Required metrics for a successful/active forum

(Helder Ribeiro) #1

Is there any study on the general ballpark number of users and engagement numbers that make up a healthy/active forum?

I know this is subjective, and two active people could use a forum solely for one-to-one conversations, but there must be convergence on some kinds of metrics on long-lived forums that could provide some insight.

What are the metrics for a healthy community?
(jcolebrand) #2

You’ve answered the question.

Only the launcher of the forum can determine if it’s successful.

(Helder Ribeiro) #3

The “I know it’s subjective” was specifically to avoid getting that as an answer.

(jcolebrand) #4

Then define metrics for us and let us determine if your metrics are useful.

Nobody has a “this is a success” metric. Never has been. It’s an impossible metric.

If you think you can change the world, then please define that metric and blow my mind.

(Peter Stoinov) #5

These numbers are highly dispersed. I’ve seen forums with less than 100 users that thrive for years and years. And i’ve seen forums gain thousands of users just to become ghost towns in a matter of months.
The number of users is not a very specific metric by itself, to decide if forum is healthy. Maybe active users per month and the general trend of this number will help you to some extent. But then again - this will not give you much insight about how thriving is this forum - maybe you just have a lots of spam bots :wink:
You need to decide for yourself what makes your forum healthy/active and if the engagement is sufficient for your liking.

(Helder Ribeiro) #6

Yeah, that’s part of what I’m looking for. What metrics are relevant, and what values are good. Maybe active users per month, maybe moderator/user ratio, maybe other things or a combination of them.

Area 51, for example, only opens new Q&A communities when there’s a certain number of interested parties, and with at least a certain amount of users with a certain reputation, etc. And they only let new communities live if in a certain amount of time they have X percent or questions answered, a minimum new questions per week, etc.

@codinghorror said somewhere (can’t find it now, don’t know if it was in the FAQ, the blog or somewhere else) that Discourse would not only provide the software, but also guidance (sometimes also baked into software) to help communities grow and be healthy. He and the other founders have great experience on Internet communities (including Stack Exchange, who created Area 51), so they must have some more insight into this.

(F. Randall Farmer) #7

Without understanding your community or it’s goals, we can only reply in the abstract:

The interesting numbers you are looking for are ratios and velocity of change for various metrics - not concrete single numbers.

For example: What is the ratio visitors to posters? What is the reply-rate? What about churn/return-visitor rate?

There are a lot of papers and a few books about the topic. Probably the best new book is:

If you want to dive deeper, check out the works of Marc Smith and Robert Kraut as a starting place.

(Jeff Atwood) #8

The difficulty here is that Stack Exchange success is measured by utility to the outside world, whereas forum success is measured by utility to the discussion participants.

While I do believe (and my Google search results will attest) most forums do end up producing some useful artifacts to the world, only the discussion participants can judge whether they have succeeded at their task of, essentially, entertaining themselves.

Compare, for example, how many times you’ve gotten a search result to a chat log, versus a search result to a forum. This is why I can’t get excited about building a chat system as a core mission.

(Jason) #9

I’ve been a part of a forum that had as few as a dozen active posters at times yet was very successful in fulfilling its purpose for at least the 5-6 years I was active on it. Obviously that is not a typical case but just serves to reinforce the point others have made about how it’s not possible to state any general metrics that you can use.

I think it would be more helpful for you if you were to give details on the type and size of the intended community, maybe along with some basic metrics you’ve considered.

(Helder Ribeiro) #10

I’m thinking of this more from an analysis than a design standpoint. I don’t have a community right now nor am building one, but I’m interested in knowing if there are such metrics and “optimal” values for them given certain desired outcomes.

So, for example, if “people not feeling like they’re talking to the wind” is a desired outcome, one might consider “90% topics with at least 1 reply within 1 day” to be an optimal value for the “average percentage of topics with replies within 1 day” metric, and from Usenet data or some other source, find that there’s a correlation between this and average active users per month, and that ~5k users corresponds to 90%.

Another example would be having “users stick around” as a desired outcome, “average month-to-month active user retention” (as a percentage of those who were active the previous month) the metric, and 40% a good value. Then one might find that this correlates with user-to-moderator ratio and that 1:10 roughly corresponds to 40% retention (less trolling, less user churn).

