Migrating an Online Community is Like Completing Someone Else’s Sudoku
by Matthew Lees
Maniac Sudoku Puzzler on the Loose
Sitting in seat 15D of my homebound flight yesterday, I opened up the airline magazine to work on the Sudoku puzzle in the down time between take off and beverage service.
Unfortunately, to my near horror, someone had already started the “Gentle” Sudoku, entering around 20 numbers, or about a third of what still needed to be filled in. Although the magazine gives three versions to choose from (Gentle, Moderate, and Diabolical), air travel doesn’t make me especially receptive to challenging mental workouts, so I figured I’d just start where the other person left off on the Gentle version.
My initial assumption was that the previous solver knew what they were doing, but either got bored or ran out of time before landing. While my personal puzzle preference leans more towards crosswords, I’m not too bad at Sudokus, so less than one minute into things, I realized that this assumption was a bad one. One nine-by-nine square had two 8s! There were two 9s in another! And how on earth could you write a 4 in that box, when there’s only one 4 given as a starting clue in the whole puzzle?
After some deep breathing exercises to calm me down from this outrage, and spending a few minutes thinking up scenarios that might explain such a poor attempt – not really knowing how Sudokus work, but giving it a whirl anyway? insanity (temporary or otherwise)? intoxication? pure mischievousness (in which case, they got me good)? – I decided to work on it anyway. After all, it was the easiest level, their pen had been black while mine was blue (so I could distinguish who did what), and they hadn’t filled in too too many numbers. So how hard could it be?
I’ll leave out the exciting details, but I completed the puzzle after about 30 minutes. It wasn’t pretty, though, as you can see from the image above. Along the way, I found that, while some of my unknown co-solver’s answers had been wrong, others were indeed correct.
Building a community from scratch is like solving a new Sudoku.
Migrating a community is like solving a Sudoku that someone else already started.
I’m currently working with a client on migrating an online community from one platform to another. Their B2B community has lived for over three years on a homegrown platform that, while impressive three years ago, is now seen as lacking essential features and functionality that the company’s users want and expect, and that the company requires to effectively manage, grow, and maximize the community’s value.
So we’re knee-deep in thinking through the ins and outs of the migration, planning how best to (1) move data (community content and conversations, member profiles, etc.) to the new platform, (2) configure the technology (reputation system, moderation workflow, single sign-on, etc.), and (3) communicate with key enthusiasts/influencers and rest of the user base. Some of these elements are informational in nature, some are technological, and others are social.
What Came Before
The social aspects are particularly apt for the Sudoku analogy. By definition, an online community that’s migrating to a new platform isn’t starting from scratch, which means it already has a culture, a shared history, and certain ways of doing things. The migration can’t help but change some of these. Ideally, all changes will be for the better, but the important thing is, successful migrations depend on knowing what came before.
If you’re involved with a community migration, you may feel that some of the things that came before were good – in the way that some of the original Sudoku solver’s numbers were correct – in which case you’ll replicate them as closely as you can. And some of what came before may not be aligned with the direction you’re going – in the way that I had to change the incorrect Sudoku numbers – so you’ll adapt.
For sure, the analogy (like all analogies) is imperfect. Puzzles have correct answers, but there’s no “right” or “wrong” way to approach online communities. There are only degrees of success based on your and your users’ criteria. But there are best practices based on approaches that tend to work.
Still, you shouldn’t be surprised if things get messy, like my smudged, cross-out-filled Sudoku. A few hurdles are okay if you still get to where you want to go.
A Final Note: If you really don’t like what came before, finger pointing doesn’t solve anything. Experience, expertise, effort, patience, and iteration, however, go a long way. That said, if you recognize your handwriting in black ink in the Sudoku above, I’d like to have a word with you…
Back to the blog
This entry was posted on Thursday, March 18th, 2010 at 3:57 pm and is filed under Best Practices, Community Moderation, Social Media Trends. Both comments and pings are currently closed.
4 Responses to “Migrating an Online Community is Like Completing Someone Else’s Sudoku”
MatthewLees says: Phil – Thanks for sharing some of the challenges Lithium regularly sees with community migrations, and the practices you recommend for dealing with them.
