Not Quite Ready For Prime Time: The State Health Insurance Marketplaces and Google

All of the state health insurance marketplaces (also known as exchanges) are online, but millions of expected users may have a hard time finding them.  Marketplaces will enable shopping and enrollment mainly through their websites. States are using a variety of promotional strategies, but most people will likely find marketplaces in the same way they find other websites—through common keyword searches on Google, by far the nation’s dominant search engine. Poor search engine results can create serious barriers to shopping and enrollment, the major measures of success for marketplaces and, by extension, the success of the Affordable Care Act (ACA).

We used standard methods to assess Google results for the 17 marketplaces operated by 16 states and Washington, DC that offer individuals, families, and small businesses a place to shop for health care coverage.  Over three days in mid-September, we looked at results for keywords that data from Google show people are commonly using to search for health insurance. We examined both unpaid (or “organic”) and paid (or “sponsored”) results. Although research shows that unpaid results get more attention, paid results can also lead to page views.

Our preliminary findings show that marketplaces for four states—Idaho, Maryland, New Mexico, and New York—and Washington, DC did not appear on the first page of Google results, which generates 92% of all page views.  In addition, both unpaid and paid search results for most of the remaining 12 states were frequently absent from page one.

With enrollment in the marketplaces opening October 1 for coverage beginning January 1, this would be a good time to focus on search engine optimization (SEO), the process of increasing the rankings of unpaid or “organic” search results. Once implemented, SEO results can be seen quickly, especially for a topic as popular and important as new health insurance options. However, it requires analysis, planning, and time to implement.

Methods for Conducting Search Engine Result Testing

To test search engine results for state-operated health insurance marketplaces, we used the five keywords most effective in producing page views of a prominent healthcare-related website produced by a federal client: Affordable Care Act, affordable health care, health care, health insurance, and Medicaid.

We further verified the effectiveness of these keywords using the Google Keyword Planner tool, which provides approximate data on the average numbers of searches per month, the average cost for paid search ads using the keywords, and the number of advertisers bidding to use the keywords (i.e., their competitiveness). Even though this tool is designed to identify keywords to improve the impact of paid searches, it is also commonly employed to identify keywords to improve the impact of unpaid searches.

From September 9-11, 2013 we performed one search per state each day. Each search was conducted at a different time–morning, afternoon and evening. For each state, we therefore collected five samples of page one search results each day for a total of 15 samples per keyword.

To make it more likely that these Google results would be similar to the ones that average users would receive, we conducted each of the 255 searches (15 searches for 17 states) using the Google Chrome incognito browser. Before each search, we also cleared our browser data, opened a new incognito window, and manually changed the physical location in our Google results page to a major city in the state we were analyzing. We then recorded whether the state’s marketplace appeared on page one as an unpaid or paid result.

State Health Insurance Market Test Results

Table 1 presents all unpaid and paid search engine results.  As it shows, the keywords we tested did not produce any page one Google results for marketplaces for Idaho, Maryland, New Mexico, New York, and Washington, DC.  They also produced limited page one Google results for most of the other state marketplaces.

Covered California is a notable exception. Common keywords frequently produced page one search results for this marketplace. In particular, in the case of the keyword, “Affordable Care Act,” it produced a paid result and two unpaid results: one to the website home page and the other to another page within the site.

The five common keywords that we tested produced a range of unpaid and paid page one results. “Health insurance” produced the most of any keyword: 22 unpaid and five paid results, or almost one third of page one results for the marketplaces.  “Health care” was a close second, generating 16 unpaid and six paid page one results, or almost one quarter of the page one results for marketplaces.

Medicaid as a keyword did not produce page one results for any of the marketplaces.  For Internet users, this may weaken the “no wrong door” approach to providing information and determining eligibility. In states where Medicaid managed care plans will be available in the marketplaces as Qualified Health Plans (QHPs), a nonproductive search for that popular keyword can create a digital barrier to shopping and enrollment.

Table 1: Unpaid and Paid Google Search Engine Results by Keyword for State Marketplaces (Click to Expand)


These findings are preliminary, but we believe they provide useful insights. We note that our sample of search engine results for state marketplaces may not reflect the impact of SEO efforts that are already underway.

We counted only page one results for marketplaces. Although the higher the ranking of unpaid results (i.e., #1 vs. #10) and the position of paid results (i.e., at the top or on the right side) on any page affect how likely users are to view them, the appearance of marketplaces on the first page of Google search results should be an important, if initial, goal for state SEO efforts.

Because Google delivers highly customized searches based on a variety of factors such as the location of the user and the user’s browser and search history, it is difficult to get a truly representative sample of search results. However, we have taken measures described above to maximize the accuracy of our results.

Elaine K. Swift, PhD is a health care researcher who leads contracts for the federal government to support public reporting initiatives.

David McCormick is a Digital Marketing Strategist at Blast Analytics & Marketing. His specialty is content optimization, content marketing and analytics analysis. David has been working with government organizations to help optimize sites for the new changes in healthcare requirements.

2 replies »

  1. The word searches used by the authors don’t seem to be ones most folks would use to search for an exchange. When you search under “state name health exchange” you get the site on the first page. Why are the authors surprised when broad categories bring up broad answers? Google appears to be doing just fine when you use the phrasing the bulk of the population would use.

  2. Hmm, interesting research. Would be curious to know what kind of search volume Google is seeing in this area so far and how things change when enrollment starts and people start linking to the official sites – from what I know, which admittedly isn’t very much – the algorithm Google uses to rank search results gives a lot of weight to the number of sites linking to you ..

    Would also like to see the $$$ numbers you turned up for keyword demand in the various markets and how much individual states are spending on optimization, for that matter ….