Working Backwards To Find What Members Need

One of my favourite tools is AnswerThePublic.

It’s a simple search tool which lets you drop in any term and see what questions people are asking about the topic, which comparisons they’re making, and what prepositions they’re using.

For example, we can enter ‘online community’ below and see what comes up.

For sure, some of it is unhelpful gibberish.

But if you’re looking to build a base of content articles to attract more people to your community, some of these questions would be a good place to start.

And if you’re looking for questions to ask to get a community started, some of these might be perfect.

I’m Sorry

The words and structure of apologising to a community is pretty easy.

  1. You begin with an unqualified “I’m sorry, I screwed up”
  2. You admit the harm you caused (if you’re not sure, speak to some members and find out).
  3. You explain why you did what you did (this isn’t a justification, it’s simply an explanation).
  4. You identify what you will do to make sure it doesn’t happen again.
  5. Then you live up to the promise.

The hard part is having the courage to take responsibility for apologising in the first place.

Introductory Videos in Communities

Videos are hard and expensive to do well (this video cost around $4k and hours of effort).

I know a dozen or so communities that have created introductory videos. None of them are any good. On a few occasions, I’ve had access to the viewing data, almost nobody makes it halfway through the video.

Generally speaking, people don’t want to stop what they’re doing for a few minutes to watch a video to get started.

If your platform is so complicated you need a detailed video to explain it, you probably need a different platform.

The better use of videos are:

  • A video library of best speeches/talks on the topic gathered by members.
  • Detailed breakdowns of a particular case study.
  • Testimonials from members and staff showing appreciation to the community.
  • A welcome introduction from the new community manager or the community team.

In short, don’t do introductory videos.

Forcing People To Visit The Community

A while back, in an internal community project, the client mentioned they had dramatically increased the number of visits to the community.

How did they do it?

They migrated all the internal documentation to the community platform and forced people to visit the community to see it.

They hoped once people were on the platform they would stick around to browse other content and participate in the community.

Not only did this not happen, it added frustrating hoops the audience had to jump through to access the information they wanted. It caused far more harm than good.

While placing the thing people want at the back of the store might work for supermarkets, it doesn’t work so well with communities. Having to register for a community to access an article (or attend an event) doesn’t make you more likely to participate in the community, it makes you more likely to find the information elsewhere.

Forcing people to visit a community portends a knowledge and motivation problem. It suggests you can create and execute a compelling vision to gain the interest and engagement of members.

Time to go back to the drawing board and revamp the concept of your community.

Insights and Advice

Insights are like unsolicited advice.

If you and your colleagues aren’t receptive to it, you’re going to reject it.

This is the obvious problem with adding an ‘ideation’ feature for members to suggest ideas. If there isn’t someone on the other end eager to receive, respond, and nurture those ideas, the entire effort is wasted.

The real challenge isn’t to gather or identify insights from members, it’s to increase receptivity to insights from the community.

It’s very hard to do this working forwards (i.e. it’s very hard to begin with a bunch of insights from the community and try to make people care about them). It’s always best to work backwards. Begin with your colleagues and find out what insights they would find most valuable.

For example:

  • Would they like to gather rapid feedback on product features, content, and marketing ideas before publishing it to a bigger group?
  • Would they like members to suggest ideas for products and features? What should an idea look like? How can the idea be nurtured to be as useful as possible?
  • Would you want to know what problems people are struggling with – and which are rising in popularity?
  • Do you want to run polls or surveys in the community to find out what they think about any particular topic?

As you can see below, there is no shortage of insights you can gather from the community.

The key is simply to work backwards from what your colleagues want and gather the insights accordingly.

Building From What’s Working

The tendency is to focus on what members aren’t doing (but you wish they would).

For example, “members are attending events and reading content, but they’re not participating”.

In these situations, the temptation is to launch a discussion forum and start pestering members to participate in it.

There are two problems with this.

First, your members might already be telling you they don’t want to participate in a discussion forum. If members aren’t doing something, the problem might not be awareness, it might be motivation. And more pestering won’t solve that.

Second, you’re not building upon the very things that are working. If members are enjoying the content and events, the obvious next step is to have even better content and events. Ask members what they like and dislike about these activities at the moment, what would they like to see in them, and build from there.

Just because members aren’t interacting in the exact way you originally envisioned doesn’t mean you’re not building an incredibly valuable community for both yourself and for them.

Reputations Are Easily Formed And Hard To Shake

A client has been treating their community as a noticeboard for months and now wants members to have engaging discussions with one another. The problem is members still see this as a place to dump information.

An acquaintance has a group of superusers who feel they’re being ignored and not listened to. This came during a period where they were ignored and not listened to. Since then, she’s hosted weekly calls where they can drop in, ask questions, and more. She’s taken time to engage with each of them individually and understand that need. Yet in her latest survey, the number of superusers who feel they are being listened to has barely budged.

Reddit has probably done as much as anyone over the past few years to rid itself of hate, discrimination, and the nasties of the web. But the public perception of Reddit is still largely a few years out of date. Facebook faces the same challenge with privacy today.

This isn’t new information. The incredible Hans Rosling used to point out how our impressions of many countries in the developing world were often 40 to 50 YEARS out of date.

Two important lessons here.

First, you have to be VERY careful in forming the right reputation from the beginning. Don’t leave this to chance. The very concept of the community has to be to foster a very specific kind of motivation. The messaging and story your community tells has to be sharp (especially to your best members). If you don’t create a story for your members, they will create their own.

Second, if you’re trying to change a perception, you have to go way over the top to change it. You can’t fiddle around the edges, make a few improvements, and hope things change. You have to begin with acknowledging and admitting the problem. Explain the thing you want to change, and solicit the contributions of members to change it.

