There’s lots of speculation about Twitter’s business model, from the serious to the comic. The firm’s backers claim the company has plenty of money for the long haul. In fact, given the openness Twitter has traditionally shown with its APIs, the model could be to let all of us speculate about it, then pick the winners.

But I’ll bite. I have an idea how Twitter could make money.

Most of the business models I’ve seen charge the publisher. Why not charge the audience?

We live in an attention economy. We’ve moved beyond the information economy — now, anyone can get access to anything. Instead, we want to know what’s worth our time. Google makes money by ranking information based on relevance; Paris Hilton makes money by pointing us at the scandalous; newspaper editors make money by selecting topics they think their readers will find interesting.

Lots of people are experts on things. I’d pay to follow someone smart and knowledgeable. Maybe only $10 a year, but in return, they’d search for useful information and tell me about it. They might be an expert on cloud computing, or web monitoring, or sustainable food, or transparent government. I’d follow them. I’d get links from them (which only susbscribers would receive, of course) to reports they’d written, or news they’d found.

Eventually, I’d follow some services. They might be stock feeds, or weather systems, or alerts — basically what we call Prospective Search, like Google Alerts or Rollyo, but tailored to my interests.

I would happily pay $50 into a “followdollars” account, and be able to spend it with a single click in order to tailore my personal set of feeds and streams. I suspect others would too. It’d be like an RSS feed from subject matter experts.

[UPDATE: As some folks have (already) pointed out, “what’s to prevent a follower from sharing with others?” The short answer is nothing — but I would provide each recipient with a unique, personalized URL (this is Twitter, so we own the data feed and can rewrite URLs) to track abuse and message spread. Then when you see a certain URL come from many addresses/user agents at once, replace the page with a “pay to follow” promotion.]

As a product manager, I’m always looking for where the scarcity is. That’s usually what people will pay for. Groucho Marx told Beverly Hills Friars’ Club, “I don’t care to belong to any club that will have me as a member.” 2009 is the year of the Velvet Rope in social networks. Twitter’s an open world in which anyone can follow anyone. Exclusive content, monetized by Twitter, sounds awfully compelling.

On the back end, Twitter would charge some fee (say, 20% of revenues) in return for handling the payment processing and providing analytics and marketing tools.

Anyway, that’s my take on it. Meanwhile, let’s see what else is out there. Most of the business model contenders seem to fall into one of three categories, each of which has its limitations.

Advertising

Flogging something is hard to do in 140 characters. Since a large number of Twitter users drink from the Twitter hose through third-party clients like Tweetdeck and Twhirl, Twitter can’t just put ad inventory on Twitter.com

It’s hard to see how Twitter can get the CPMs it needs without constraining its APIs in order to force people to use its website–which will hurt the openness that has brought it so far and trigger user backlash.

Analytics

Seeing where messages are going is useful and interesting to companies that want to understand their markets. Semantic interpretation and natural language parsing is a great way to generate market intelligence.

If you want to tie what people say back to what they do, however, you need more than just reports of message popularity. The source (Tweet) needs to send the destination (a website running analytics) a tag (such as a URL parameter) telling it which message triggered the behavior.

Note that the bit.ly/Tweetdeck combination is a very enticing advertising and analytics platform, with bit.ly already finding commercial customers to use its shortened-URL analysis for campaign tracking. We became acutely aware of this while working on the community section of the Complete Web Monitoring book.

I think this is Twitter’s best hope, since they have unique access to message spread and virality. I also note that the one thing that forces me to go back to Twitter’s website is the “new follower” notifications. I haven’t dug into the API to see if this is something that’s explicitly not part of the data they share, but IMHO anything Twitter isn’t sharing yet (such as new follower notification) is core to their upcoming business model.

Pay to publish

Charging companies to get their message out misses the point: Social networks aren’t about broadcasting data, they’re about encouraging interaction. Twitter’s about letting anyone access anyone in an asymmetric manner; if the big guys have to pay to publish, it breaks the pact of celebrity access that fuels much of Twitter’s mainstream appeal.

There are plenty of other ideas — name registration, calendaring applications, suggested follows, and so on — that are ripe for development. Which is one of the things that makes Twitter fascinating: No startup ever makes money doing what they set out to, but the hardest thing to generate is attention, and they’ve got that in spades.