Skip to content

20% of Searches on Google are new… another SEO Myth?

I got a marketing email from Hubspot today, with a headline about detecting “bad seo”. It then made an argument based on the “fact” that 20% of the queries on Google’s search engine have never been searched before. That was enough for me to recognize Hubspot as a “bad seo” vendor.

The 20% number is pure Google propaganda. “Good” SEOs aren’t so easily led astray.

I assume that at the time Google announced that “20%” statistic, it was good for Google for frame the issue that way. Tomorrow, when Google wants to give an alternative impression (for whatever reason), it will be easy to re-position that data. SEO people need to be smarter than that. Don’t believe everything Google says. Better yet, “critical thinking” is essential for SEO.

Maybe today I search “shoe repair Fort Lauderdale, Florida” looking for a shop to resole my awesome c. 1981 biker boots. That’s a new query and is counted.

Next I search “shoe repair Fort Lauderdale Florida” and, technically, that query has never been searched before. Merely a comma different, yet “not the same”. If I’m looking for a large number for my “there are a lot of new queries every day”propaganda, I increment the count.

Maybe someone searches “shoe repair Ft Lauderdale Florida”.. another “brand new, never before searched phrase”. And then how about “shoe repair Ft. Lauderdale Florida”, with the extra period? The count continues to rise.

We can see how very quickly we get way up there in numbers….especially when we are reporting percentages:

“shoe repair Fort Lauderdale Florida”
“shoe repair Fort Lauderdale, Florida”
“shoe repair Fort Lauderdale Fl”
“shoe repair Fort Lauderdale Fl.”
“shoe repair Fort Lauderdale, Fl.”
“shoe repair Fort Lauderdale, Fl”
“shoe repair Ft Lauderdale Florida”
“shoe repair Ft Lauderdale, Florida”
“shoe repair Ft Lauderdale Fl”
“shoe repair Ft Lauderdale Fl.”
“shoe repair Ft Lauderdale, Fl.”
“shoe repair Ft Lauderdale, Fl”
“shoe repair Ft. Lauderdale Florida”
“shoe repair Ft. Lauderdale, Florida”
“shoe repair Ft. Lauderdale Fl”
“shoe repair Ft. Lauderdale Fl.”
“shoe repair Ft. Lauderdale, Fl.”
“shoe repair Ft. Lauderdale, Fl”

Suddenly 20% is not a very surprising figure. There are many, many more variations but I stopped documenting permutations above because quite frankly, the way Google has been over-reacting to market dynamics lately, I’m afraid my site will be banned for spamming the local shoe repair SERPs for Ft. Lauderdale, Florida.

Is it really remarkable that 20% of searches on Google have never been searched before? I honestly don’t know, because Google only released that junk statement and no good supporting data about how it was obtained. Junk science, or junk SEO, in this case. Meaningless as best, and distracting at worst.

If you start your SEO newsletter with some assertions based on such “facts”, I think you lose credibility as an SEO.

Critical thinking, folks. Think of it as a necessary tool for survival in web marketing and SEO.


  1. Thank you professor Andrews for the reminder to think, although it’s so much easier to eat the mush on the spoon in front of me, hope your biker boots are doing swell.

    Monday, October 10, 2011 at 12:13 pm | Permalink
  2. Brian Kenyon wrote:

    I wouldn’t label HubSpot as a “bad seo” company just because of the 20% statement. I reviewed their service over a year ago before their recent investors jumped in and felt it was a very thorough service.

    Tuesday, October 11, 2011 at 1:20 pm | Permalink
  3. john andrews wrote:

    Well Brian just as you would not label them “bad” they probably shouldn’t label others as “bad” as a way of marketing themselves. On one hand maybe they don’t believe in keyword meta tags, while they do believe Google’s propaganda. Is it really their place to say what is and isn’t correct? Google says unequivocally that kw meta tags are useless… and that 20% of searches are brand new. Which one is more correct… or less false?

    Tuesday, October 11, 2011 at 8:22 pm | Permalink
  4. otto wrote:

    Hi John,

    Not sure how 20% new and bad seo are related.

    The way I see it, there are loads of new products coming out daily. Existing brand new product means loads of new search terms.

    As you have shown, with plenty of possible variations, depending on what counts as unique.

    If you take into account search terms disappearing, a larger percentage becomes new.

    In terms of SEO, all this new sounds like opportunity to me.


    Wednesday, October 12, 2011 at 2:45 am | Permalink
  5. otto wrote:


    meant to say: existing brand new product means a lot of new search terms

    Wednesday, October 12, 2011 at 2:46 am | Permalink
  6. john andrews wrote:

    @Otto I was referring to blind acceptance of the claims put forth by Google, without questioning actual meaning. A straight up “wow.. 20% of queries were never searched before” is deceptive… and should not be repeated by an SEO company without some clarification about the importance of that. It clearly does not mean 20% of queries represent new searcher intent, as I demonstrated above.

    As Google continues to expand it’s switching e.g. “results are shown for the query we decided you might have meant, instead of the query you actually entered” this is more and more important.

    Is that 20% actually 10% Or maybe 5%? Or perhaps even less than 5% when you remove the duplicated intent? And if it is less than 5%, is it remarkable?

    Tuesday, November 1, 2011 at 12:14 pm | Permalink