Building a Better SXSW

A team of Viget folks spent a long weekend in Texas for South By Southwest. We saw some interesting panels, reunited with folks we hadn't seen in months or since last year, and got to meet and talk to dozens of folks we hadn't ever seen. It was definitely a great time, but it was also lacking a lot that we look for in a great event, especially one being paid for by our professional development budgets.

South By Southwest's Interactive Festival is certainly the biggest interactive conference every year. In fact, I dare say that the majority of people in the industry look forward to heading to Austin each March. But as SXSW and its attendees have matured, comments about the conference have gone from "I heard some amazing talks at South By" to something more like "well, you don't go to South By for the panels, you go for the socializing and networking" or "it feels like a big family reunion." It may still be a must-attend event for many, but it's become a social event, not a conference.

Wisdom of the Crowds

Part of the SXSW pitch is that it is, at least in large part, a user-generated conference. Panelists aren't paid (except with free admission), panelists tend to be folks who would be attending SXSW anyway, and the panels are ostensibly chosen by the SXSW community. This whole panel selection process happens in September, a full half a year before the event itself.

The first problem is that the sample size is too large to make for an interesting conference. While the voting of 100 people would yield some safe topics being presented by experienced speakers, there would also be more edgy panels that address the need you didn't know you had. With 10,000 people voting, the panels that folks choose are much like our politics — moderate and safe. Talking about design inspiration and defending your work in the face of clients is a reasonably safe topic, and you might leave with some decent takeaways, but it's not a talk that is likely to blow your mind or dramatically change your process.

The Web Changes Quickly

The second part is probably more significant: by the time it gets to SXSW, it's old news. Unless there's a product launch involved, few people are willing to wait six months to release their content. So a girl writes a great post on a given topic in June, getting some significant buzz. She proposes a talk in July or August, and folks look at the list in September. The buzz is still around, so folks give it the nod. However, by March, it's nothing new, and it's either a panel you're happy to skip or one you regret having attended.

This lack of new, relevant information may be somewhat common at conferences, but it's especially apparent at SXSW. Ideally, content is evergreen, and six months doesn't really mean anything. But the web is still so young, and six months probably represents a period during which you revised the way you work at least once or twice. You might even have revised your "specialization" or research interests more than once.

Again, for some talks, this timing thing doesn't matter. If it's about culture in the workplace, working with clients, or something like that, it's probably relevant anytime. But those are also the safe talks.

Grow with Your Audience

One of the things we've talked about with DC Design Talks is growing with our audience. After two, three, or five years of SXSW, the veterans don't need to see another talk about standards-based front-end development. We get the gist of the common themes in the community that SXSW addresses. But still, you get to feeling like it's the same old thing, year after year.

The Human Social Network Works

In the end, the value of SXSW comes from the people who are there, especially when they aren't inside the Austin Convention Center. Having a discussion over tea and cake at Halcyon (before it was too cool for school), talking about something that had happened in the last week or two, with smart folks you may or may not have known, is what I go to Austin for. It's in those conversations that real problems are uncovered, solutions proposed, connections made, and you get excited about the industry and community in which you work.

SXSW is just a catalyst for these discussions. That alone makes it rather valuable. But it's a lost opportunity not to have a great conference alongside the great social and networking event, and I think that lost opportunity is beginning to wear at folks.

Just as I was departing my hotel in Austin, one of the A-listers in the web community (who won't be named unless he decides to comment and name himself) was telling me that it just wasn't that great an experience this year. I'm rather sure that he's not alone, based on the fact that a number of other A-listers chose to skip out this year.

Can it Be Saved?

Maybe. Here are a few ideas, not all of which go together cleanly:

• Skip the Panel Voting. Voting-style user-generated content is cool for things where popularity really is the end game, like mainstream news (or even one category of news, like technology on Digg), but it's awful for instances where you really want to push the limits. Industry conferences should be pushing those limits, not telling everyone what they [should] already know.

  Instead, still let folks submit panels as they do now, but just let the advisory board pick all of them. In fact, only let people who have already registered for SXSW submit panels, just to drive the value of investment in the conference.

• Instead of submitting panels, allow users to pitch themselves as speakers or moderators. If the advisory board sees you're doing good work, then you'll probably have something to talk about and share with others. Especially if your blog content is strong, or you've been speaking on interesting topics in the past, this would be especially apparent. In this event, I guess the advisory board might want to survey attendees asking about topics they'd like covered.
• Bump the panel process closer to the event. Choose panels in January or something, which would make the content fresher and more relevant.
• Give more weight to the veterans. This is vaguely elitist, though finding a SXSW veteran isn't exactly a high bar. If we're voting for panels, give the least weight to folks who aren't registered, more weight to those who are, and the most weight to people who have attended in the past. This probably happens to some extent in the process now, but it should be more blatant.
• Fewer panels, more talks. I've found that, with a couple of exceptions, the panels have been of less value to me than the solo or duet talks. This probably has to do with the focus you can give a talk when it's just you, rather than coordinating focus among four or five individuals.

In the end, I'll almost definitely attend SXSW next year, because it's still very valuable as a social event. But I wish the content were equally valuable. Thoughts? Other ideas for improvement?