Insight from online communities: 8. Quick polls

For the final in our series on how to get insight from online communities, we are looking at using what is very much an insight tool but that can be included in any online community: quick polls. Easy to respond to and simple on the community, getting online polls right is actually more difficult than you might think. If you want to get real insight from them, you need to know what questions to ask, and what answers to offer.

Quick polls offer a way to get high-level feedback from your community members on simple quantitative questions. You can understand what people think and can often get feedback very quickly.

There are four steps to make quick polls successful and a useful source of insight:

  1. Define what you want to find out – you have only a quick poll and a limited number of words to explain what you are asking. Define a question that is actually useful to you and that is specific enough so that people understand  what they are being asked.
  2. Choose your words carefully – how you ask the question is very important. You need to be clear, specific and direct. Make sure you are asking only one question otherwise it will be difficult to analyse the results.
  3. Offer specific answers – in a quick poll you probably list a set of answers from which people will choose. Make sure the answers you offer are discrete and different from each other and that you offer all the combinations people will want to choose from.
  4. Use the poll to spark a forum discussion – the poll itself can only tell you what people think. To find out why they think this, you should start a related forum discussion where people can discuss the poll, their answer and the issues it raises.

Quick polls can be a great opportunity to get relatively quick feedback from the community members and real insight into a question that is important to you. It’s important to make sure you make the most of this opportunity and produce data that gives you real insight.

Subscribe to updates from the FreshNetworks Blog

Insight from online communities: 7. Discussion events

So far in our series on how to get insight from online communities, we’ve looked at the kind of depth of insight your can get from profile information, the discussions on the site and the language people use, ratings and voting and from photos and photo-based activities. For the penultimate post in the series we want to look at a different type of activity that you can run in your online community – discussion events.

The nature of online communities typically lends them to asynchronous discussions, with forums often the centre of the community and the most vibrant and popular parts. In fact, this is one of the real benefits of online communities – they foster debates, discussions and support between people who are disparate temporally and geographically. However, sometimes there can be real benefit from getting members of your online community onto the site at the same time to take part in a discussion event.

As an online community matures, you will find that people start to adopt patterns of use. Some people will always talk about and comment on the same subjects, some people will talk in conversations with their friends, and many members will show clear patterns of use. They will go to the community at the same time during the week and will do similar things when they are there. This pattern of behaviour is one that should be capitalised upon from an insight perspective. If you have a group of your members coming onto the site at the same time every week, then this is a great opportunity to engage them in a new way. Rather than having them discussing things asynchronously, use your existing features to run a discussion event.

As with most things online community, it’s best to start small. Watch when people are most likely to be on your site and then advertise a discussion event to match one of these times – a Tuesday evening chat session, for example. Choose a subject that’s topical and related to the theme of the community and invite people to come onto a forum thread and discuss it for half an hour. The first time you might get a handful of people, but persist. Run them regularly and more and more people will come. Before long you’ll find that this is used as a real catalyst for discussions for the rest of the week. You can get a depth of insight from a range of your community members on a topic that you choose at a time that you choose it. You can then help to direct the community on an ongoing basis by regular, targeted weekly chat sessions.

If you want to really maximise the benefit you get from these sessions you should report back to the rest of the community what went on, what was said and what you think of it. You’ll gain a depth of insight and reinforce a sense of community that can really help to continue to grow and develop a community, even when it’s reached maturity

Subscribe to updates from the FreshNetworks Blog

Insight from online communities: 4. Rating and voting

We return to our series on getting insight from online communities with a look at how you can get insight from those members of your online community who may not want to begin posts or be regular commenters in public forums. We’ve looked already in the series at the information you can get from profiling data, discussions and the language people use online. Today we want to look at how you can get insights from ratings and votes.

It’s a commonly accepted fact within online communities that many of your community members and visitors will not want to initiate or publicly respond to public discussions. They are happy reading the content and are important as without all these readers, those who do contribute would have no reason to do so. This behaviour is encapsulated in the 90-9-1 rule: in any community of 100 people, 90 will be readers, 9 will edit and add to content and just 1 will initiate discussions or add new content. The best communities find ways to make the most of each of these types of people, recognising that different people behave in different ways and accommodating that.

Most online communities allow people to vote for or rate content – say that you like a certain post or rate a photo or other piece of content out of five. However, too often these tools are overlooked as sources of insight. This is a shame. For those 90 people out of every 100 who are unlikely to contribute to discussions publicly, votes and ratings are ways of letting them have their say. Making this easy to do and encouraging people to rate or vote for content will maximise the benefit you will get from it from an insight perspective.

Whilst such ratings and votings shouldn’t be thought of as representative of the community, they can capture the collective wisdom of the members. If you want to know how important a discussion is, looking at how many people voted for that thread, or at its average rating, is a way of helping you to understand the mass of opinion. If lots of people have voted for it, or rated it highly, then this is a great sign that it’s a discussion you should be reading and digesting.