Over time, as existing communities mature, as technology platforms continue to develop, and as the industry consolidates (as it inevitably will), we’ll be seeing more and more migrations. While there are a great many similarities in launching a brand new community and re-launching a community on a new platform, I expect the expertise in helping with migrations — particularly around the social, technological, and organizational aspects — will be highly valued. (Could be solid business for vendors, consultancies/agencies, and individuals with direct experience…)
A great anecdote, Matthew. So often I hear community projects related to a “garden” or something like that, so a partially (and possibly drunkenly) completed Sudoku analogy made me chuckle.
Community migrations are pretty tricky, but not for the reasons I always find to be immediately obvious. The technical challenges can always be a burden because the customer specifically wants to come onto the new platform for a lot of feature / functionality / extensibility improvements, but along the way, they bring their “baggage” with them.
And by baggage, I dont mean bad data. The “baggage” can be things that led to failure or trouble on the last platform. For example, maybe the community really failed on the previous platform because of a lack of promotion, or a lack of a real Community Manager. Maybe it was actually a good community on a good platform, but the business never really ran internal metrics against it, and never took a deep dive into how to reconcile business objectives with community objectives.
Ownership (and investment) issues oftentimes plague a business when they’re first starting a community. And I’ve found that those issues do not magically resolve themselves just because the business has decided on a new platform or a new vendor.
When a migration comes my way, my first objective is to suss out the Community Management team. It is so easy to get bogged down in the all the technical details, the change management, the platform opportunities, etc. Those things are vitally important, but I like to go straight to the heart of the matter by finding out who the team is, what their experience is, and how they hope to operate in this next new phase.
This may seem so cliche (and at times, untrue), but I’ve found that you can tell a lot about a town, state, or country by the mindset of their leaders. Sometimes it takes a few generations of leadership change (or a change of attitudes within leadership) to see civic success.
MRowland says: Nice analogy… We’ve just watched as two of our clients migrated to new community platforms over the past two months. The first was for a company that went from Prospero to a white label Social Network and now to a new platform all in less than three years. The other went from a platform that they were continually re-writing code for to one that had better out of the box functionality and features. We worked with both in the transition, but acted as counselors for the team during the process more than anything else.
Our focus was on getting expectations set properly. The first client was down right giddy to be getting away from a really bad platform and thought this new one would do everything perfectly. The second was ready to change, but wasn’t thinking about all the great new things they could do; instead thinking about how to make the new platform look and function like the old one.
Both were mistakes on opposite sides of the expectations spectrum. A new platform is nice, but won’t solve all your issues. Focusing on new technology in this case is like believing that if you trade in your Chevy for that BMW, you’ll be able to date Megyn Fox’s twin sister. But buying a BMW and then wanting it to operate like a Chevy is a mistake as well.
Want your experience to go well? Have realistic expectations and think about what your outcome should be. That will involve more than the platform, it will involve updating inefficient processes and methodology to reflect the new platform’s strengths. It will involve trying new features out with your community based upon member input, but also your organization’s goals. In short, test, measure, update, test, measure, update, etc. You must keep an eye on your organization’s goals above the technology. Technology is a tool to help accomplish your goals. But the behavior and engagement of your members is where the value is in offering a community or social network.
Now, I have to go buy that BMW… 
Nice analogy.
At Lithium, We have a lot of experience in this area, and we can’t underscore the importance of #3 — communication with your most enthusiastic users — enough. These people will make or break your community, and if they don’t feel that they’ve been consulted, they are likely to be more critical than they would otherwise be about the new system.
Some other things we’ve learned:
* Don’t let your desire to rebrand the site make the new one too confusing for users. A fresh canvas is every artist’s dream, but remember that the canvas isn’t fresh for your users. They have been trained about how the site “should” work by what you had before.
* If the source reputation system has rewarded something, make sure there’s a reasonable analogy in the new system. People may get upset when behaviors that they’ve been rewarded for in the past no longer seem important. If the analogy isn’t perfect, make sure it gets explained.
* Set up an area to solicit feedback, and act on it quickly. We have a product that lets people suggest and vote on ideas, which turns out to be a very nice way to figure out what people like and don’t like about a new system. It’s also a nice way to keep griping focused and productive, rather than generalized and petty.
Hope that’s helpful for your readers. Let us know if we can help!
Phil