What makes Wetpaint’s pitch stand out is that the company has a public lab experiment it has built out to prove its point: It built Wetpaint.com, an entertainment news site, solely so it could show off the Facebook traffic techniques it has figured out.
Now that site — which looks a whole lot like any other site that covers pop-culture candy like “Twilight” and “Real Housewives” — has 12 million uniques, many of which CEO Ben Elowitz acquired via Facebook. And Elowitz has been able to use it start wooing clients who want his insight.
The first: Germany’s Hubert Burda publishing conglomerate, which has signed up to use Wetpaint’s technology for a multiple-year deal.
If you don’t have Burda’s checkbook, you can still get a free taste of Wetpaint’s insight here. Last week, I chatted with Elowitz about some basic Facebook does and don’ts:
Ben Elowitz: The first thing is who you have working on it. There’s tons of of supposed “social experts” out there. But most of them have experience with one particular brand or audience. What you really want to do is hire for analytical talent, for empiricists, not hire for any kind of supposed social expertise.
The more they say they’re an expert, the more they’re clouded by a few things that worked in a prior role. Instead of saying, “let me go see what works in this case.”
You’re talking about 22-year-olds right out of college? There are empiricists in that group?
There are. We hire them.
Do they have to be 22? Could a 40-year-old do this?
It helps to be in the demo. Because the mechanics of it, when they’re natural and you speak it fluently, as a first language — you have good intuition about what kind of things to try. And then if you’re oriented analytically, all you care about is running a bunch of experiments, and seeing what works and what doesn’t.
So what do you have these 22-year-olds do?
Experiment a lot. That’s how you generate all kinds of insights that you want.
And then really institutionalize what works and what doesn’t. That’s been really hard. In the old SEO days, you had a guy sitting off in corner, who told you “Oh, use an underscore instead of a dash in your URL.”
Now what’s much more effective is to have folks testing things, measuring them, and then making sure that when you learn something, it’s institutionalized in a playbook your entire social media team can refer back to and implement. The institutionalizing turns out to be where all the value is. It doesn’t help you to know what works for the audience if you don’t implement it every day.
Everything you’re talking about here is about framework and structure. What about the content of the actual stuff you put on Facebook? I remember hearing that people like sharing happy stuff more than negative stuff.
That’s true, but it changes by audience. We ran a set of experiments on whether happy stories or stories about life’s foibles resonate more. It turns out that for “Real Housewives” fans on the East Coast, all the stories about foibles were getting about 30 percent higher performance. On the West Coast, all the happy events — the weddings, the birth announcements, were doing far better.
That’s why I say that anybody who has a general rule is probably leading you in the wrong direction.
And how do you ensure that people don’t just see your stuff on Facebook, but actually click through and end up on your site?
There are two pieces to measure: Impressions and how much visibility it got, and then click-through rate.
Impressions are largely based on what came before — not how that particular post is doing. It’s based on what Facebook thinks the audience interest is going to be. They make a guess, before that post shows up, of whether it’s going to be interesting, mostly based on how strong your relationship is with the brand that’s publishing it.
So this is either a vicious or virtuous cycle? If I’m putting out stuff that people aren’t clicking on, then Facebook will promote it less, and fewer people will see the next one?
What happens when Facebook changes up what it thinks it is important? They’ve been pretty up front about the fact that they’re constantly fiddling with their dials.
There’s parts of Facebook that are super volatile — like the social readers. But the basic relationship between the brands and their fans is much much more stable.
Let’s say I work for a Web site that covers technology. How do I Facebook better?
The most important thing you can do, to start off with, is to look at how many posts a day you are putting out there, and what times of day. That let’s you create an analysis of what’s worked and what hasn’t before. You can get started manually with a giant Excel table and a lot of data entry.
That seems like a lot of work.
The opportunity is way bigger than most people realize. For most sites, they get about 4 percent of their traffic from Facebook. We’ve been able to 10x that.