Of course, then you have to live and breathe that story.

Solved vs. Helped

If I ask you where I can find an open locksmith at this time of day and you tell me none are open until tomorrow, you’ve helped me.

…but you haven’t solved my problem (and you certainly haven’t provided me with an acceptable solution).

This is the problem with measuring whether an answer solved a problem or was ‘acceptable’ to the recipient. Often there simply aren’t any answers which can do either. A community and its membership shouldn’t be negatively judged for that.

If your product is so badly damaged it can’t be fixed, the only help a community might be able to provide are recommendations for replacement products. You might not be happy about it, but the answer still ‘helped’ you.

As in the original example, you’ve saved the person the time, energy, and frustration they would experience looking for other options. You’ve removed the uncertainty. The member can make a decision based upon the information they’ve received.

In many communities, ‘help’ is simply about being there for someone, listening to them, supporting them emotionally.

A lot of communities (and vendors) measure themselves by % of questions with an ‘accepted solution’ or ‘solved my problem’. For most, it’s a mistake.

Instead of asking ‘how many problems did this community solve?’ a better question might be ‘how many people did this community help?’

The Peril of Top-Down Approaches To Community

We should know by now the evils of taking a top-down approach to a community.

If you come up with the idea for the community, develop the platform, and begin sending out invitations without having spent 40+ hours engaging with prospective members first, you’re setting yourself up for failure.

This typically leads to ‘life support’ communities kept alive by the occasional pulse of activity from an increasingly frustrated community manager.

It often leads to technology that members find fiddly, confusing, and far less convenient than email.

It often leads to staff who struggle to initiate engaging discussions because they have little idea what members find engaging.

It often leads to dull, templated, content filled with articles members really don’t find interesting.

And it often leads to the community manager sending out invites that fail to spark any excitement or interest.

The better approach is to begin with a blank slate.

Spend time with members and notice what comes up. What do members find frustrating and exciting? What kind of people do they want to connect with? Who do they admire and respect? What kinds of words/language do they use to describe their problems? What technology do they tend to use?

Now you can begin with the platforms members already use. You can invite people to connect with the very people they’ve said they want to connect with. You can use the words and language they use to guide their contributions. Better yet, you won’t need to start new discussions, you can simply continue discussions you’re already having with members.

If you’re struggling for activity, the problem is probably not spending enough time engaging with your members.

Time To Dig In and Build Something

I know someone who has spent the past decade jumping from big brand community to big brand community.

She joins, stays for 12 to 18 months, and then quits complaining she didn’t get the support she needed to succeed.

I suspect having so many big brand communities on her resume is actually a help rather than a hindrance. Yet I doubt she can point back to any contribution she’s made over the past decade and say ‘yup, I built that’.

Here’s the deal, no one building communities today has everything they need.

No one has all the resources, staff, respect they need (or want). No one has the obvious career path from community to where they want to be.

At some point, you need to stop hopping to a new lilypad each year at the first sign of difficulty and start building something that lasts.

You need to win people over. Earn respect and credibility. Increase understanding and awareness of community. And, most importantly, deliver results that get you to where you want to be.

Not as easy as quitting in disgust, but what will you be most proud of in five years’ time?

Creating Guardrails For New Community Behaviors

You can find plenty of communities that invite members to do something non-standard like:

  • Share your story.
  • Share your experiences.
  • Create a case study.
  • Post a best practice.
  • Add your photos to the gallery.

Most of the time, these requests fail and members keep posting short discussion posts as before.

The problem isn’t that members don’t want to do it, but because any new behavior creates a number of questions which the request doesn’t answer.

For example, members will have questions like:

  • Where do I even begin?
  • How long should my story be?
  • What does a good story look like?
  • Should it be written in the first, second, or third person?
  • What are some examples I can learn from?

If you want a new behavior. You need to design the experience to be more encouraging. This would usually include:

  • Highlight a specific kind of story/topic you want to feature each month and give a time limit for members to submit them.
  • Be clear about the length, style, tone, and content of the story.
  • Share examples (you or your confederates have created) which members can follow.
  • Let members submit drafts for feedback/ideas.

Once these things are up and running they tend to take on a life of their own. Members can see what’s been published in the past and copy each one. The problem is getting these things up and running in the first place.

Launching A Bounty Program

Bounty programs offering members rewards can combine extremely well with public communities. You have a lot of people enthusiastic about you (and your topic). You can set a challenge and offer them a reward.

With a former client, we offered $10 to any member who identified an error in the community after a migration had caused some issues. The speed and scale of response was fantastic.

You can easily expand upon this to your main site too. You can set a bounty and challenge members to identify any errors on product pages, in the customer experience, or even tiny things like SEO optimisation (most communities include a few tech-heads).

A few tips for making this work.

1) You need a central system for members to ‘file’ the issues they see. Otherwise, several members might report the same issue. Whoever files the issue first gets the credit. It helps if members can see which issues have been identified already.

2) You need to review and approach each of these quickly and easily. Ensuring consistency of approving issues is the real challenge.

3) You need to cap the limit any member can earn in a month. This should never become a substitute for employment for any participant in the program.

4) You need to be clear about what counts as an ‘individual’ bug. For example, if a website isn’t set up correctly, the same problem might show up on thousands of pages. That’s one issue, not thousands of issues.

5) Trial this in one area first. Target one small area of your community or invite only a select group of people to participate. Make sure there’s no problem in paying members, you have a system that works, and overcome any early challenges before expanding to the entire community.

©2021 FeverBee Limited, 1314 New Providence Wharf, London, United Kingdom E14 9PJ FEVERBEE