I don’t know if there’s enough data about online communities out there to allow for this much detail, but there might be some acquired intuition about these things among veteran online community participants/organizers, or at least some anecdotal data from specific cases (“our community only ever ‘took off’ – for some definition of ‘took off’ – once we hit X milestone”).

(jcolebrand) #11

This almost sounds like you’re talking about conversion. Does that seem at all correlated with what you mean?

(Helder Ribeiro) #12

Yep. Conversion is the very general term. You could rewrite the topic’s title as “Required conversion metrics for a successful/active forum”. “Conversion” on its own tends to be used more for user acquisition than engagement metrics, but even retention can be seen as measuring conversions from previous-month users into current-month users.

The problem of finding out which things are best to measure conversion on, and which values are considered good still exists though. Email marketing open/click rates measure a form of conversion, and sometimes companies such as Mailchimp publish industry-wide benchmark numbers so you’ll get an idea if your campaigns are working well enough or if there’s room for improvement. I’m trying to find pointers to benchmarks about forums/online communities.

(jcolebrand) #13

Can you find pointers/benchmarks to IRL conversion rates?

For example: Phone company growth. What defines conversion there?

Adult novelty stores, what do they consider conversion?

Willy Wonka candy bars, do they have conversion?

Your local .NET User Group, do they have a conversion rate? What does it consist of?

Notice each of those has a different goal. They can’t possibly have the same goals for conversion.

I’m curious why you think there is a global formula for conversion rates amongst online communities? When you say “I don’t have a community right now or am building one” then it sounds like mental masturbation, which is at best a fruitless exercise.

I really am trying to guide you, but I can see you’ve decided I’m no help at all, so I’ll shut up after this post on this topic, unless replies are directed at me. But here’s the thing, that you keep ignoring:

The community decides what is useful, usefulness dictates conversion. If you want to facilitate conversion, remove trolls, increase useful content, and create a welcoming environment.

Otherwise, you’ll never see growth. Those rules hold true for every group, every community.

(Adam Davis) #14

Nope, no study, and no convergence. There are things you can tweak and control, but success depends more on how you advertise the forum, the forum topic, and how you moderate it than the number of users. A good small forum can live forever generating lots of great quality content with the right advertising, topic, and moderation. Likewise, a huge forum can thrash around and die even though it has hundreds of active participants due to bad advertising, poor topic, and/or bad moderation.

And the three are interrelated. If you have a great topic, but advertise it poorly or have terrible moderation, you’ll have lots of brief visits, and few people that will stick around.

My limited experience suggests that the best answer to your question is that if you have 15 people that visit the forum and generate content daily, and 100 people that lurk and contribute occasionally, then as long as the other aspects of the forum are good, then it will not die. It may not grow, either, but so much depends on factors other than participation and engagement.

Fewer than 15 daily content generators, and you’re likely to stagnate. I don’t know the anthropological reason, but my WAG is that those 15 people won’t be only friends, but friends of friends, which induces an amount of diversity that the topic at hand will be discussed from more than one perspective, which will incite cyclical response. Good moderation is needed not just to tamp flames down, but to encourage a warm fire that won’t get out of hand. You want minor disagreements. Otherwise it becomes “Opinion!” “yep” “uh huh” “True, dat!” etc. Of course you can have a unified forum as long as you also have one minor troll, but usually that requires more than 15 active users.

If you want hard data, I’d look toward the vast number of books that exist on building community.

(Dan Neumann) #15

I think the most daunting part of a forum is simply hitting that critical mass where you don’t need to create fake activity with your army of fake accounts anymore.

So the % of activity made up by a forum’s creators is an interesting metric over time. Decreases with success, increases with desperation. I wonder how many @codinghorror alternate accounts I’m talking to right now. :baby_chick:

(Adam Davis) #16

He knows!


(Jeff Atwood) #17

Trick question! You are all alternate accounts of mine. Hi boys!

(Chris Hanel) #18

Wow. My whole life is a lie.

Or, wait, is @codinghorror saying that, pretending to be me? Or is @codinghorror actually referring to himself? In the third person? I’m so confused. Or he is. Or both.

I/We need to lay down about this…

(Helder Ribeiro) #19

Should support for sockpuppeting be built-in as a feature for admins who are bootstrapping a new community? Seems to be very effective.

(Cameron Martin) #20

A quick google search returned this: