There's a reason why Mark Twain popularized the saying, "There are three kinds of lies: lies, damned lies, and statistics." Data is malleable, and easily misinterpreted, if it's even gathered accurately to begin with. And that's assuming you even want the data to give you an honest answer in the first place. Sometimes, it can feel like you're relying on your gut to decide which truth is the right one.
When I wrote my blog post, “In Search Of… Those ‘Other’ Search Engines,” on October 21st, I was actually a little nervous. The blog post was only worth anything if I could look at two different data sets that differed from one another pretty substantially, and make an accurate judgment call as to which one was right. Or at least, ‘more’ right.
On one hand, comScore gets a LOT of press online for their monthly research on the ranking shares of various search engines in the United States. Their latest report for September 2014 gave Google a 67.3% desktop market share in the United States, with Bing and Yahoo trailing at 19.4 and 10.0% respectively. Those numbers were perfectly in line with comScore’s report for August, without a change of even a tenth of a percent.
Whereas, in my “In Search of…” blog post, I decided to favor the data released by NetMarketShare. I did so because based on my research into their methodology, I came to the conclusion that the NMS data would be more relevant to our purposes, and the purposes of our readers. Just as you wouldn’t want to air a commercial on TV, only for one couch potato to watch it 38 times, knowing how many searches each search engine handles isn’t nearly as important as knowing how many people they serve, which is what NMS focuses on.
But what’s the point of me going on about this? What’s the difference? NMS and comScore are in agreement about what’s blatantly obvious to just about everyone: Google is in the lead by a margin of victory most Kentucky Derby winners would envy. Where they disagree is just how much distance Google has put between it and the other two major contenders, Yahoo and Bing, and how the silver and bronze medal winners relate to one another. comScore says that Bing has nearly doubled up on Yahoo in market share (19.4% vs 10.0%, as noted), while NMS indicates that Google has a 71% desktop search share, globally. Meanwhile, Yahoo and Bing are digging around in the couch cushions for a combined 12% market share, with Yahoo winning by a hair.
It should be noted that the NMS study measures traffic on a global basis, while comScore operates at only the national level. So that’s a pretty big difference (and the two have different means of coming up with their data). But still, the only thing that’s really shaking up the global rankings appreciably versus the national is the presence of Baidu, which has a stranglehold on the search traffic of a country with over a billion people in it, giving them more than 15% market share. Aside from that, comScore and NMS agree on the top five (with Ask and AOL playing the roles of Blockbuster and Radioshack).
For the sake of comparison, let’s do a quick and dirty adjustment with the NMS data, subtracting out Baidu and pretending that it doesn’t exist (which, to be fair, outside of China that’s pretty much what the rest of the world does). So basically, we’re ignoring anything to do with Baidu, and just looking at the total population that uses all other search engines, and seeing what percentage of them uses each search engine. This is what we come up with.
To be fair, we’re still taking a study of global behavior, and trying to compare it to a study of American behavior. However, from what I’ve read and researched, none of the ‘Big 3’ has search shares in other countries that are so dissimilar to their American rankings that this global data average wouldn’t be pretty darn close to the truth for us U.S. folks.
There’s no question Google is dominating the market. And it’s even more ridiculous on the mobile side of things. If you refer to my previous blog post, you’ll note that NMS’s global data indicates that Google has a nearly 92% mobile search market share.
So why does any of this matter? Well, if we take comScore’s word on the matter, Google controls “only” two-thirds of the desktop search market in the United States, meaning it would make sense for marketers to spend as much as a third of their search advertising budget on Yahoo and Bing, combined. So for every 3 dollars spent, 2 goes to Google and 1 goes to Yahoo and Bing. But if the reinterpreted NMS data is closer to the truth, then it’s a different story.
That is a pretty significant difference. And that’s just with regard to desktop search.
There has actually been a lot of discussion in our own office about how much time we should dedicate to targeting search rankings on Yahoo and Bing, versus placating the big G, so this kind of data is integral to our operations, and how we serve our customers.
The reason why this all came up in the first place was because of an article that popped up on Search Engine Land three days after my “In Search Of…” blog post, entitled “Who’s Really Winning The Search War?” The author had run into the same confusion that I had while I was writing my blog post–the disconnect between the two different sets of statistics. He had a lot of resources on hand, and so after he mulled over things for a bit, he decided to do his own study.
His results? Google was the primary search engine for 80% of respondents. Then it was Yahoo at 8%, and Bing at 6%. That’s pretty darn close to the graph above. The ratio of Google users to combined Yahoo and Bing users comes out at 5.7 to 1, pretty close to my suggested division of marketing expenses. Even the also-rans were the same, with the exception of DuckDuckGo making an appearance (with a share of about 0.5%).
All in all, this made me pretty happy with the choices I made for my blog article. This independently done study revealed results that were much more in line with the data I’d sided with–Google with a fat lead, and Yahoo having a tiny edge on Bing–as opposed to what I’d perceived as being the commonly accepted reality, that Google held two-thirds of the market, with Bing beating out Yahoo by a hefty margin.
The general object lesson here is, when you’re doing research, think about how you’re going to use the data. What exactly it is you’re looking for, and why? This will define what sorts of data are most relevant to your research. My focus was on search users, not on searches, and that was what guided my choices.
And secondarily, what’s today’s take-away from revisiting this particular topic? Like it or not, the vast majority of potential customers are using Google. Not just 65%, but more like 80-92%.
For most companies out there, 80-85% of the time and money you’re investing in search marketing should be going towards Google, with the remaining 15-20% being split pretty evenly between Yahoo and Bing, with the former getting a bit more attention than the latter.
However, there are exceptions to this. The ‘Search War’ blog post noted that respondents over 45 years old tended to favor Yahoo and Bing more than the general population did, so if that age group is your core target market, then it may pay off to do some deeper research into the search behaviors of different demographic groups, to see how your attentions should be divided between different search engine sites for the best return.
Again, it’s all about you, and your customers.