Month: March 2012
There are broadly five approaches to developing a community platform.
1) The perfect community platform. This approach uses the wealth of information about online communities to develop the perfect design. This is the approach taken by most enterprise platform providers.
2) The needs-driven community platform. This approach assess the needs of the target audience and develops the community to those specifications. A lot of bespoke communities follow this route. It's the classic of user experience literature.
3) The objectives-driven community platform. This approach assess the goals of the organization and develops a community platform to match. If the goals are to solicit feedback, facilitate advocacy etc.., then the platform is orientated around these elements.
4) The emergent-approach. This approach launches with the minimum viable platform to facilitate interactions and then develops to suit the growing needs of the audience.
5) The platform-less approach. This approach neglects to use a platform at all. This uses existing platforms and connections to stimulate, develop, or facilitate a stronger sense of community amongst members.
So which approach should you take?
Assuming we're not competing against an existing community of fans, we generally advocate the emergent-approach. It's the least risk, least cost and has the highest probability of success. It lets you test and tweak at high speed. When members want something, you add it.
It's also easier for existing members of a community to tell you what they want based on their own experience in the community, as opposed to prospective members.
However, for organizations that lack resources to take an iterative approach to community platforms (or the authority to do so), the perfect community platform, or the needs-driven community platform can be more suitable.
You can't directly invite people to join forever.
You will either run out of time or contacts. Perhaps worse, you'll upset your contacts with repeated requests to join the community).
This means, you need to encourage growth from existing members. You need to penetrate through the existing connections of your target audience. Academics call this the diffusion model of growth. Every member is a node, the message flows through the network until most of the nodes have turned purple. There are a lot of great strategies to boost word-of-mouth.
But what happens when this source of growth declines? This is where it gets interesting.
Growth can slow for two reasons.
1) Most members with an interest in that topic have already joined.
2) Most members within those pre-existing connections have already been reached.
If it's the former, that's fine. You've won. Congratulations, you've created the perfect community. Optimize your conversion ratio, work on your big wins, move to the mitosis phase and try to become self-sustaining. If you want to growth further, you need to create entirely new communities. That's your source of long-term growth.
If it's the latter, then you've hit a self-induced wall. You've only reached out to a single network of people, now most of them have joined. You need to use promotional tactics to reach new segments of your audience not connected to existing members.
This is the key point here. As your community moves into the maturity phase, you need a combination of word-of-mouth AND promotion to optimize growth
The focus will still be on word-of-mouth, but intense periods of promotion targeted at those outside of your current audience's demographics, habits, or psychographics will ensure you didn't miss potential members.
The key to scaling and sustained growth isn't to have a large number of volunteers with similar job descriptions.
The key to scaling is to to have an increasingly large number of volunteers each responsible for their own areas of the community.
If you want highly engaged volunteers, you need to give them ownership over elements of the community. You don't want a large number with similar, overlapping, roles. You want voluteers who need the responsibility, power, and recognition that comes with this ownership.
Over 10 years ago, I spent 50hrs+ a week volunteering on a community. I had to, I was responsible for a large chunk of it.
From an early stage you need to identify the people that write knowledgeable posts and/or have passion for a particular topic. Invite these people to take responsibility for that topic within the community.
You might create categories/areas of the site for this. You might simply give them responsibility for initiating discussions, responding to discussions, and soliciting contributions from others on this topic.
Take the FerrariChat community below.
Here you want to have a volunteer responsible for each specific model discussion. They stimulate regular discussions, encourage people to respond, create content about what's happening in their area of the community, and recruit people to join.
There are no shortage of volunteer responsibilities. It doesn't solely have to be category-based. We covered a few last week. You can ask members to write an opinion post every thursday. Or a summary of the news beat for their niche every Saturday.
A volunteer can host a weekly live discussion on a specific theme related to their niche.
There is no shortage of areas in your community volunteers can be responsible for, but they do need to feel responsible for it.
You can begin right now. Find one person within the community that has a passion for a particular niche, and ask if they would like to take responsibility for it.
Sharon recently mentioned she had a clique problem.
She had a group of highly active regulars who were preventing newcomers from participating and feeling engaged. How can she fix this?
First, you prove the assumption. What data supports this assumption?
The conversion funnel
Look at the conversion funnel, where are newcomers dropping out?
Is it between visiting and registering? This would suggest members aren't finding a reason to interact and engage with other members. It might be poorly targeted traffic, or a lack of interesting discussions. It may also imply the registration process is overly complicated.
Is it between registering and participating? This would again imply that members who register aren't being guided to make their first contribution. They might lost interest during the process or struggle to find the discussion they wished to participate in.
Is it between their first participation and their second? This usually suggests the response to their first contribution didn't engage them, the notification system is poor, or they simply had one question to ask and not come back.
Is it between their first week and their first month? Between the first month and six months? This suggests something affected their long-term engagement. They didn't feel a sense of community. They didn't befriend others.
You have to measure both activity on the platform to identify where people are dropping out (first contribution, second, fifth, tenth etc…) and length of time (1st week, 1st month, 6 months). Some members are highly active for a few weeks and then vanish.
Is Sharon right?
Now if Sharon's assumption was right, you would expect the drop out rate to high be between the first few contributions, or the first few weeks/months. This is the stage when newcomers can identify the regulars and begin to feel they're being crowded out.
If this is borne out by data, you need to understand why. That means individually interacting with members that dropped out at this stage to ascertain what they felt was wrong. Ask them why they dropped out? Do they mention the clique problem without prompting?
Now you know whether the clique is a big problem (and whether it is your biggest problem in converting newcomers to regulars).
(A clique, by the way, is a group of 2 – 12 members whom interact with each other at the expense of others).
Resolving the clique problem
This is where we design an intervention to solve the issue.
This intervention might be to introduce newcomers to members of the clique. It might be to give the clique there own place ot interact. It might be to talk to the clique directly about the issue and see if they can help. It might be to give newcomers discussion topics. It might simply be to try and form several cliques, instead of just one.
There aren't a shortage of possible approaches here. Each can be tested until you identify what works.
What's important isn't the intervention, but the process of diagnosing a community problem, pinpointing the biggest issue (a clique is rarely as big a problem as most people suspect), and then designing a measurable intervention to act.
When this becomes habit, your community gets much better.
Jones et al. (2008) (don't you love academic referencing?) found empirical evidence that information overload significantly constrained interaction between members.
They discovered that 40 participants interacting within 20 minutes was the maximum number which could be sustained.
As the volume of messages increases. Users are:
1) More likely to respond to simpler messages (shorter, dumber, fun).
2) More likely to end active participation.
3) More likely to generate simpler responses.
40 participants might not be the exact number, especially within forum platforms.
But there is a number…and that number is very important.
Beyond that number the quality of interaction plummets, the number of active members plummets (in favour of fewer, highly active members, posting silly comments to each other).
The challenge is to identify that number and take steps to prevent information overload. Developing sub-groups of friends or interests would help. You can split one popular category of discussion into several and reduce the danger of information overload. Dissipating activity is your best weapon.
You might also tighten the moderation policy, disallow dumb comments, increase the quality of discussion taking place.
You can also turn a major topic of discussion into a weekly live chat for interested members. This might boost activity without decreasing membership. We recommended this to WarriorCats a few weeks ago. Likewise, you might add a chat room to a forum, or a forum to a chat room. Now you can have a relaxed policy in the chat room and a stricter policy in the forum.
Your mileage with any of these routes will vary. What's important is the number. It exists. It might be 40 participants in 20 minutes, it might be 100 in 10. Make sure you identify your number and prevent information overload.
Let’s divide those that come to the community into three categories.
1) Visitors. Those that visit, but are not registered members and don't participate.
2) Lurkers. Those that have registered, visit, but don’t participate.
3) Members. Those that have participated within the last 30 days.
Visitors heard something interesting about your community and decided to check it out. They might visit repeatedly. They’re information-seekers. They quite happily satisfy their information needs.
Lurkers saw something interesting on your platform and decided to register. Yet since registering (or even participating once or twice) they lost motivation to participate. They have reverted back to their information needs. At some stage, the interest that caused them to register was lost.
Members are those whom have made a contribution in the past 30 days. They keep finding something interesting on the platform they wish to participate in. They're trying to satisfy their social needs.
Two thoughts here. First, stop using the number of registered members as your community’s number of members. This is outdated. Use the number of people that participate. By the former definition, most of your members might have completely forgotten about the community yet still be counted as members.
Second, pinpoint the stage lurkers dropped out. Was it after 1 contribution? 2 contributions? 5 contributions? Design interventions at that stage to keep members engaged.
If you know a larger number of members drop out after 3 contributions, or two weeks, you might have an automated message at that stage about something a member can do at this stage. Perhaps you have a ritual after the 5th contributions or after being active members for 3 months. Maybe you give members access to forums for different specialisms or access to the same forums veteran members usually participate in? Maybe after 3 months they can apply to be a moderator in the community?
Or perhaps you simply have a big event every 3 months? There is no shortage of interventions, you just need one that matches the stage members drop out.
A few organizations we’ve worked with over the past year have trouble making the content/discussion mindset shift.
The following scenario is fairly common.
An organization will have a few experts generate activity in the community. The experts will write lots of content. The organization promotes this content to its audience. The audience visits, but doesn’t participate.
Why? Because they just came to read. They're information-seekers.
The organization's experts are stuck in the mindset of giving people information. It’s worked for previous marketing efforts, so why not now?
Because developing a community isn’t a marketing effort.
The skillset (and mindset) required to attract an audience is different from building a community.
In this scenario, you don’t need experts writing terrific content. You need experts provoking terrific discussions. This is easier than writing content (quicker too), but tough for people enamored with producing content to adapt to.
This goes far beyond lazily adding ‘what do you think, let us know in the comments below?’ to the end of news posts.
It means identifying the issues that people care about and initiating discussions about these issues.
Experts are perfectly placed to identify these issues and ask questions about them. These are issues that provoke an emotional response. These might be issues about the audience’s biggest problems/challenges, the audience’s experiences, or their hopes and ambitions.
Resist the temptation to create a question and give an answer in the same post. Ask a question, explain why you’re asking the question, keep it short, then respond later with your opinion. Participate in the discussion, don’t dominate it.
In communities you need to forget about giving people information. You need to figure out how you can provoke a discussion.
The value of lurkers is misunderstood.
Lurkers offer absolutely no benefit to the community. If they don’t participate, how can they offer any benefit? Other members don’t know they exist. When calculating the health of the community, lurkers aren’t included.
A community with 10 active members and 10,000 lurkers is in considerably worse shape than a community with 50 active members and 0 lurkers.
However, lurkers may offer a benefit to the brand.
The ROI of the community with 10,000 lurkers might be much higher. Lurkers may be more likely to purchase from the organization. They may become more loyal. They may be more likely to give feedback and they may tell others about the brand.
Too often we confuse the metrics used to calculate the health of the community with the metrics used to calculate the ROI of a community.