In Counting Actual People, Hyperlocals’ Smaller Traffic Numbers Can Be Big
Check out this chart (image below) based on an analysis by Borrell Associates of one community site’s traffic. UVs are a hefty 881,227 in a month. But actual people are a small fraction – 181,091 (the red sliver in the pie).
In short, many of the “unique visitors” were not unique at all, as Greg Harmon, Senior Research Analyst at Borrell Associates, points out in his takedown. On the Internet, a single user can lead multiple virtual lives. It happens when the user works from multiple computers over the usual 30-day tracking period, and deletes computer “cookie” trails.
But even the 181,091 number is a distortion of what’s actually occurring on the site being measured. Harmon’s research found that “core loyalists” – those who visit the site most frequently – represented only 48,960.
Yet these incredibly shrinking numbers don’t have to be a turn-off for a potential advertiser. A business could well be getting a very good return on investment by placing an ad on the site. That’s because those 48,960 “loyalists are a more cost-effective universe of potential customers than the 880,000-plus UVs who melted away under Harmon’s scrutiny.
With Small and Medium-Sized Businesses (SMBs) and Very Small Businesses (VSBs) looking at more and more digital marketing options in their search for customers, community news sites have to be aggressive in selling themselves to potential advertisers (and raising revenue that will pay for editorial and other improvements). How can they make the most of their traffic numbers, which are much lower than those at most commerce-based sites? I put some questions to Harmon:
A single-community news site whose clients are mainly VSBs needs to focus on “making the phone ring” rather than branding messages, we are told. How should this small site present its traffic numbers to attract VSBs?
To the extent that a local site is competing with other similar media outlets (other newspapers, radio, or TV), they need to acknowledge the competitive numbers, like page views and unique visitors, from Omniture, Google Analytics or whatever is being used, but only if the client is aware of them. We recommend that, to make its case to most businesses, a community website concentrate on the numbers of actual people from the local area that use the site. It should collect its own proprietary numbers.
National and regional advertisers and large agencies only concern themselves with impressions as delivered via page-view reports, so they don’t much care about accurate sizing of local audiences. But local advertisers do and should.
Will Google Analytics give this small site the people numbers it needs — or does it need to go to a company like yours or Quantcast, Chartbeat and KissMetrics that can provide deeper audience metrics?
Google Analytics will provide any site using the platform with data, regardless of size. But these are the page views, visits, and unique visitor data, so not particularly useful for sales.
Quantcast etc. will typically require a registration, but can even work for smaller site and of course, we think these are better numbers, but continue to believe that the best numbers are from direct on-site research. We do that. If a community site doesn’t want to hire an outside party, it should at least do its own research.
Can the small, independent site that is doing okay on reaching toward sustainability, but doesn’t have a lot of financial resources, afford what it takes to make a credible numbers presentation to advertisers?
Most local media can probably afford what expense there is to develop data for good, solid presentation numbers.
Should a regional or national network of community sites that targets not only VSBs and SMBs but also regional and maybe even national businesses make a different pitch when it comes to traffic numbers?
I do not think the pitch would vary between the two types as to what it has to say about audience: “We deliver X size total audience with Y local people and Z loyalists.” Beyond this, the pitch is all about ROI and audience behaviors, affinity and interests.
What about user “engagement” numbers based on various actions the user takes on the site (e.g., number of pages visited, length of time spent on each page and cumulatively and click-throughs to advertiser messages). Aren’t those numbers sometimes better than even “people” visits?
Click-throughs not so much. But the other engagement factors are useful. It is also use a good idea to report real audience numbers because even a highly targeted niche site that might have only a few thousand users is going to be powerfully attractive to a local vendor or related products and services. And the return on investment will be much better than at a larger site with a more dispersed audience in terms of interest.
Tom Grubisich authors The New News column for Street Fight. He is editorial director of LocalAmerica, which is partnering with InstantAtlas to develop sites that will present how communities rate in livability. Local America is featured on the Reynolds Journalism Institute’s Pivot Point site.
Top image courtesy of Flickr user Mark Fischer.