like "I am friend with some other specific users and love to chat here" and "sunken cost withdrawal syndrome" that we can't really expect to work at large we are left with a question: Why should a bunch of strangers that feel nothing for each other want to stick together here? There is only one answer. The strangers have a shared objective and they gain something from being here together.

Fourth: What makes the community "Healthy"?

Considering that the strangers "tolerate each other" just to reach an objective, it is clear that what really keeps the community going is the ability to reach that objective.

Fifth: Are we healthy?

Our group of strangers is actually TWO groups of strangers: the "users" and the "company". Based on what we said before one would expect that to be "healthy" these two groups should also be willing to "tolerate each other" and "stick together". Therefore I will identify three sub-questions.

Fifth point One, Is the users community healthy?

I will be blunt here. Ironically the users community currently is quite healthy. Not only they recognize their objective, but the recent events even managed to make them reflect and polish their understanding of those key values. The strikes should also prove that those strangers have come together closer than they did before, because they fell like they have a shared enemy. You.

Fifth point Two, Is the company community healthy?

I fear there is not need to answer here. The company doesn't appear to be healthy. Between lay off, employees not happy with what they are forced to do and ex employees openly talking against the direction the company leaders are taking... the scenario we get is quite objectively one of a company that has much internal fighting.

Fifth point Three, Is the company and users bond healthy?

I think it should be clear by now: we would not be here asking yourself all of this if it was. The bond between the company and the user base is severely damaged and bleeding

Sixth: Why isn't the company and users bond healthy?

Again, let's think back at point four. We are not zealot that would buy a company product base on their name even if it is bad. We aren't friend or relatives of the company that will stay here because of our ties. We are strangers. And in order for a stranger to do something, that something must be useful to them in some way. It may be an actual quantifiable monetary gain or it may be a "spiritual gain" ("I know I helped someone" for example) but there must be a reason or some sort of gain for some unrelated strangers to be willing to work together

There is therefore only one conclusion left.

The users objectives and reasons to be here and the company ones are not the same

I am sorry, again, but I don't think this is going anywhere. People already told the company multiple time why we are here. Joel original launch announcement is still there. And every site Tour Page starts with the same formula.

[...] is a question and answer site for enthusiastic [...]. It's built and run by you as part of the Stack Exchange network of Q&A sites. With your help, we're working together to build a library of detailed answers to every question about [...].

Therefore, my humble conclusion is that the question you should ask is just one.

What can I do to make my objectives and the users one go hand in hand?

Sadly, the only thing I can make out about "Your" objectives right now is something like "make money with AI, using SO data as training data" so I can't really help you there.

PS: as usual I have to add this line specifying that "You" means "the company" and is not aimed at any specific individual.

  • 2
    "this question feels like another time waster" - no doubt. Kudos for spending time on it anyway. Commented Nov 28, 2023 at 14:11
  • New users who have posted a bit & very new users who are just posting & users who have so far only come after googling are all reasonably part of the community but reasonably not part of the community as you have characterized it. Can you address them & their difficulties in onboarding / adapting / culture shock?
    – philipxy
    Commented Nov 28, 2023 at 23:33

from your perspective (whatever it is, moderator, community member...) how do you define a successful community?

A community people want to be a part of for more reasons than sunken cost or it being the only place where one could hope to maybe get a response from a real human being.

Are there any obstacles that prevent your community from being successful? (I'm particularly curious if you have insights on specific Stack Exchange communities you are a part of). Do you think these obstacles can be solved either by tooling and/or by help from the Community Management Team? What would it take to accomplish that?

At this point? I don't know. The problems Stack Overflow has have been problems forever. As a long-term user who's been here for a while, we have a set of tools we can use when someone posts something that is off topic that in their entirety are overwhelmingly seen as negative to the rest of the community. We're left with just ignoring that people hate us trying to preserve what this site is and continuing to dig that hole deeper year after year... or just giving up. There've been dozens of initiatives over the years to work on the close reasons, to change how close reasons are presented, to coax people into "being nice" or leaving comments more often... but none of it really addresses the hatred that is planted squarely on us for using the only tools we have.

We've just shrugged it off this whole time as people just being lazy, but in the end... if more people are giving up year after year than are joining/participating we're moving in the wrong direction.

What frustrates me about all of this is instead of working on tools to address this... we continue to instead work on segmenting the community, or getting people to click more things rather than work on this image problem.

