So, if you're following the recent posts, you'll realize we're cruising through another round of the taxonomy debate.
Our team came up with a clever proposal. Our community didn't exactly take to it as we thought.
Which, of course, resulted in an interesting discussion around "what problem are we trying to solve here?" that was perhaps the most interesting learning from the whole experience.
What's The Problem We're Trying To Solve Here?
If a nice taxonomy is an answer, then what's the question?
Turns out that the consensus was primarily around orienting new users. Right now, our "home page" presents an alphabetical list of all communities with no real heirarchy.
Since we have a business goal of getting new users to interact with the platform, it's obvious that a logical, well-ordered taxonomy would engage users in a productive way.
Or would it?
What Behavior Do You Want?
What do we want people to do on the platform?
Is it more like a library, where the goal is to find what you want, as quickly as you can? Or is it more like a bookstore that encourages you to browse, explore and linger?
I weighed in pretty strongly -- the ideal behavior, from a business perspective -- is to encourage people to explore and find things they're interested in that they might not have considered initially.
Put differentlly, let's say my job at EMC primarily revolves around Topic X. Sure, if I go to our platform, I'd be interested in anything to do with Topic X (call this a directed search, for lack of a better term). There's value in that, right?
But, what if along the way, I got exposed to Topic Y and Topic Z and this incredibly cool discussion around Topic G that I didn't know was going on? Call this undirected exploration, for lack of a better term.
Which primary behavior creates more business value?
I think that really hinges on what kind of company you're trying to be. If you're a classic, 1.0 company, it's all about process improvement, and finding the best possible answer (or people) as quickly as possible. Taxonomies and heirarchies might matter more in this world, right?
Or, if you see yourself as a company of knowledge workers, and more value is created when disciplines hybridize (rather than the disciplines themselves), more value is possibly created with the second scenario.
Put differently: sooner or later, I'll find what I was interested in -- but what else did I find along the way?
One example from EMC: we have a security division (RSA), but we also have a very broad business where security discussions matter. If you're interested in RSA stuff, it's pretty clear where that is, but where do you go to find out what everyone else is doing about security?
Well, there'll never be a nice, neat orderly taxonomy bucket for the "everything else" category, you'll have to go exploring, won't you?
Interaction And Unintended Effects
Make it too easy for people to find what they want, and -- perhaps -- they'll be less likely to go exploring.
Or, at least, you can ask them to use some of the 2.0 techniques to find what they're looking for: search, tags, asking someone, looking at user profiles, etc. -- rather than relying on an Independent Authority to put everything in neat buckets.
The cool stuff in life doesn't live in neat buckets, I've found.
I won't go into all the inherent problems associated with taxonomies -- but, I'd be willing to consider revising them if (and only if) we had a model of "desired behaviors" associated with this platform. If we had that framework, we could clinically discuss whether they were a good thing, or a bad thing -- but in the context of what we were trying to achieve.
Actually, We Do Have Such A List ... Sort Of ...
But I don't think I ever got around to writing it down, which might be part of the problem.
Some behaviors are obvious: transparency, honesty, communicative, etc. Others, maybe not so obvious: curiousity, exploration, engagement with people outside your circle, and so on.
And some are maybe so subtle they need more explanation: things like self-responsibility for organizing information, or not expecting that everything on the platform has been scrubbed for accuracy, or encouragement of diverse opinions and perspectives, and so on.
I found myself restating to the team one of the big reasons we're doing this behind-the-firewall platform: to learn new behaviors.
Well, We Still Have The Problem With New Users, Don't We?
Yes we do.
To recap, once you've been on the platform a while, and are an experienced user, there's no real problem in finding out what's interesting to you. Sure, we could do better with a variety of techniques, but it's not a glaring problem right now -- maybe later, but not now.
New users need orientation and engagement. We don't want them cruising over to the site, taking a look at the chaos, and saying "sheesh, that's not for me".
But, if we framed the problem differently -- and kept in mind the desired behaviors we were trying to engender -- different sorts of answers pop out.
A Sampling Of The Creative Ideas That Resulted ...
One idea was simple sending a "welcoming email" to someone who'd just registered. The email could be personal and warm, encourage people to explore a bit, and have a set of links to basic orientation materials.
We've also got Clearspace 2.0 in house -- but not in production yet -- and we've got all sorts of interesting new ways of customizing the home page, including "popular discussions", "most popular blog posts", etc. -- basically, getting people engaged in a way that makes them want to explore, rather than helping them to immediately seek out their comfort zone.
Another idea was multiple taxomonies -- put several different ones out there, and let people submit their own. If people want taxonomies, fine, have at it. Just don't ask for the Authoritative View.
And yet another idea was a weekly newsletter or blog post, e.g. "This Week On EMC|ONE" that highlighted interesting conversations, new communities, and other points of interest for the general reader. Of course, anyone could create such a distillation -- the more the merrier!
All of these work for me because they solve the stated problem in a way that encourages the behaviors we want; reinforcing the new, rather than the old.
It Doesn't Matter What The Best Answer Is, Does It?
No, it doesn't, does it? There is no "right answer", just a statement of the desired outcomes, and different ways of getting there.
But, if this experience has taught me anything, it's that we need to be even more clear about what we're trying to do here. There's a ton of pre-conceived baggage around what an enterprise portal should do for its users. Part of the magic here is that we're redefining traditional expectations in a very progressive, 2.0-ish manner.
Some people will like it, some people won't.
But, sooner or later, they'll all have to learn to be proficient in a 2.0 world.
Maybe there should be a few different welcome emails to choose from each reflecting different users in the community. The all-business welcome email is primarily a search oriented user who subscribes mostly to business oriented discussions. The more socially oriented welcome email is primarily a user who joins the Green community and Water Cooler. The curious welcome email is primarily a user that had his/her foot in both worlds. Or maybe even have preset dummy user configurations/sign-ins so that they can go see for themselves what kinds of different information each perspective sees.
Posted by: John Prichard | April 21, 2008 at 05:24 PM