Extracting Key Metrics to Make Your Hyperlocal Site More Valuable
Say you’re the publisher of a hyperlocal news site that’s a year old. You’ve gained some traction with the audience in your community, but revenue remains the big issue. When you do get merchants and other businesses to listen to your ad pitch, they often come back with a version of this: “You’re one of umpteen sites looking for my dollars — prove you’re special.”
You can quote strong counts in unique visitors and page views, but are they enough to lift a community website above the digital publishing clutter that confronts small and medium-sized businesses in most markets? Lars Lofgren, marketing analyst for the analytics company KISSMetrics, says no.
“In the basic business model, you’re selling impressions,” Lofgren says. “You want to raise your page-view count as high as possible. You chase news cycles in the hopes of going viral. You’ve got a content machine. But five minutes later you need another piece of content.”
“Unique visitors and page views work well with this model. But there’s a big disconnect between those numbers and everything you need to know as a publisher: Who are my visitors? What sections on my site are they most using? How can I get more visitors to do that? The answers require a new model – one that matches qualitative data with quantitative data.”
“Quantitative” data — those UVs and PVs — work effectively for advertisers who want volume, Lofgren says, but most hyperlocals aren’t big enough to deliver multi-million numbers in audience. “Qualitative” data, he says, tells you who your customers are — by their activity on the site, as it’s actually happening — or their identity, with information neatly served up from sign-up info. But it can tell much more if the publisher surveys a sampling of site visitors with multiple-choice questions like:
You didn’t make a purchase today. Can you tell us why not?
— Just browsing – did not intend to make a purchase.
— Did not find what I was looking for (please explain).
— Prices were too high.
What if a publisher wants to find out information that’s more elusive, like what kind of editorial content will attract visitors who will be more valuable to increasingly choosy advertisers?
“You’re talking about validating your idea,” Lofgren said. “That can be difficult, but you can do it. You test two different articles with two halves of your audience. Firms like Visual Website Optimizer and Optimizely can set this up.” After the split-audience testing, “that’s where KissMetrics becomes essential,” he continued. “KISSMetrics keeps track of which people move through each version of an A/B [split audience] test, I can pull a full list of people that went through version A, and look at individual reports to see how they’re using the site. I can also look through reports on people that went through version B and compare the differences.” From those comparisons, says Lofgren, emerge recommendations regarding higher-value visitors.
Now, all the publisher (who very often is the editor at independent sites) has to do is come up with editorial content that will attract enough of those higher-value visitors. That will take a lot more than even the best analytics.
Tom Grubisich authors The New News column for Street Fight. He is editorial director of LocalAmerica, which is partnering with InstantAtlas to develop sites built around how communities rate in livability. Local America was recently featured on Reynolds Journalism Institute’s Pivot Point site.