Community board structure

(Olivier Le Pord) #1

Hi all,

I am embarking on an examination of our community structure. The community forums have been in existence for >15 years. Some boards were added, others were archived so that we have now over 300 boards catering the product support needs of about 100 products. Some of them have a huge user base, others… not. On top of that, we serve 9 languages (English, German, Japanese, S Chinese, Turkish, French, Spanish, Russian and Portuguese). The boards are grouped by categories and subcategories.

We feel that we have too many boards and we need to consolidate, or archive some. Possibly split some if it makes sense.

2 issues (well, at least):

1 – I am having a philosophical debate with myself while trying to correctly define the problem to solve.

Since most of the traffic comes from Google (lurkers), I wonder if the structure of board matters that much as long as they find what they need. The structure may matter for a minority of users who use a wide array of products and go from a board to another. And it matters definitely a lot to us, the community management team. In other words, a well-structured community and the time invested to achieve this, may matter to just a handful of employees. Is it worth it? Thoughts on this?

2 – I am going to do it anyways, so I need to define how to declare a board healthy.

I am tempted to approach this by retaining a series of monthly metrics like

  • of visits

  • of posts and replies

  • of active users

Additionally define tiers according to:

  • Board age: a 1 month old board is still in a build-up phase, vs 1 year.
  • Board language: I expect less traffic on a Turkish board vs. English just because the audience size is different
  • Type of product: well established in the market or relatively new and “strategic”

Having these tiers would lead to different acceptability thresholds, for example.

Thoughts from you all?

(Sarah Hawk) #2

I think you’re correct – members don’t care. I’m interested to hear why your staff are so strongly invested in the categories.

I know of communities that have done research into the performance and engagement rates of category pages vs activity pages (linked ours as examples) and the activity pages always come out on top. That said, in a product support community that activity may differ. But if, as you say, the majority of traffic is direct and organic then categories don’t matter.

Do you have the ability to dynamically list your categories in order of popularity?

Your rationale around diagnosing health seems sound. I think in these situations we have difficulty being harsh. If a board isn’t being daily is it necessary?

Are you comfortable linking to your community? It might help with context.

I know that @rebeccabraglio went through a culling process recently. Any advice Rebecca?

(Richard Millington) #3

That feels like the absolutely key question here.

What problem are you trying to solve?

…and is changing the board structure the best way to solve that problem?

For me this begins with the hypothesis

1) Members can’t find what they’re looking for.

Then finding evidence to support / refute that hypothesis.

2) What does your data / anecdotal evidence show?

Then deciding what is the best way to resolve the original problem.

Then determining what is the best way to help members find what they’re looking for.

That last point is really critical.

Changing the board structure might help, but in most communities 80% of the discussions are about 20% of the topics.

So you might instead what to identify what brings most people to the community and feature those discussions right at the top? Or create unique guides for them? Thus it will be easier for most people to find what you’re looking for.

If you’re going to do it anyway for internal reasons, that’s fine. But if you’re looking to achieve the biggest result for members, I’d begin with those 3 questions first.

(Olivier Le Pord) #4

Thx @richard_millington and @HAWK . Great points. I will have a meeting with my team mates next week. This should be very interesting!

There are many scenarios to consider when it comes to workflow. We know that most come to the community from Google. They typically land directly on a thread. If they don’t find the right content, they may post a question or start navigating internally and click around on the community (and perhaps post something as well).

Anecdotal evidences tell us that the navigation experience is not great. We have started a usability study with actual users, at the thread level, to better understand possible page design flaws and navigation issues.

Reading your reactions, I am now thinking that this usability is the proper context in which to place the board structure story, instead of a separate / disconnected issue.

I’ll try to share the results of my upcoming discussion :slight_smile:

(Joe Velez) #5

At one point, we had over 350 forums. Today, we have less than 200!!

I started decreasing forums 3 years ago and have plans to continue.

The goal was/is to be better prepare for the increase in mobile readers. At the time, we had very little mobile readers - today, our mobile readers make up 60% of our traffic.


People hate change.

Pitch your idea early and often. Pitch your idea to individuals first hear what they have to say and make a note of it. When the time comes to make a presentation to the team you have answers.

Go BIG but be prepared to pare it down. The goal is to get the team to agree on something (Phase 1).

