In my last Search Insider
article, I briefly mentioned comScore's Universal research presented by James Lamberti on the Orion panel at
SES NYC. Considering that this may be one of the most significant studies on Universal Search queries to date, I think it is worthy of sharing some of their findings with Search Insider readers.
Here is an overview of methodology and findings taken from both Lamberti's talk, and data from the PPT presentation.
Lamberti first explained how comScore's qSearch data was utilized, and
included queries, searchers, search pages, and Universal placements within the set. Queries were divided into two buckets: Those containing a Universal result, and queries with no Universal result.
With the Universal results set, several placements were isolated, including video, news, images, maps and stocks. Data was taken during one week in January (exact dates not specified), and the
findings represent full U.S. market projections.
advertisement
advertisement
Out of 1.2 billion queries surveyed, 220 million (17 percent) contained a Universal search results. 87 million people searched during the
measured period, and 58 percent saw a natural search placement. Within the Universal results, individual placements broke out this way:
Video 38% News 34% Images 19%
Multiple placements 15% Maps/Stocks/Weather 10% Lamberti also noted that the appearance of Universal placements was off when compared to the number of queries for the
respective category. So while video was found as a result 38% of the time, there was a higher share of search interest for video (53% against the entire set), with a segment that went unmet with a
video result. News share was also higher, with a 38% query share against 34% SERP penetration. In the other areas (images, multiple, maps, stocks and weather), the appearance of one of these
placements exceeded search interest.
I have two thoughts on why video results do not appear to be meeting search demand. The first is that there may simply be not enough video content to
meet all interest. It's early in the game for video right now, and this demand may be naturally met over time. Second, it could also be that Universal video is taking a bite out of Google's primary
revenue stream, and the company's strategists have pushed back on the number of video results shown in order to preserve paying clicks while paid video ads are being tested. This may sound like a
conspiracy theory to some, but I give this idea more credence based on a statement from Marissa Mayer in a recent New York Timesblog post. Here's how writer Saul Hansell paraphrases what Mayer says about testing video ads in
the sponsored results: "Text ads are not as effective on pages with search results that include images and video. The eyes of users automatically gravitate to the images more than the text, she said.
Now that Google's main search results pages include more images, video links and other elements, it is more appropriate, she argued, to have corresponding advertising formats [in paid search]."
So perhaps Google has backed off on serving video a bit, while a little more testing is in order. Backing off from cannibalizing the company's revenue stream isn't evil; it just might be a
matter of exercising good business sense while other possibilities are being tested.
Lamberti went over his final conclusions based on the research findings (to which I add my own thoughts
below):
"Organic search is increasingly critical." With the answer being provided right at the query level, search engines are now destination sites. Weather
forecast, on the SERP. Watch video, right in the SERP. Get the latest stock quote, right in the SERP. The data supports what many of us SEM evangelists have been talking about for the last year.
"Paid search will become more competitive." As an aside to this observation, paid search is about to get very interesting with a competing PPC video placement, and
I would like to see a follow-up to this report once Adwords video is in full swing, especially when compared to other organic Universal results. "Searchers are being
retrained on how to view search." Lamberti also made one other interesting quote, in that comScore found that 80% of searchers knew the difference between paid and natural results, though
this contradicts research conducted by Pew in 2005, as well as other studies. While previous studies on pure text-ordered results lists show higher trust, views, and searcher CTRs for the
highest-ranking assets, searchers are reorienting themselves to myriad possibilities on the page, from maps, to images, or even video results in the middle or bottom of the page. Again, Adwords video
will continue to add a tremendous dynamic to this retraining, especially if image icons are added into the paid video results (see Times article above). "Google-owned content is prominent today." The question here is: will Google go against its mantra of serving the most relevant result, or will it continue showing the best result
that it controls, or the natural result with the most potential for monetization? It's a question that Battelle keeps hammering away at, and one that we should all continue to keep asking as
marketers and critically thinking searchers. I think the question has already been answered in the search results to a slight degree. The greater question may be one of "how far" Google properties
will go to control the future SERPs, and what is the best balance of their own content versus the algorithmically "most relevant" content. Overall, this was a very thought-provoking study,
and the panel raised many interesting questions. While this gives us a first glimpse of the changing results landscape, don't expect the dynamics of search engine marketing and searching to settle
down anytime soon.
