Keyword “Not Provided” By Google Spikes, Now 7-14% In Cases

google-g-logoGoogle’s new encrypted search for logged in users now appears to be blocking a much higher percentage of search terms than when it initially rolled out two weeks ago. In some cases, it might even be higher than the 10% or less figure that the company initially predicted might be impacted.

Blocking Search Queries

Two weeks ago Google announced that it would start encrypting search sessions of anyone signed in to Google.com. In practice, this means that Google stopped passing the organic keywords that referred traffic to websites whenever users are logged in Google and conducting searches.

This change caused strong reactions, both in the web analytics industry (as this means that organic search becomes less trackable) and also in terms of whether Google was protecting privacy fully (since advertisers still receive this information).

Single Digit Impact Predicted

Google predicted that the change would impact 10% or less of searches:

Google software engineer Matt Cutts, who’s been involved with the privacy changes, wouldn’t give an exact figure but told me he estimated even at full roll-out, this would still be in the single-digit percentages of all Google searchers on Google.com.

In fact, the week the changes started rolling out, various reports put the “Not Provided” percentage — which is what those using Google Analytics see if terms are blocked — at around 2% to 3%.

High Single Digit To Above 10% Now Happening

However, as of October 31, we have seen a very significant increase on the Not Provided figure here on Search Engine Land. It’s not just us, either. Looking at data from several websites across industries, we see a range of 7% to 14% of total organic keywords now being blocked.

Below you see how the “Not Provided” figure has suddenly spiked for Search Engine Land:

Encrypted Search not provide search term

Below you can see how the percentage of Not Provided for the total organic keywords for November 1st to the site is above 10%, 12.87% in all:

Encrypted Search Analytics

The figure is even more dramatic, however, when you consider it as a percentage of Google-driven keywords. In other words, the 12.87% figure above means that for ALL keywords from ANY search engine to Search Engine Land, 12.87% of them were blocked.

As this blocking is only happening by Google, what’s the percentage of only keyword traffic from Google? That works out to 14.2%.

Of course, one might expect Search Engine Land to have a higher percentage of search-driven traffic than other sites. But as said, we’ve also looked at sites beyond Search Engine Land.

Danny Sullivan’s personal blog, Daggle, had 13.65% of its Google-driven keywords blocked. One non-tech site had 7.1% of its Google-driven queries reporting “Not Provided.” Another non-tech site we know of had 8.83% of all its keywords reporting as “Not Provided.”

We’d love to hear what others are finding — please comment below, if you’d like to share your figures.

Postscript: Conductor posted a few days after our report that it found 6.5% of Google-driven searches from high traffic sites they monitor had keywords blocked.

Rollout Still Happening; Percentage Could Get Higher

The rollout was supposed to take place over the course of several weeks. The process is still happening, and it seems as if it was suddenly enabled for more users on October 31.

Google wouldn’t confirm that, nor say how complete the rollout is at this point. In general, the company said:

As we noted, this change will occur over the next few weeks. Traffic figures will naturally vary depending on a website’s audience. What we provided was an estimate.

If there are significantly more people not yet being included in encrypted search, the percentage of Not Provided queries would likely grow over the coming weeks.

Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.

Related Topics: Channel: Analytics | Google: Analytics | Top News

Sponsored


About The Author: is the Founder of Conversion Journey, a Google Analytics Certified Partner. He is also the founder of Online Behavior, a Marketing Measurement & Optimization website. You can follow him on Google+ or Twitter.

Connect with the author via: Email | Twitter | Google+ | LinkedIn



SearchCap:

Get all the top search stories emailed daily!  

Share

Other ways to share:
 

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. Comments may also be removed if they are posted from anonymous accounts. You can read more about our comments policy here.
  • http://natangesher.com gesher

    So Daniel, why don’t you use some of your magic protektzia and get this fixed?

  • http://www.IgniteSocialMedia.com Jim Tobin

    4.45% in the last 7 days for us.

  • http://www.IgniteSocialMedia.com Jim Tobin

    Oops. Just rechecked since Monday (instead of the last 7 days) and it’s 11.53%. Today alone it’s 15.76%. Wow.

  • http://fr.linkedin.com/in/zinebaitbahajji zaitbahajji

    OMG, you’re right! my GA graph looks definitely like yours, with a huge increase of (not provided) keywords on the 31st of october.
    Encrypted data represented less than 3% of all the keywords generating traffic on my website before the 31st. Now it’s 7.45% !

  • http://www.jaankanellis.com Jaan

    Google should be embarrassed for rolling this out in the first place. not it thinks it can pull a fast one and triple the % of (not provided). Dumb.

    I feel bad for people that need keyword data to survive.

  • http://cfni.org CFN

    The graph for almost all the sites I manage started spiking on October 30th. 1.76% on 10/30, 3.26% on 10/31 and 7.56% on 11/1. It’s currently sitting at 10.55% on 11/2. That’s insane…

  • http://www.gjmarketingproductions.com Glen Jackman

    (Not Provided) traffic on the site I manage went up to 4.5% from .5% in the last few weeks. I’m hoping that this is not going to trend upwards and I can’t understand while Google would want to limit our abilities to use search data.

    I have a feeling that high tech website will be more affected by this due to more users being logged into Gmail and Google+.

  • http://www.seo-theory.com/ Michael Martinez

    Glen wrote: “…I can’t understand while Google would want to limit our abilities to use search data.”

    I should think the answer would be obvious. They are taking a lot of heat over user privacy issues and this is one way to address the concerns that have been raised.

    If one wants to play the conspiracy theory game, this could also be a good way to reduce the amount of “made-for-search” content that Google has been slogging through for the past few years.

  • Joe Horan

    Well think about it. Users get online to check their email, a lot of which use gmail. Then they go onto search for whatever it is they search for. Who logs out? Especially in gmail where it automatically refreshes when you get an email.

  • http://www.gjmarketingproductions.com Glen Jackman

    @Michael Martinez

    I understand the outside pressure for privacy but I can’t see how this keyword connects to the individual searching for the term.

    To my knowledge we are not able to identify a term and then find Gmail and Google+ users who searched this term so why not let us know how many people are searching for it?

  • http://www.katemorris.com Kate Morris

    Smaller site, in one week (last full week of October) it was 1.8%. This is not a high tech target community, more the day to day person.

  • Rob Snell

    But if they’re NOT blocking paid ad keywords, how can it be a privacy issue (other than eyewash)? I don’t understand. They’ll only protect my privacy unless someone isn’t paying, then my privacy is for sale? O_o

  • http://www.linkedin.com/in/danpeskin Daniel Peskin

    Just looking at the past two days of organic search traffic, 12.3% comes from “not provided”. The worst part is that this set of keywords under “not provided” has a 36.8% conversion rate. This makes up 13% of my conversions for the past two days. That is pretty damn significant in my opinion.

  • Paul

    Just checked a variety of my websites and the percentage of “not provided” ranges from 0.08% to 0.85% over the past week.

    Mind you that none of my websites have anything to do with seo/marketing/tech.

    It only seems normal that seo/marketing/tech websites have much higher % “not provided” as indeed most of those visitors are always logged into Google, but as I expected non-techies don’t seem to be logged into Google, so for “normal” websites I don’t think the impact will be severe or even noticeable.

  • http://www.squidoo.com/greekgeek G.G.

    At the same time as it’s doing this, Google has also apparently decided to discontinue HTML5 standard rel=”me” and rel=”author.” I implemented it as per their instructions a little over a month ago with a regular Google account (not Google+, as I have serious issues with how Google has implemented it).

    For a few weeks, my search results were showing my author name and icon correctly. Now they have been removed. This is true for most of the authors I know. Ironically, many of us were creating websites and posting articles before Google incorporated.

    It looks almost as if Google has decided not to count our authorship unless we’ve signed up with Google’s social network and shared publicly the personal information that Google has now decided we must share with the planet: our name, photo, and email address posted on *every page of our content* or at the least our author profiles.

    Who died and made Google the W3 Consortium? Why are we letting its redefinition of rel=”author” and rel=”me” prevail over web standards? Why does Google get to decide how we prove we’re authors of our own work?

  • http://www.netage.co.za Iblis

    G.G., what option do we have? Are we going to stop using it or optimising for it because of this? Or are we going to adapt to it the same way we do to algo changes?

    The bottom line is the bottom line. Google isn’t a charity or an NGO or open source. It’s a business. The more people use it, the more we have to cater to it. Like the quality score thing a while back…they’ve got us by the short and curlies to a certain extent…

  • TimmyTime

    “But if they’re NOT blocking paid ad keywords, how can it be a privacy issue (other than eyewash)? I don’t understand. They’ll only protect my privacy unless someone isn’t paying, then my privacy is for sale? ”

    Questioning Google’s motives??? I thought I had heard it all.

  • http://www.seomoz.org/users/profile/155620 Cyrus Shepard

    We have seen Google push harder and harder to get folks to create and log into Google accounts – Gmail, YouTube and Google+ for example. If they continue to succeed, we can expect the percentage of (not provided) keywords to rise over time.

    David Naylor did an interesting experiment to determine just how many users were logged into Google who visited his site. The results were much higher that the reported (not provided) keywords in his GA.

    Not to link drop, but traffic to my website notprovided.com has increased 10 fold, seemingly in reaction to the news.

  • http://www.advanced-web-metrics.com Brian Clifton

    Also seeing a sharp increase on “not provided” from 31st Oct in US and 1st Nov in Europe

    Impact on my clients varies widely from only 1-2%, upto ~10%. This appears to be down to regional differences i.e. the impact is greater from US visitors. I expect this difference to iron out and be more uniform globally pretty soon though…

    Best regards, Brian
    Author, Advanced Web Metrics with Google Analytics

  • Alistair Dent

    Daniel/Brian,

    The analysis I’d be interested to see (for one of these US sites that are seeing a large impact) would be to have prior keywords broken down into brand vs non-brand, to see if either one of those types has dropped by more and is driving more of the “not provided” keywords.

  • http://www.NJStangers.org ECMC

    This new update stinks. I saw a big jump yesterday into not provided, a total 15.6% of my keywords were not provided yesterday thanks to this update…. :(

  • http://www.artscienceweb.com Justin Howley

    We’ve had some issues with Panda, but we still see a nice amount of traffic. I see a 287.48% increase in (not provided) keywords and 2.10% of those are new visits. This means that the majority of our traffic is return users. It’s very disappointing in not knowing return visit keywords.

    Just thought I’d share with you all.

  • http://www.uscollegesearch.org CliffordB

    We had 21% yesterday. And just thinking about how 10% of our traffic is mobile and 3/5 of that is from an Android device is killing any analysis.

  • http://www.vincodo.com Vincodo

    The 15% is what I projected two weeks ago on my blog (http://www.vincodo.com/blog/2011/10/google-shuts-down-referral-tracking-privacy-protector-or-data-horder/). This figure is consistent with Google’s market share in the free email space. What we have to become concerned is Google’s coverage of the market should Google+ get accepted by the user community, where this could leap easily to over 50%.

  • http://dynamical.biz/blog/ Ani Lopez

    Last 4 days (oct. 30 – Nov 02) this is what I see:

    - Highest percentage seen is 9.17%
    - Lowest percentage 0.72%
    - Average 5.01%

    but growing fast, unfortunately
    all numbers here http://dynamical.biz/blog/seo-technical/non-provided-keywords-38.html

  • http://www.fahrenheitmarketing.com WilliamG

    Our agency site his 37.2% yesterday. Holding at about 32-33% today. (http://www.fahrenheitmarketing.com/seo/googles-big-lie-about-the-impact-of-keyword-data/)

  • http://www.bradfry.com/ bradfry

    A popular music video site is getting 1% (not provided) since Nov 1

  • Winooski

    I’m seeing an average of 9% “Not provided” on clients’ sites. I’m as dismayed as everyone. (gesher: I thought protektzia only worked in Eretz Israel. [;-)] )

  • http://cxinteractive.com Brian Robinson

    Wow, that graph looks exactly like one for our main client which is the only one I’ve checked but it’s still only 0.02% (was 0.01%) of Google organic traffic because the site caters to people who aren’t tech savvy and probably don’t have Google accounts.

  • http://www.thedr1ver.com Nick

    My personal blog, consisting of mostly tech related tips, has had a huge spike after the 31st.

    http://i.imgur.com/1sSNv.png

    18.06% of my traffic from yesterday alone came from (not provided)

  • http://europeforvisitors.com Durant Imboden

    Google Analytics shows just under 3.2 percent of our keywords as “not provided” since the first of the month.

    For what it’s worth, our editorial travel-planning site reaches a general consumer audience, with about half of our readers coming from Europe and the rest split between North America and the rest of the world. (Is it possible that searchers outside the U.S. are less likely to be logged into Google than American searchers are?)

  • http://www.universetoday.com Fraser Cain
  • http://about.me/alexedlund Alex Edlund

    I can only see this number growing, but keeping an eye on the adoption rate of Google+ will be key in its growth. Some users log out before making a search but that’s a minority. Don’t forget that this only affects Google.com, so if you’re running analysis, make sure to exclude other Google domains. Looking well beyond 10% for our properties.

    Seems to me that we are already seeing a number of new advertisers in the space, possibly created from the lack of insights with encrypted organic search referral data. The resulting increase in CPC’s comes as no surprise.

  • Davor Bomestar

    Our sites 7%, 14%, 20%, 33%.
    And growing.

    Can we all just remove and abandond +1 to lower the amount of users being logged in?

  • http://www.for-rent-nerja.com FRN

    Just checked my site, for November it’s 21.35% not provided! 1 word.. LAME!

  • http://www.edeninteractive.com E.I.

    We need a pluggin to block the “not provided” traffic segment. If enough of us built this..and make the experience crappy enough for the Google Logged in User. That user will get the message and log out– Google Loses.

    Everyone keeps crying about not being able to measure traffic. Well guess what. People visiting your site are guests. “No Shoes, No Shirt , No keyword URL” No Services. Just like the annoying “Please Login to access our site” – The message would explain our position and why we are blocking the organic traffic.

    Perhaps somebody out there would like to provide this service and create the plug-in.

    Searchengineman

  • http://gnarld.com G.D.

    Wow. For November average 25% not provided, with the highest not provided single day right at 30%. (Roughly 65% of my site traffic is from search engines)

    Back in October these numbers were around 3-5% mind you.

  • http://www.onewavedesigns.com Paul Coffman

    Arrrgghhh. WHat’s with google? This would almost appear to be an act of pure greed. They’ll give us search terms from paid add, and they’re still selling our info to companies, but we can’t see. What the hell?

  • http://finepointwriting.com Erika Gimbel

    On my personal site, this week’s average (not provided) was more than 35%. On another website I write for, which gets about 1.5 million visitors a month, the percentage was about 2.5.

  • http://www.thebauergallery.com Christopher Bauer

    In the past two weeks, my non-tech website is running a cool 17% (not provided) and thats from ALL search engines combined…

  • Frankie

    New clients website currently at 26.74% , I don’t know how to keep explaining to my clients to “just trust me, you’re being searched for”. This is shenanigans.

  • e

    Today, November 15 – (not provided) 402 19.98%
    This is very frustrating to say the least.

  • http://www.findwell.com F.W.

    It started off small in October on our site, and on Nov 1, quickly climbed to 11% of our searches having the keywords hidden.

    Not cool Google. “We’re not going to share the keywords with the publishers of these sites for the protection of your privacy. They might do something malicious with it. Exceprt for us, since we can still keep using your keywords to sell ads through our own network…”

  • http://searchquant.blogspot.com S.Q.

    One data anecdote: for my mainly-SEM blog, 73 of the last 110 Google organic visits were stripped of the keyword. That’s 66.4%.

    I do think Google will find its way to a point where for mainstream websites, over half of keyword data is stripped.

  • James Waulk

    Thanks for the great article.

    I am getting 100% “Not Provided” on my site I am currently working on at BestFreeNet.com.  It is my first site and I am just now getting into SEO.  I get very little organic search traffic less than 20 hits a day but they are all from Google and not a single one of them tells me the search words.

    This is vary frustrating because my site covers many different categories of free products and services that can be found on the internet. and without this data I do not know how to find out what people are searching for to find my website, so I know which categories to improve on.

  • http://twitter.com/jmz360 James

    93% of my traffic comes from search engines. 27.85% of that is ‘Not Provided’, my site is aimed at web developers and I guess most of them are signed into google when they search…

  • http://www.facebook.com/people/Tim-Eansor/100001889198457 Tim Eansor

    We were seeing around 3-6% which I was sort of ok with, because we were getting the numbers back through a couple of reverse engineering techniques from various blogs, but now our percentage is this:           (not provided)    10.1%      121 visits!

    How can anyone run a success business online with out knowing what your top converting keywords are? HELLLOOOOOO GOOGLEEEEE? you have cut your own throat here!

    Not with Google at ALL on this one. Basically they are trying to hide there network from other 3rd party sources that are trying to utilize Google for there own products and tools, however they have almost completely wiped out the hand that feeds… their CUSTOMERS!

    Bing anyone? anyone? Bueller?

  • http://twitter.com/TheSaganaki Jason Poulos

    Would using something other than Google analytics allow you to see what keywords are used to reach your site? I’ve only used Google analytics, so would different tracking software solve this issue?  

  • thomasfals

    Google is not sending the referrer data along at all so there is no way of getting hold of the data regardless of platform used.

  • http://www.facebook.com/profile.php?id=624594102 Neil Pearson

    35% not provided…

Get Our News, Everywhere!

Daily Email:

Follow Search Engine Land on Twitter @sengineland Like Search Engine Land on Facebook Follow Search Engine Land on Google+ Get the Search Engine Land Feed Connect with Search Engine Land on LinkedIn Check out our Tumblr! See us on Pinterest

 
 

Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States

Europe

Australia & China

Learn more about: SMX | MarTech


Free Daily Search News Recap!

SearchCap is a once-per-day newsletter update - sign up below and get the news delivered to you!

 


 

Search Engine Land Periodic Table of SEO Success Factors

Get Your Copy
Read The Full SEO Guide