For example, if you want to remove 100 forums break it down into a 3 phase project. You can learn from the first phase - what worked / what didn’t.

Once Phase 1 is complete the team will realize that it wasn’t as bad as they thought it would be. It’s easier to get them to agree to prune/merge more forums after this.


1 ) Announce plans to your readers. This will provide a place they can post their complaints, suggestions, etc. It will also give you a place to merge into any newly created topics about the changes.

2 ) Mass Move all topics from inactive forum(s) to active forum. Then, prune the forum(s).

3 ) Create Redirects from old to new location. Keep the redirects for 1-2 years. This will help you keep most of your ranking juice.

Over time, these pages will lose their “juice” because the topics are buried deep - which will lead to decreased traffic for these topics. For example, these pages may be found in new Category → page 250 vs Old Category → page 5.

4 ) Highlight Popular Threads - old and new. This will make sure any moved popular topics will continue bringing in traffic.

Make sure months have gone by before you start the next phase.


After all these changes our organic search is still healthy. Search is 80% of our traffic. So, this is a great relief to all of us. :slight_smile:

I mention organic search because if done wrong - too much too quickly - it could hurt overall traffic.


We are reprogramming our readers. We want to change how they do things on the site. Some readers will complain. By understanding why people are complaining you will be able to address their concerns. This will pass quickly.

Be proactive. Each category (with multiple forums) should have 1 main general posting forum. This will protect you down the road whenever you want to decrease number of forums in category.

Mobile is the future. If your mobile numbers are low now it’s guaranteed to increase. Because of this Pageviews and Pages Per Session will decrease. Offset these losses by using social media and other traffic channels to promote popular / engaging topics.

A different approach … We have a U.S. category with 50 state forums. I want to get rid of these - but doing it differently.

I will mass “label” each topic in each state forum. I will even set the system to add labels according to state (GEO) to new topics throughout the site. On new topics posters will have the opportunity to not include state label.

These labels will be hot linked and found within each topic. A click will redirect readers to a listing of all STATE topics. eg. Florida will list all topics labeled with “Florida”. Once the traffic in the U.S. category slows I will start the Steps above. The goal here is to reprogram our readers before the change not after. Having less of an impacting on Organic Search traffic.

I did it again - ugh - I hope I stayed on topic :).

(Sarah Hawk) #6

Awesome advice Joe.

I’d add here that it’s important to make sure that people don’t expect that just because you ask for feedback that it will all be followed. Communities aren’t a democracy, although members often feel that they are. People have a right to get angry if they feel ignored.

(Olivier Le Pord) #7

Many thanks for this very practical guidance. Definitely helpful!

(Olivier Le Pord) #8

Good point @HAWK. They are not a democracy, and yet… bears strange resemblance. Everyone has a voice and can be heard. But not everyone makes decisions. Community managers are not elected, etc.

OK, I’ll let the philosophy for another day.

(Sarah Hawk) #9

Right, and that’s why I think people get upset. They feel like the place is a democracy. They are asked for opinions and feedback, they are kept in the loop with decisions, there is often a democratic structure. But the reality is that [most] communities need to be run like businesses. That’s why the honesty around the solicitation of feedback (and how it will be used) is important.

(Rebecca Braglio) #10

Why yes, I did! My recommendations didn’t get implemeted before I left, but I was able to go from (I can remember the number of boards but I think it was 50+) down to 4. Four.

I’m looking at this challenge again in my new job, but this time the communities are very young, which is good.

(Sarah Hawk) #11

So if the communities are young, was this a case of ‘launching with every possible board’ rather than ‘let’s see what people want’?

(Rebecca Braglio) #12

it was let’s think of every possible board. Many of them make perfect sense, just put up too early.

(Olivier Le Pord) #13

On this initial topic: after reading what you guys had to say and talking through this with stakeholders, we decided to reduce the number of boards and avoid a desert town affect of the small boards. I am pretty much following the advice from @Joe_Velez (thx!). I am trying to go from 300+ to possibly half on this, getting rid also of subcategories of boards.

  • Facilitate internal navigation
  • Encourage engagement (“Ah, some activity, here. I’ll get an answer somehow…”)

(Sarah Hawk) #14

Good luck. I’m keen to hear how this works out for you.