In a more proactive sense, you can use voting and rating alongside comments as part of a process of co-creation. Getting people to comment on photos, articles, concepts or any piece of content will capture the opinions from a proportion of your community members. Encouraging them to vote too will allow more people to have a voice.

Voting and rating is often used as an engagement tool in online communities, but it can also be a source of valuable insight. See how people rate the different discussions, or the votes that different pieces of content get. You’ll learn something new.

Subscribe to updates from the FreshNetworks Blog

Insight from online communities: 3. Learn their language

We’ve already looked at the insight you can get from profile data and focused discussion in online communities. For the third in our series of how to get insight from online communities we are going to look not at what people say, but how they say it.

Communities drive discussions, be those the organic discussions that will begin between members, or discussions prompted by questions, content or other stimulus from the community manager. There is a lot that we can learn about the community members from how they respond in these discussions. What they say, the ideas they give and the opinions they express. But often overlooked is to examine the language they use.

There is a great value to seeing and  understanding the language people use when talking to each other about issues, products and brands. Organisations often have no clear idea of the language people use, the words they choose and the way they discuss their product or talk about an issue. It is difficult and has traditionally been hard to really see how a mass of people discuss and talk about what you do. With online communities you get a real spotlight into this, not only the language people choose but how they talk about and describe things to each other.

Observing and understanding this can be really valuable. One of our clients at FreshNetworks was able to identify significant problems in it’s marketing by watching how people discussed their needs and the different products in an online community. When none of the language they used was chosen by community members we saw that there must be a problem, asked the members why they hadn’t used this language and then realigned the client’s marketing message using the language that customers were using. In this case the real insight from the online community was not so much what was said, but how it was said.

Subscribe to updates from the FreshNetworks Blog

How to avoid convergence collision

At the E-consultancy Future of Digital Media conference last week the focus was on two magic words “relevance” and “engagement”. In Ian Jindal’s stimulating and lively talk he correctly pointed out that marketing hasn’t actually changed much, even though where we choose to communicate our messages may have. Consumers are at saturation point so the only thing left is for companies to get better at attracting customers from their competitors. The ‘How’ was the focus of the day. There were three themes:

  1. engaging customers around content using publishing techniques
  2. how to add a layer of social discovery around your brand
  3. finding ways to manipulate data to do our marketing for us

Many Intermediaries, product manufacturers and retailers are developing strategies around multimedia content to engage customers. Thomas Cook showed a particularly impressive video of their new 360 degree marketing strategy, complete with store front touch screens, co-browsing between customer and agent and every conceivable video clip you could imagine to show you what your resort will be like when you get there (which somehow I felt took the discovery out of it). The issue for most companies with this approach is the sheer cost of it and the skill sets needed to become programmers and publishers, as well as focus on your core business. The other issue was that there was one viewpoint that appeared to be missing from the strategy – the customer’s! Thomas Cook is still in broadcast mode.

The jury was out when it comes to engaging customers effectively using social network sites like Facebook. Brands are just not managing to attract the levels of fans they would like and I believe that it’s because people are, well, hanging out with friends and simply not in buying mode. Companies that have developed non intrusive, useful and engaging tools like Mydeco’s Roomshare seem to be having more success. Considering 56% people go to a brands home page to check out information first, it might be better to find a way to encourage people to stay there. For sectors like Insurance, where aggregator sites like Confused.com and cashback sites like Quidco.com have contributed to a reduction of customer lifetime value from 3-5yrs to 1-2yrs, the focus has to be on finding better ways of retaining their customers once they are delivered to their door.

Understanding data is essential to running any business and more importantly what action you take from looking at it. The new kid on the block is APML or attention profiling markup language. It is a common standard to describe your interests, likes and dislikes and how much each means to you (weighting). The idea is that your attention profile is owned by you and is portable so you can decide which websites you want to share it with for more relevant surfing. Currently Amazon is the only company doing a good job of recommending to us what we’re interested in but they own your profile and the process isn’t transparent. In theory it’s a great way to reduce information overload but it feels like a very long way off, if at all feasible. When I think about how I tag my del.icio.us bookmarks with words like ‘cop’ (which to me means community of practice, not policeman) and when I think about how often I change my likes and dislikes, it feels like someone is going to have to do a lot of work to maintain this profile (either me or a very clever programmer). I would love to hear your thoughts on this whole area!

Don’t get me wrong, as data fragments into smaller, accessible pieces, there are many innovative ways to play it back to your customers in really useful ways. Some good examples that add a valuable social layer for customers include the new AMEX Members Know community where members can see the most popular hotels and restaurants based on anonymous purchase data and Flickr’s use of the Exif data hidden in your digital photos about which camera was used to take a photograph which is the served up as a league table of the best digital cameras. We’re going to see a lot more of this.

I tend to agree with Simon Waldman of the Guardian who commented that sometimes qualitative measures like “Why are you on our site and did you get what you came for?” are equally important. Having conversations with your customers might just reveal what they actually want from you!