Blog Post

'Visits' doesn't equal 'enrollees'—but traffic data can be useful

October 3, 2013

    Dan Diamond, Managing Editor

    CMS really doesn't want to give out a much-desired number: How many people have signed up for health insurance through the Affordable Care Act's new exchanges.

    Sure, we've only been living in the era for just over two days, as I write this. But several states have been very transparent around how many people have applied for coverage in their exchanges, and CMS has heavily promoted some other numbers, which I'll get to in a second. Meanwhile, the agency's caution was magnified during an odd press call on Tuesday, when reporter after reporter was rebuffed when asking about the number of exchange enrollees. (As amusingly chronicled by Bloomberg’s Drew Armstrong on Twitter.)

    It's possible that CMS is being reticent because the number of enrollees is quite small, as reason's Peter Suderman suggests, given all the functionality problems with the federal exchanges. Several diligent reporters—the Chicago Tribune's Peter Frost, the Huffington Post's Jeffrey Young, and the Washington Post's Sarah Kliff, among them—have searched for people who signed up for health insurance through the federal exchanges...and have been essentially stymied.

    Don't judge the exchanges' fate too soon: Join us for a conversation on November 15 as our experts perform a pulse check on Obamacare.

    Another scenario is that federal officials are deliberately waiting to release the data as part of some broader messaging campaign. (A third option is that the exchange's systems are so goofed-up that CMS truly doesn't know how many people have applied for coverage yet. That isn't impossible given all of the other IT challenges, although counting enrollees would seem to be a pretty basic piece of functionality.)

    Can we learn anything from Web traffic data?

    Meanwhile, CMS has heavily promoted the number of Web visits to—nearly 5 million visits on Tuesday alone.

    There's been some debate among health journalists over why the agency is trumpeting traffic and if those numbers (and state-level data on Web traffic, too) actually matter. Especially because of the obvious showroom effect: Just because someone browses your website doesn't mean she's actually going to buy anything.

    Even if officials are only seizing on traffic data because the numbers are so robust, they hope it proves that the exchanges' launch went well, the Web numbers could provide a window into some users' experience and even a back-door path into the number of exchange enrollees.

    Clues to user experience?
    Look at California's data, for example. The spokesperson for the state's exchange is in some hot water because of a minor kerfuffle over whether state officials exaggerated traffic to their website. As it turns out, Covered California didn't see 5 million unique visits on Tuesday, as originally reported; the site saw about 500,000 unique visitors, who browsed about 5 million pages.

    That's useful data, although not necessarily revealing without context from users. Was the page-per-unique visitor count high because users were frustrated by the technical problems and had to reload or make return visits? (Based on anecdotal feedback, we know the answer here is yes.) Or was it high because visitors found a reason to keep browsing? (This is where CMS wants to get to.)

    Clues to enrollees?
    The traffic numbers also might—might—give us a sense for how many people enrolled through the exchanges after all.

    For instance, Kentucky's state exchange website saw nearly 78,000 unique visitors on Monday, who visited 939,000 different pages; about 4,700 applications were created and somewhere in the neighborhood of 2,900 people ended up applying for coverage, or about 3% of unique visitors to the site.

    That's not a constant algorithm, unfortunately. Nevada's exchange, meanwhile, has been reporting a much lower ratio of pages-to-unique visitors, as well as a lower ratio of unique visitors-to-applications. But it's possible that looking across the various exchanges might surface some common threads around how many would-be shoppers end up sticking around to buy coverage. It's certainly something that the Daily Briefing team is planning to do!

    Perhaps CMS will make this post moot by becoming more transparent about the number of exchange enrollees, before we have to do our own rough accounting. But don't count on it.

    Our exchange tracker: Which sites are glitchy?

    Here's the Daily Briefing's rolling look at which exchange websites are still having glitches, as of Oct. 3, 4:00 p.m. ET.

    Have a Question?


    Ask our experts a question on any topic in health care by visiting our member portal, AskAdvisory.