Monthly Archives: July 2008

How To Build Community Around Your Fiction

A Community of Monopoly HousesThe Internet is really only fulfilling when you’re a part of a community. Admit it: one of the main draws of the blogosphere is waiting for your readers to comment, to squeal at your hero’s bravado or to laugh at your jokes. The attraction is there for all of us, really – at Novelr I enjoy nothing more than going through a storm of particularly furious commenting, reading arguments and evaluating alternate viewpoints, sometimes even laughing at the occasional lolcat. There is joy in communication and love in community. We all crave it. Hell, we all want it in our blooks.

So how do you go about creating community? Much has been said in Novelr about the whys of community, but little has been said about the hows. This post is my attempt to rectify this lack: it contains all the observations I have made and some of the things I have done to create communities (at Novelr and elsewhere) over the past three years.

Finding A Suitable Metric

The metrics usually thrown about when we talk about blogs are comments, RSS subscribers and visitors. The truth is that community cannot be measured by any one of these metrics alone. Subscriber count and visitors may indicate the number of long term readers, but if they don’t speak up then they’re merely observers in the community of your blook. Comments, perhaps, are a better indicator of community, but we have to remember that 100 ‘tat was amzng lol!’ comments does not mean you’ve got a great crowd on your hands (maybe an unintelligent one, but that’s a different story).

I measure community with two interlinked gauges. The first is the interaction between readers. When your readers argue, joke (even flame) each other they ensure that your blook/blog isn’t about you anymore – it’s about them. But we have to remember that reader-reader interaction is an inconsistent metric – there will be months where the discussions and arguments will be plentiful, and there will be months when nobody’s posting. This is normal, though entirely unhelpful – if you think community equals activity then you’re going to get one hell of a headache trying to cultivate a constantly high level of it.

The second gauge is interlinked with the first, and can be summarized with a few simple questions: do your readers talk about your site and the community of people around it? If so, what do they say? And how often do they do it? These questions touch on the intangible quality of ‘we’ness – the integral core of any group.

One thing you’d notice is that this second gauge can be found in all strong communities. Matthew Haughey, creator of the Metafilter community blog, points out that people will start talking about the site and the community sooner or later if it’s good enough, so he provides a place to do so on his sites. Conversely, if they stop commenting on the state of the site (or if they stop cracking inside jokes) then your community has more or less disbanded. Not died, certainly, but it has lost its soul.

I consider a site successful in community the instant people start talking to each other, and about each other (us). Now I’m not going to mislead you: getting there is hard, but not impossible; maintaining it once you’ve got it is satisfying, but hard in a completely different way.

Floods and Streams: Where Traffic Comes From

TrafficI logged in a few weeks back to find Novelr’s stats behaving strangely. Reinvigorate reported that my traffic had leapt from 20 a day to 300 – a stunning figure, considering I hadn’t updated in a few days. So I checked the referrals, and I found that an article was making the Stumbleupon rounds. I was nonchalant. That particular article wasn’t very good.

Things got worse. Less than a week later Novelr’s traffic spiked at 900. I checked my bandwidth and breathed a sigh of relief to find that I still had a gig or so left. The viral word-of-mouth hadn’t killed my hosting package. Oh, how wrong I was. Traffic spiked at 3000 per day shortly after and lasted four days. Novelr got killed in the process.

What I found interesting in this wasn’t the traffic spike (that happens pretty often, of varying magnitude, to any website) – it was what happened afterwards that mattered. Novelr’s daily traffic remained at a constant 150-170 per day, a huge difference from the 40 daily visitors from before. I was surprised at this – blooking and digital fiction is a highly niched field to be blogging about, and I hadn’t expected so many new visitors on a day-to-day basis.

Here’s what happened: Novelr had found itself a traffic Stream. The initial Stumbleupon outbreak was more like a Flood: it took my hosting package down by consuming 5 gigs of bandwidth. But let’s take a look at both kinds of traffic, separately.

Floods

Floods are short term bursts of traffic that usually come from a linkup in a) a major blog, or b) a social news site (Digg and Reddit). The traffic from these Floods leave behind a trail of particularly unintelligent comments, have a high bounce rate, and they trickle off fairly quickly. What you do get from Floods, however, is wider exposure – and you’ll find a corresponding upsurge in your RSS subscribers and daily visitors for a few days after. Novelr has experienced two Floods: once for Problogger’s writing project (with this article) and the second being the initial Stumbleupon spike.

Streams

Streams are long term sources of traffic. Good examples of these are blog networks like 9rules, advertisements (as long as you can keep a consistent number of them running) and bookmark sites like Del.icio.us and Stumbleupon. I’m actually quite surprised at Stumbleupon’s ability to constantly point new visitors your way – I’ve always thought of it as a copycat Digg – but then again I am not a regular user of either so I can’t say. What all these examples have in common is that they aren’t particularly influenced by time. Whereas traffic from Digg and a linkup in, say, Boing Boing, dries up after awhile, these sources consistently bring you new traffic, even if it’s for an article that’s one year old.

Which Is Better?

I’d like to suggest that Streams are the traffic sources that people should aim for, simply because they’re more consistent and are thus more reliable than Floods. You’d notice that amongst the really big blogs (I’m thinking of Kottke and Daring Fireball here), traffic gets redirected quite often. These linkups count as Streams, not as Floods, because to each other these blogs act as consistent sources of new traffic.

I also believe that a site (be it blook, blog or company splashpage) will do well if the owner takes steps to convert as many Floods to Streams as possible. If Blog A links up to one of my stories, for instance (sending 1000 visitors or so my way), it’ll make sense to put it up on Stumbleupon as well. And my experience with the service is that the traffic spike may come months later, but that good content on Stumbleupon attracts new visitors regardless of how long it’s been ‘out there’.

A closing thought: how successful your blook is depends on how many eyeballs you command. And while finding Streams and enjoying Floods as they come is fine and dandy, in the end your job is to make sure new visitors love your stories, and come back. Like fruitflies on a particularly sticky banana pie, your job is to make sure they become regulars – that they become glued right on and that they can’t escape.

I’m preparing glue as I write this. Are you?