There aren't less devs out there with problems to solve, nor are we reaching some kind of "complete" state where we have all the answers, people just aren't joining/returning. Maybe some percentage of dummy questions are being answered by AI, sure, and that's a fun thing to point at as a cause of less activity in the past year, but we've had dropping activity since 2014. Adding in our own chatbot won't somehow fix this or alleviate the long-term problems we've had for nearly a decade. It's just another toy.


There are many factors involved with the general SE community, such as trust described in dan1st's answer, with two other important ones for having a successful site community being:

  1. There is a general consensus among the members about most aspects of running the site. Too many significant disagreements can create a lot of acrimony which makes it harder for the community to work together to help make the site successful.

  2. Most major disagreements among members are resolved relatively amicably. Especially in smaller communities, having even several members significantly not getting along with each other can tend to dampen the overall community spirit. This is where having actively involved diamond moderators can make a big difference.

A couple of specific tooling and/or help from the Community Management Team that would make a significant difference with these particular factors are to provide appropriate moderator tools (such as the fairly recent Plagiarism flag and moderator tooling has launched to Stack Overflow!) and to ensure there are always enough active diamond moderators for each site, e.g., promptly organizing and running an election whenever there needs to be more. In addition, have enough Community Management Team members to be able to provide good support, not only for the diamond moderators, but also for ensuring prompt handling of most general community member contact support requests.


Based on the wording of the question, I'm going to assume that a "healthy community" and a "successful community" are the same thing. I'm also going to specifically talk about healthy and successful Stack Exchange communities, which exist to build libraries of questions and answers (and, in the case of Stack Overflow, with Collectives, Discussions and Articles).

I don't think that defining what a healthy or successful community is as difficult or nebulous as the question makes it out to be.

Some aspects of health or success are easily quantified:

  • Total number of open and active posts. This is the overall size of the library and how much information the community is currently curating and managing.
  • Number of new posts. For all sites, this is questions and answers. In the context of Stack Overflow, I'd also include discussions (and responses to discussions) and articles. To build a library of content, people need to be contributing to content. The most basic measure is the amount of content being contributed per unit of time or as a rolling average.
  • Number of posts closed or deleted. It is insufficient to just have contributions. The contributions need to be curated. This is ideally expressed as a ratio of the number of new posts. Not all contributions will be accepted, but the majority should be. Ensuring that visitors understand the nature of the library being curated, the minimum quality bars, and can be onboarded to be contributors and curators.
  • Curation activities. Libraries need curation. Curation activities include edits (perhaps splitting out edits to closed questions as a separate measure), votes to close, votes to delete, moderator flags, up votes, down votes, review queue reviews.
  • Community members. There are quite a few ways to slice participation. Rate at which new people join the community. A raw count of people active per unit of time. Number of people contributing questions. Number of people contributing answers. Number of people who contribute N questions or answers in a unit of time. Number of people performing each of the curation activities. I'd even suggest tag-level participation metrics to understand the actual activity within the scope of the site and if there are participants with knowledge in those tags.

These are the five biggest types of things to monitor to understand the health or success of a community, quantitatively. Some of this data is even available on a today. Exactly what healthy numbers look like will vary by community.

But there's also a qualitative aspect. I think dan1st and starball nail this one. All of the quantitative metrics are about people giving up their time. A lack of trust between the owners and operators of the platform and the individuals who's activities would be measured would drive down participation. People don't want to participate and freely give up their time and knowledge if they don't see it as advantageous. And if you don't have the experts contributing their knowledge, people seeking that knowledge won't see the value in asking their questions.

I would suggest that regular surveys and questions to understand the biggest problems would be the right way to make sure that the relationship is healthy and find problems. However, I don't think you need a survey - there are plenty of actions that are still identified and open - starball's post highlighted several recent items, but even more recently, you still haven't addressed how you intend to work through the backlog of (the, as of the time of writing, 208 Meta SE) posts following the recent layoffs as well as the remaining items from the mod strike.


How do you define a successful community?

Pretty simple, a healthy community is a community where people participate. The most important tendency that affects participation of any community is that users tend to leave at some point. As a result the goal of the community team should always be to make it so that your community has more users who start contributing than users who leave your site.

If you would like to be more precise in the type of users you need, you can do a breakdown by the level of user engagement and see how many users leave or join a particular engagement group.

Are there any obstacles that prevent your community from being successful?

I think the first question to ask here is why do people join a community and keep returning to the site. For a community of practice (any Stack Exchange site is a community of practice) the answer lies in the space of mission and goals of the community. If one cannot articulate the mission to the users, they do not have a clear answer to “why” they should join the community at the first place.

Do you think these obstacles can be solved either by tooling and/or by help from the Community Management Team?

Better software never leads to a better community per se, though it helps a lot to keep a community being good. So it is not about software, software is just the means.

I think the CM team needs to focus on making the mission of the SE communities more visible through storytelling and direct the work of fellow employees to support the narrative the team tells to the community.

What would it take to accomplish that?

Besides a ton of work on reshaping the way the company makes decisions at the moment, I think you will need to invest a lot in educating moderators (create or lease something like this), if you want to grow all SE sites. You cannot grow a couple hundred of communities with the team of a dozen CMs without help of volunteers. To make the volunteers successful in what they do, you need to prepare them for the task.



All answers seem to focus on the main Q&A site so far. However, chat is important as well, and should not be forgotten about.

Chat can be a good place to talk about the site and other related things in a more lenient manner.


I was relatively long deciding, to answer on this OP or not.
I am not maybe an active user for a long, but I am a passive user of SO for years.
I patiently waited to join only after I could write something useful.

From my point of view a successful community is able to correct itself up.
Such as a healthy body is able to cure itself from a disease.

That takes me to thought about revision of a downvoting mechanism.
When I first looked at reputation gains and loses, it surprised me, that you get -1 rep. for downvote but 0 for upvote. That effectively restricts ability to downvote to users with higher hundreds of reputation points, whose are much less frequent than users able to upvote. According to statistics by an order.

When I did some research, I found that I am not the only one confused by the downvoting mechanics.
People complain, that downvoting is often politically driven, presenting options, how to handle that.

I agree with the idea, that downvotes should be conditioned by an anonymous comment, explaining why post was downvoted.
To mitigate "xxxx" and "Because you're a buffoon!" type of comments, they will have to be shown publicly.
It does not duplicate simple comments functionality, because nobody read comments on the first impression, but upvoted posts are preffered.

Please do not hurt me, if you do not aggree with me🙂.
It is relatively possible that I will have a different opinion after several years of active using SO/SE.

  • "When I first looked at reputation gains and loses, it surprised me, that you get -1 rep. for downvote but 0 for upvote" - I'm not sure what you mean. Are you suggesting that users should lose (or gain) reputation when they upvote other users' posts? "That effectively restricts ability to downvote to users with higher hundreds of reputation points" - note that you don't lose rep for downvoting questions, only answers (and you only need 125 rep, which isn't that much). [1/2]
    – F1Krazy
    Commented Mar 10 at 19:26
  • [2/2] "I agree with the idea, that downvotes should be conditioned by an anonymous comment, explaining why post was downvoted." This idea has been brought up and rehashed and shot down more times than I can care to remember. I'm afraid it's just not going to happen.
    – F1Krazy
    Commented Mar 10 at 19:27
  • Thanks for the comment. [1] I mean that it is unbalanced, when you get penalty for downvote and nothing for upvote. And for that reason you cannot effectively downvote answers once you gain that privilege. I think that both votes should be equal - no penalty. Downvote is the same expression of opinion as it is upvote. [2] Ok, I will read that article. And I was thinking I came up with something new🙂. So I just reminded that again.
    – Jerry
    Commented Mar 10 at 20:17
  • I have read that article, which focuses on commenting alogside downvoting, not penalty for downvoter on answers/articles. It seems to me like downvoting is used more as a moderation tool (if I exclude politically driven downvotes), not to benefit users - to help choose the best answer/post.
    – Jerry
    Commented Mar 15 at 10:11

First, I think we should acknowledge and appreciate that this is a very difficult problem. Stack Exchange somehow manages to turn a large number of individual volunteer contributors into a functional entity. This is already a miracle.

Second, these people are not just answering questions, they training future generations and probably future AIs too. The work people do on this site is extremely important.

Third, we want people to feel like they are part of a close-knit community and we want a diverse range of opinions to be represented. These may appear to be opposing goals but they are both important. To make that work, I think we need to find ways to get people talking more to resolve conflicts. For example, breaking off to have a private chat with someone or even initiating a voice call like you can with colleagues at the office. Right now you can't even ask someone "Why did you downvote my post?" There is just not enough open and meaningful communication going on. Lots of people are jumping to the wrong conclusions. Others are acting based on their suspicions rather than facts. This can only lead to chronic and pervasive distrust.

tweet from Jeff Attwood, Stack Overflow Founder A tweet from Jeff Attwood, Stack Overflow Founder

Fourth, in the interest of increasing communication and mutual trust among members, we could reward people more for making constructive and actionable comments and for successfully addressing concerns that are raised by commenters. This might involve decreasing voting button relevance and increasing comment relevance since the voting buttons exchange less information than the comments. People won't stay in the community long if they get hammered with negativity when they post. Or, worse they will stay but learn to only write bland and boring posts that don't add as much value. Or just engage in what is essentially circular reporting. Basically "slow death" activities.

