In a strange confluence of events, yesterday I: wrote a post about filtering and connecting when you’re a small enterprise; then Clay Shirky wrote about almost exactly the same subject in a post on algorithmic authority; then I ended up talking about the same topic with Paul Moynagh from the innovation consulting group Tough Problem. When I combine all that with something that John said today, it’s helped me clarify my thinking.

The original point that I was making was that while algorithmic aggregation and filtering has some real strengths, there still might be a time when you want the opinion of an expert rather than everyone. When Paul and I were talking, the way that I framed it was by saying that google is great if you’re trying to find out what millions of people like, but it’s not quite as good when you’re trying to find something that hundreds of people like. In some ways, this is like going back to the original model of yahoo’s – where every link indexed was checked first by an actual person. Immediately after I posted that Shirky then said this:

It’s also worth noting that algorithmic authority isn’t tied to digital data or even late-model information tools. The design of Wikileaks and Citizendium and Apache all use human vetting by actors prized for their expertise as a key part of the process. What seems important is that the decision to trust Google search, say, can’t be explained as a simple extension of previous models. (Whereas the old Yahoo directory model was, specifically, an institutional model, and one that failed at scale.)

At first I thought that he had demolished my argument. But then I realised that his examples of Wikileaks and Citizendium are exactly what I was trying to get at – smaller organisations that filter in a particular, small area. In other words, if you can’t be google and aggregate everyone, maybe the thing to do is to be an expert, and filter one particular thing. John said this then in a conversation about Rupert Murdoch’s efforts to turn off google:

It’a actually an interesting case for thinking about competing value propositions and strategy. Murdoch thinks he can make money out of what he thinks is the “best product” from professional journalists, but it’s a fool’s gold game. Other customer value propositions are lowest cost (not an option here because the lowest marginal cost is 0) or customer intimacy (specific content to a targeted audience at low scale). I think that any business model needs to be based on the last value proposition, to be viable. This is how we would approach it as a strategy case.

That’s a framework that he uses in his strategy classes, and it’s a pretty useful one. What I was arguing for without really realising it was that the way to implement a successful filtering strategy as a small firm is to go for customer intimacy. Find something that is important to a small group of people, aggregate the information that pertains to the topic, filter it to identify the most useful ideas, and connect people up with the ideas. And with each other, and with you.