Of course, the way we got this job done was with meetings - big meetings, little meetings, hundreds of meetings! The thing we always tried to do in these meetings was to encourage everyone, no matter how shy, to speak out, hopefully (but not always) without being subject to ridicule. We wanted to make sure we had not overlooked any legitimate input.

NASA SP287, "What Made Apollo a Success"

My recent experience with how some people write comments today:

"That is a myth [referring to the notion that anything launched from a railgun will burn up when it hits the atmosphere] perpetuated by certain people who are interested in defending the status quo of launching payloads with chemical rockets." That is quite an accusation. I would ask for evidence for this assertion, but I won't bother because I know there is none, because this claim doesn't pass the giggle test.

(Note: I did respond to this comment with evidence, which you can see here.)

Fifth, change "vote to delete" to "vote to hide". When someone is actively trying to improve an answer that has received some downvotes, it's unfair to delete it. Let them continue to work on it to make it better. Hide votes should expire after a while if they are not renewed and if there were changes to the answer.

Sixth, hold an open conference so that people meet each other, see each other, and talk about some of the deeper issues we are facing. It could be held online, but it needs to be live presentations with lots of Q&A and break-out sessions.

  • 7
    In essence, you mean make Stack Exchange a social platform where the focus is the users, the people, and not the content. This can have one outcome only: SE losing its one and only advantage being quality of the content, and degenerating into collection of rubbish posts that nobody will use anymore. Commented Dec 2, 2023 at 20:58
  • Heck no! As you probably garnered from my other post, I think that the danger is that the content will suffer if we allow the communications, and exchange of viewpoints, to suffer. Somehow SE has to rise way above "Internet Accepted Practice" when it comes to publishing misinformation. SE content has to be more trustworthy than Google and ChatGPT results, which may seem easy now but will become increasingly difficult moving forward. To get there we'll need lots of open communication, exchange of ideas, collaboration, and acceptance of differing but valid viewpoints.
    – phil1008
    Commented Dec 3, 2023 at 2:00
  • 7
    Sorry I don't understand what you say, but what you suggest e.g. "initiating a voice call like you can with colleagues at the office" is ultra social, even much more than Facebook where most people don't really make voice calls just message each other. Also, here is the typical contents of voice call: "Hi, why did you downvote?!" "Because your answer is wrong" "you're toxic!" "no you are rude!", and two people are even more angry than before. Sorry but it's just a terrible idea, that's all. Commented Dec 3, 2023 at 7:27
  • I always found that conversations tend to diffuse disagreements more than escalate them. But you could be right. Some people on this site might have self-selected for the messaging approach. They might not communicate as well in a conversational setting. But, I was not proposing that voice calls would be forced on anyone. It would be optional and, initially, experimental.
    – phil1008
    Commented Dec 4, 2023 at 2:01
  • 5
    "...we want a diverse range of opinions to be represented..." In general, in Q&As whose themes revolve around information technology and exact sciences, we don't seek opinions; we aim for good questions and answers supported by credible documentation. The use of comments on posts should also avoid a social tone and must be objective, aimed at guiding users, and also curating publications. But, more informal communication can take place within the chats linked to each SE site; indeed, chats are a good space to discuss and learn about rules, usage concepts, curation, and moderation of our Q&As. Commented Dec 4, 2023 at 3:38
  • You should post this question and see what happens. "In C, what does x = x++; do?"
    – phil1008
    Commented Dec 4, 2023 at 8:23
  • 2

Defining a successful community is tricky, but here's my take: a good community is where people feel welcomed, involved, and respected. They share ideas and help each other out.

In Stack Exchange communities, challenges can pop up. Things like keeping discussions helpful, dealing with different opinions, and making sure everyone feels heard can be tough.

We can tackle these challenges with both tech solutions and human touch. Better tools can make things smoother, but it's also about how people manage the community. Clear rules, fair decisions, and listening to what users need all play a big part.

To keep a community strong, we need to keep learning, talking, and finding ways to make everyone feel like they belong. It's a team effort!


Are there any obstacles that prevent your community from being successful?

I'll just raise one particular issue: excessive commenting. This may arise in various ways:

  • Someone has their personal idea of how a site should be run, and copy/pastes a comment whenever they perceive a violation. This results in hundreds, if not thousands of such "don't do this" comments. This is especially frustrating when no consensus is established.

  • Someone feels that virtually no questions (except maybe their own) are on-topic, and no matter what question you ask, and no matter how much effort you put into your question, they'll explain why it's off-topic and poorly written. At the smaller sites, they can leave such unwelcoming comments for every single question.

  • Someone feels they're in a kind of "competition" towards another particular user, and comments on practically (if not literally) every single one of their posts. If you reply, they're guaranteed to argue irrespective of how wrong they are, just to make sure they have the final comment and therefore "win".

  • Someone overestimates their expertise, so leaves comments that they think are helpful, but are totally useless or even time-wasting.

A small number of users, when I see their name pop up in comments, make me groan "oh, not again". In this way, a single user can make me not want to use an entire site.

In my mind, comments should be far more heavily throttled. It shouldn't be possible for a single user to comment on virtually every single post at a site, nor on virtually every post by an individual user. It shouldn't be possible to copy/paste comments en masse. It shouldn't be possible to post many comments without also posting many questions and/or answers. To inhibit arguments, there should be a time delay (say, half an hour) before someone can add a second comment on a post [learn to say what you want to say in one comment then move on], with the time delay redoubling for subsequent comments. And it should be possible to block users.

But if not those features precisely, it would be nice to have some kind of mechanism to combat excessive commenting.

  • +1, most of this answer resonates with my experience with SE. Browser extension to hide all the activities of a given user on Stack Exchange Commented Nov 19, 2023 at 5:11
  • 4
    Your bullets claim commenter feelings, thoughts, attitudes & motivations that are very hard to justify given the small bandwidth of comments & do not allow that it is the repeated appearance of questions with the same poor qualities that merit the same (neutral helpful) comments.
    – philipxy
    Commented Nov 19, 2023 at 5:23
  • 3
    Point number 2 is occurring on several sites I visit, Movies:SE in particular seems to suffer from an excess of downvoting and snarky comments. It's almost as if users don't want questions to be asked on their sites.
    – Richard
    Commented Nov 19, 2023 at 7:35
  • 12
    Any proposals in this answer would basically result in a free-for-all-crap on smaller sites as the most active users there will become unable to moderate posts effectively.
    – Tinkeringbell Mod
    Commented Nov 19, 2023 at 8:53
  • 6
    These are all issues of a lack of actual moderation. Blocking them doesn't make them go away, your choice to not flag, to not make meta posts, and to not moderate the content or manage the user is what makes this continue. Do your job as a user, as a curator, and as a moderator, and the problem will go away.
    – Nij
    Commented Nov 19, 2023 at 10:14
  • 4
    Bullet points 1 and 2 are especially frustrating if that user is also a moderator on the site. Commented Nov 19, 2023 at 12:05
  • @Tinkeringbell Can you give an example of a small site where it is necessary to comment en masse in order to moderate effectively? (Are you thinking Skeptics?) Why would commenting en masse be necessary for moderation (when we can already downvote, close vote, flag)? Nij Excuse me. My job? I contribute plenty to Stack Exchange as it is, thanks. Blocking them would absolutely make them go away instantly (freeing users to do more productive things), and if it were implemented properly, it would provide actionable data for moderation (e.g. a single user blocked by multiple other users). Commented Nov 19, 2023 at 22:40
  • 2
    The point is, are these "mass comments" technically relevant to the respective posts? If not, these comments are spam and should be treated as such. First, warn the user, and in case of recurrence, suspend them. If they are technically relevant, they are just an expression of freedom of speech, a manifestation of individual contribution to the collective. However, obsolete comments can and should be removed as they are no longer necessary. Commented Nov 20, 2023 at 0:39
  • 12
    @RebeccaJ.Stones Well, the sites I moderate, for example. How else is a user going to know the site has a citation-expectations policy and that is why their post is down- and delete-voted? Or that their post is blatantly off-topic? No comments going with down-/close-/deletevotes is one of the biggest complaints around.
    – Tinkeringbell Mod
    Commented Nov 20, 2023 at 7:22
  • 2
    1) the points you raise sound like something that you, as a moderator, should address by warning and suspending folks, when they go too far. we even have a stock mod message for excessive discussion in comments, if you don't have it on your sites, you can ask in TL for a template; 2) especially your point #2 is why I stopped contributing to Chinese.SE. "Anything goes" is a nice recipe to drive people out.
    – blackgreen
    Commented Nov 25, 2023 at 23:28

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .