{{ searchResult.published_at | date:'d MMMM yyyy' }}

Loading ...
Loading ...

Enter a search term such as “mobile analytics” or browse our content using the filters above.

No_results

That’s not only a poor Scrabble score but we also couldn’t find any results matching “”.
Check your spelling or try broadening your search.

Logo_distressed

Sorry about this, there is a problem with our search at the moment.
Please try again later.

A few weeks ago Google started to encrypt search data for logged in users. This essentially means that website owners will see no keyword data for visitors referred from Google.

We didn’t think too much of it at the time, given that just 0.68% of our visits were affected in the period immediately after this initiative came into play.

However, it seems that things have significantly worsened in November, at least for Econsultancy. 

Why keyword data matters

I have been looking at some of our top performing pages, to try to understand what makes them consistently popular. Ultimately my goal here is to identify certain words and phrases in search queries, in order to optimise our content (and therefore our business) strategy for 2012.

Knowing what phrases perform the best is crucial to making our content and our website better. Which keywords ultimately lead to sales? Which ones result in high levels of social sharing? Which ones have the highest bounce rates? 

Why Google has chosen to restrict our view is beyond me. It seems to go against its aims to improve the web experience.

Let's look at the detail...

Page-level keyword analysis 

One of our most popular pages is a blog article called ’25 brilliant examples of Facebook pages’. Despite the fact that it was published in August 2010 it continues to attract thousands of visits and pageviews every month. Last month the post generated around 9,000 pageviews. 

Econsultancy has an internationalised website. We currently have US and UK versions of our site (econsultancy.com/us and econsultancy.com/uk). This is a little annoying when looking at the analytics (and also because the Twitter counter only tracks US retweets, rather than grouping in the UK action too), but it is revealing too, as we will see.

Below are two charts for the US and UK versions of this article. The date range covers the past month, during which time Google started to hide keyword data for logged in users. 

US page

Suddenly we’re seeing that ‘not provided’ sits at the top of our referring keywords for US visitors, representing 431 pageviews of the total 3,651 (that’s about 12% if my mental arithmetic is correct).

UK page

You can see that this issue isn’t quite so serious in the UK, with 82 in the ‘not provided’ field against a total of 2,419 pageviews (around 3% or thereabouts).

So the first takeaway, based on this small sample, is that the problem is magnified by a factor of four in the US, compared with the UK. I guess our US visitors are four times more likely to be logged into Gmail, or some other Google account.

November spawned a monster

What happens if we restrict the date range to just November? That’s when we think Google rolled this out on a widespread basis. The data below shows that the story considerably darkens... 

US page

 

A staggering 362 pageviews out of a total 1,138 are showing no keyword data. That’s around 33%. One in three search referrals from the US do not pass on any search query data. My jaw has dropped to the floor.

UK page

Again, the ‘hidden’ number has jumped. 70 pageviews out of a total 839 have no search referral data attached to them. Around 8%.

The big picture

What happens if we look at the site through a wider lens? Google sends Econsultancy.com around 40% of its traffic, so what proportion of that is now camouflaged? Let’s look at the entire site (including US and UK versions, and all pages), and let’s restrict the date range to the first 11 days of November…

 

Wow. The hideous ‘not provided’ is now showing as our top search referral. It has usurped our brand terms. It represents more than 10% of our total keyword referrals. And I fear it will worsen, as Google convinces (or requires) more people to log in to its products.

I guess the caveat here is that Econsultancy is a B2B website aimed squarely at the digital industry, so we might attract a greater proportion of users logged into the likes of Gmail and Google Plus (which could become a big deal as far as the rankings go, though social search shouldn't evolve at the expense of this kind of standard data analysis). But there is a problem here, any which way you look at it, and I know that we're not alone. B2C sites will be experiencing similar issues, even if the numbers look different.

So now it’s your turn. How does this look for you? You can use Avinash Kaushik's fantastic custom report for Google Analytics. Do let us know in the comments below, and it would be helpful to know what kind of site / audience you have…

Chris Lake

Published 11 November, 2011 by Chris Lake

Chris Lake is CEO at EmpiricalProof, and former Director of Content at Econsultancy. Follow him on Twitter, Google+ or connect via Linkedin.

582 more posts from this author

Comments (72)

Comment
No-profile-pic
Save or Cancel
Avatar-blank-50x50

James Gurd

Hi Chris

Thanks for sharing. I'm still convinced that this has something to do with a bigger commercial plan from Google as like you, and many others, I can't see how this has anything to do with protecting privacy. How does keyword information compromise privacy?

There have been a few 'conspiracy' theories around why Google would want to do this. If it doesn't apply the encrypted lens to PPC, arguably it could be a drive to encourage more keyword bidding.

Another proposed reason is that they want to restrict the amount of keyword data available to ad retargeting companies who mop up on display revenue by retargeting non-converting searches based on matching ad copy/creative/placement with the original keyword search. Could Google be wanting to get more people spending through its ad streams?

I don't know for sure, none of us do but it smacks of smokescreens to communicate to experienced search marketers and web managers that they're slashing the value of your keyword data to protect online searchers. If that was truly the case why would you have to be logged-in for that? Why wouldn't they just blanket apply the secure search regardless?

And here's my final conspiracy suggestion - it's a big elaborate ploy to encourage more people into Google accounts by making them think their privacy is at stake, to drive the + network and make us all complete and utter Google slaves.

cheers
james

over 4 years ago

Avatar-blank-50x50

Neil Yeomans

Hi, this was a really interesting article & enjoyed looking through the data, so thanks. With regard to your point below:

"So the first takeaway, based on this small sample, is that the problem is magnified by a factor of four in the US, compared with the UK. I guess our US visitors are four times more likely to be logged into Gmail, or some other Google account."

Most signed-in users in the UK default to Google UK, and Google SSL is only applied to .com sign-ins. However, there are certain signed in users in the UK that do default to .com. This is why we are seeing a lesser proportion of (not provided) data in the UK.

over 4 years ago

Avatar-blank-50x50

Sebastian Cowie

The problem is definitely getting worse and there is a significant spike from the 1st of November onwards.

October 21st -> 31st provided 21,402 organic visits. 227 Visits (Not Provided) - Approximately 1% of total organic traffic from Google.

November 1st -> November 11th provided 21,345 organic visits. 1,340 Visits (Not Provided) - Approximately 6.3% of total organic traffic from Google.

That's an increase of 490%.

This is an international entertainment basd site and attracts traffic from primarily the US, but has a large fan base in UK, Germany, Canada and France.

How long will it take before we lose access to 50%+ of our data and we're forced into adopting PPC campaigns to assist in our SEO endeavours?

over 4 years ago

Anna Lewis

Anna Lewis, Google Analytics Analyst at Koozai

Thanks for sharing the data, it's good to see a how different sites have been affected, but very worrying how much this number is creeping up and up.

At Koozai this week we have seen (not provided) account for 15% of organic traffic, 77% of which is from the US.

It's creeping up day by day and UK is also rising too. I'm really not looking forward to the time when UK searches are encrypted too.

over 4 years ago

Chris Lake

Chris Lake, CEO at Empirical Proof

Thanks Neil, you're right. Definitely worth factoring in.

James - I'm wondering if Google Analytics Premium users would get to see this data? Presumably not, but that's another theory to throw into the pot.

Sebastian - I guess you can live with 6.3% but as you say, when it hits 50% how much dumber are we all going to feel?

Anna - it could be that B2B sites have higher rates. 15% is far from inconsiderable, and given what you do I bet you're all rather annoyed by this!

over 4 years ago

Avatar-blank-50x50

Steven Clayton

Hi Chris,

A really interesting article, thank you. The data Google provides us is awesome when you consider it's free. I am not surprised that Google is taking steps like this, however worrying, when they could make a lot of money.

Thanks

Steve

over 4 years ago

Avatar-blank-50x50

Phillp Marquez

I seem to recall Google saying that keyword data can still be gathered reliably from Google Webmaster Tools -- it's just one extra step for the analyst or for a person that has access to GWT for the domain.

I haven't had the time to attempt some sort of comparison between GWT and GA data to determine if this is actually true, or if it's just the same data with the "(not provided)" entries culled, but would be interested if someone else has had the time to do so...

over 4 years ago

dan barker

dan barker, E-Business Consultant at Dan Barker

thanks for the great post, Chris.

I really hate that they've done this. A sad step, and I'm sure many within Google don't like it either.

It's very sad the way some parts of the web have changed over the last few years. Amazon, Facebook & Apple created their giant content/communication malls, sat within nicely walled gardens.

Google stayed standing there for a while, like an old public city centre.

Now it looks like they're buying a few plots of land too, diverting a few roads to suit their needs, and erecting a trellis around the whole thing.

over 4 years ago

Avatar-blank-50x50

batracio

You can get the referral data if you enable https in your site. Google passes referral information between http->http and https->https, but not between https->http. And that's a privacy feature. I think that in the long term, google's plan is to force everyone into https, and I can't see that anything but beneficial.

over 4 years ago

Avatar-blank-50x50

Steve

@Steven... I think you are on to something. It seems like only a matter of time before some of these internet giants start charging. At the same time, they seem to be doing alright without the small man paying.

over 4 years ago

Avatar-blank-50x50

Brandt Dainow, Director at ThinkMetrics

It is clear that the degree of impact this will have varies largely according to the type of site. I have checked 20 different sites, .com and .co.uk, B2B and consumer covering travel, clothing and B2B products/services. All of these sites get more than half their traffic from Google, most around 70%. The impact varies from 0.3% to 0.8%, but never reaches 1%. For example, on a B2B site, of 100,000 unpaid search visits, a total of 135 had keywords blocked. It is worse for .com than .co.uk. I guess it depends on your content and how people find it.
What concerns me the most is that the percentage of visits blocked forms a neat curve if you graph it. I find it hard to believe that the percentage of people signed in could vary so neatly - surely a given percentage of people sign in at any time? I'm not sure that Google are applying this consistently. It wouldn't surprise me if this turns out to be another dumb initiative they haven't thought through and slowly/quietly drop.

over 4 years ago

Avatar-blank-50x50

Isaac

Awesome - not!

Oct 21 - 31 - 2% not provided
Nov 1 - 14 - 11% not provided

over 4 years ago

Avatar-blank-50x50

Peter Young

Nice summary Chris, and certainly its something I think many people in the industry are starting to see.

We at Brilliant have used Atlas to track organic search data for a while, however the figures both you and the wider industry are reporting are way beyond the single digit factors Google first mentioned at launch. As many people said the 'privacy' issue appears to be very much a red herring for what is definitely a bigger picture - and one has to assume Google are working on somethign to address this as the last thing they will want is for things to return to the days of the Wild West - somethign one can't help feeling may happen.

Whether or not they will charge for data moving forward - I have to say I would doubt that given the whole can of worms they would open up from a monopoly perspective further to which lets face it they scrape OUR content every day so theres a bit of a pot kettle black there.

over 4 years ago

Pat Wood

Pat Wood, MD at TruffleShuffle.com

I've just checked our sites stats, and so far, not too bad... gone from 0.06% in the last two weeks of October to 0.3% in the last 14 days.

We run a retail clothing website and our demographic is made up of normal people... most of whom have no idea that Google+ exists.

I think the scary stats are for those whose target customer is an internet professional - seeing as this demographic makes up a large proportion of the few people who seem to have signed up for Google+ and are therefore consciously logged in.

So, in my humble opinion, due to my broad demographic of customers, as long as Google+ remains the insignificant outsider, our keyword data remains intact!

over 4 years ago

Graham Charlton

Graham Charlton, Editor in Chief at ClickZ Global

@Pat Yes, our user base does make the stats scarier for us, but presumably most of your traffic is UK-based. When Google begins to encrypt UK searches, things may change.

over 4 years ago

Avatar-blank-50x50

JM Camiade

I don't understand...
By hiding keywords, Google disables anyone's capacity to choose an efficient Adwords campaign. I want to pay for keywords I know as being the best ones for conversions, I won't pay for "SEO-hazard"!
Is Google search the Adwords' worst enemy?

sorry for my "frenchie english"

over 4 years ago

Avatar-blank-50x50

Brent Nau

In response to batracio comment I wonder if you would get all the keyword data if you redirected a "Google Log In" user to the https version of your web site.

over 4 years ago

Avatar-blank-50x50

Ciaran Oliver

What I don't know could fill a warehouse. But what I do know is the Google is pushing hard with Google+ and I expect there will be a way of accessing the data through some kind of payment in the future, when we are all really squeezed for good information.

The world of Analytics has some pretty big players using free data and this is where the free ride halts.

over 4 years ago

Gerry White

Gerry White, Techical SEO Director at SiteVisibility

@Pete Young - see there are far more potential personal privacy issues with using Atlas, we used it at RazorFish and whilst I am sure your responsible, there is potential for privacy issues.

When you are searching for medical information related to cancer or similar and then go searching for health or life insurance you can see the ramifications.

From an analytics background it was surprising how clustering people who searched for keywords would help with analysis, these clusters are going to be far less accurate with anything approaching a 33% drop in data, I don't truly understand the privacy issues though.

over 4 years ago

Avatar-blank-50x50

Keith

Still not seeing anything near 33%, part of me thinks GOOG is still playing around with this (and that is what is worrisome)

over 4 years ago

Avatar-blank-50x50

GrumpySEO

WMT doesn't give you the same level of keyword referral data as Analytics, and the numbers reported there are way off in any case. Google doesn't want you to have this information. That's what this is about.

over 4 years ago

Christopher Rose

Christopher Rose, PPC Marketing Director at Rose Digital Marketing

The main problem with Google is that is has undergone a major change of attitude over the years.

When it started, its goal was to help people obtain better search results and "do no evil".

Now it seems to want to only help itself and has become a major part of the machine it once stood outside of; "do no evil" has been completely forgotten.

I have stopped using it as my first choice for search and am now using Bing and Wikipedia to get better, more relevant results for many searches.

Maybe it is time for the "Occupy" movement to diversify a bit and occupy Google!

over 4 years ago

Avatar-blank-50x50

Abby Hardoon

Thanks Chris for taking the time to write and pointing out this issue.

I've just crossed check Google Analytic stats for my company and it appears that we are not as effected. This is probably due to the fact that we are focused on UK traffic - data not provided accounts to 1.3% of total "non-paid" traffic.

I can't really fathom the reason as to why Google would be doing this other than that armed with such data a website owner can optmise their website and via SEO activities able to climb the natural listing ranks. But then again that also doesn't make sense as traffic is already being sent using previously known non-paid keywords - so whats the point of masking it all of a sudden?

One thing that's for sure, Google gives this perception that whatever they do is for the good of all internet users - and that's rubbish - its all about their bottom line most of the time. Let me explain although this is not related to this topic, when the joint venture between Google and BT started giving out free .uk domain names in the last 2 years or so they were bidding on the brand keywords of the company I work for !! Not only that they were also bidding on the keywords of the products that we sell causing bid prices to go up - talk about biting off the hand that feeds you !

So in this instance I wouldn't be surprised if there's a financial reason behind all this.

over 4 years ago

Avatar-blank-50x50

@jwlls

Google is monetizing Analytics - they will make you pay for results.

Makes perfect sense - they have two traits as a company: 1. Monetising products and services they own 2. Improving the functionality and connectivity of information.

Think about every product/service they've owned - it's a no-brainer

over 4 years ago

Partha Bhattacharya

Partha Bhattacharya, Owner at 2WebVideo.Com

Nice insight Chris. Thanks.

Perhaps using open source analytics like Piwik is a solution...as also the basic solutions like AWStats, Webalizer, etc.

over 4 years ago

Gerry White

Gerry White, Techical SEO Director at SiteVisibility

Partha - this won't help, there is a post by Peter Young on Hollistic search which shows that the referral data simply isn't there - it isn't Google Analaytics not tracking it - it would be ANY analytics package.

over 4 years ago

Avatar-blank-50x50

Sanjit Chudha

@ James Gurd + @ Chris Lake - spot on post Chris and thanks for sharing this info; ditto your comment James.

I agree with you - My gut instinct is that this is all about driving sales for Goggle Analytics Pro, plus a drive to increase ad spend through Google. No other explanation makes sense.

over 4 years ago

dan barker

dan barker, E-Business Consultant at Dan Barker

Re the "Do no Evil" comments - sadly the employee who came up with "Do no Evil" left a few years ago.

Wonder if you could plot the change in sentiment to the point he left?

(it was Paul Buchheit)

over 4 years ago

Partha Bhattacharya

Partha Bhattacharya, Owner at 2WebVideo.Com

Gerry - yes, I see your reason. Thanks.

over 4 years ago

Avatar-blank-50x50

@jwlls

@ Dan Barker, @ Christopher Rose - I don't see how monetizing a service which costs them money to run and maintain is evil?

It is after all a business-to-business service.

"Do no Evil" never meant Google was a NFP - it's listed which means it serves in the shareholders best interests

over 4 years ago

Avatar-blank-50x50

Richard

I'm not seeing "(not provided)" anywhere in my entrance keywords report for my home page for the whole of October. In November so far I'm seeing "(not set)" accounting for 123 pageviews from a total of over 9,000 - not sure if that's the same thing but not a huge percentage if it is.

Not defending Google btw, it seems a mad decision.

over 4 years ago

dan barker

dan barker, E-Business Consultant at Dan Barker

@jwlls I think that's what they call a 'straw man'.

over 4 years ago

Adam Cranfield

Adam Cranfield, Chief Marketing Officer at Mynewsdesk

Excellent and useful post, Chris. I don't suppose anyone from Google would like to comment on this?

over 4 years ago

Avatar-blank-50x50

Yannis

Hiding information on the back of "privacy reasons", with a view to charging in the future... in which case people will be buying the information anyway, rendering any Google argument about privacy pretty much irrelevant.
It all stands to financial reason, but I feel it is too early (their social presence is still low to start hiding data this early) and that they would be risking court action in countries where abuse of position of power is not received well..

over 4 years ago

Avatar-blank-50x50

Kiril Bunin

I think you answered your own concern in the last paragraph. E-consultancy is not just a general b2b site. It's a website solely focusing on internet marketing and web analytics meaning that majority of your audience would use at least Google Analytics or Adwords (or Gmail & Google + as you mentioned)

The same situation can be observed with any e-marketing focused content site. The general situation is not so dramatic at all, and from what I've seen is close to mentioned by google 1%-2% of all keywords.

over 4 years ago

Gerry White

Gerry White, Techical SEO Director at SiteVisibility

@kiril - today, but wait a month or two with the rolling out of G+ and it being all over everyones results even my non-techy, non-facebooking relations have created a profile as they use gmail as their primary (non work) account and it pushed them into it...

over 4 years ago

Avatar-blank-50x50

Matthew Tod

What is the problem? I'm sorry but I really don't understand why this is such an issue!

I stand prepared to get flamed, but let me explain.

Instead of having a notional 100% of all activity recorded in the data you now have a SAMPLE of data. The sample in the worst case scenario covers 66% of all traffic, so compared to many techniques this is a huge data set that offers a rich environment in which to glean appropriate insight.

So as I see it you can still find out top keywords, create segments based upon them and do all the usual analyses you would normally do.

To get an estimate of the total number of searches you now have to use some basic stats instead or reading off the totals. These techniques will give you a probable range of the total number of visits at a level of confidence.

So regardless of the nature of the reasons behind the Google decision there is still sufficient data in Google Analytics to help you make highly informed decisions.

So yes we all have to change and upgrade our techniques but this is hardly a disaster.

over 4 years ago

Avatar-blank-50x50

Wordpress Training

Excellent post. I am curious to see Google's response on this. I am going to start paying attention to this now because knowing which keywords perform better is sort of the lifeblood of my business. I hope this isn't permanent.

over 4 years ago

dan barker

dan barker, E-Business Consultant at Dan Barker

Thinking about this a little more, here are 2 silly ideas for tools to help overcome this:

1. NEAREST NEIGHBOUR

Build a tool that tracks the search terms bringing visits to all pages on a site, and also captures the location of the visits which bring people from those search terms.

Whenever someone visits with a '(not provided)' search term, the tool simply inserts the most likely /real/ term that has brought people to the site when they've landed on the same page from a similar location.

2. ALL-TIME TRACKING

If a visitor is known to the site, this tool would simply reinsert their last 'known' search term as the source keyword, instead of (not provided)

Both of these tools would be fairly easy to build. They'd both be slightly more invasive, and privacy-creeping than the current system of just looking at genuine search terms, but I bet quite a few things like this start to crop up now.

dan

over 4 years ago

Avatar-blank-50x50

@jwlls

Giving it some more thought, as I am convinced that they will be monetizing it (it would go against their business strategy to restrict access of information) - and without doing much research in to it - I would be interested to see how they would do the costing structure for this information.

I think it's possible they could build two models, just like with AdWords - where people are accredited for their skill of AdWords and can administer multiple accounts (i.e. a professional package where one SEO professional can receive access to multiple website analytic pages - sort of like how it exists already) and a standard company account where people, as a company, can pay to receive detailed keyword information for their own website.

It will be interesting to see.. Not that it's worth me rambling or conspiring about, especially considering SEO isn't really my past-time, passion or pay. @jwlls

over 4 years ago

Dave Chaffey

Dave Chaffey, Digital Marketing Consultant, Trainer, Author and Speaker at SmartInsights.com

Hey Chris, I enjoyed the way you worked in a Morrissey reference into a post on analytics - nice one!

Dave

over 4 years ago

Chris Lake

Chris Lake, CEO at Empirical Proof

@Dave - thanks and good spot ; )

@jwlls - I think it's perfectly reasonable for Google to monetise Google Analytics. I'm less sure about it controlling the flow of data. Also, this is apparently a privacy-related initiative, so I'm not convinced that it is about money as such...

@Matthew - a sample comprising of 66% of the total isn't the end of the world, but things may worsen further, and for a B2B site with relatively low search volumes a point may come where the sample data is considered flaky. At any rate, I preferred it when we didn't have to resort to guesswork.

over 4 years ago

Avatar-blank-50x50

Alex Wares

Good post Chris.

We have been keeping an eye on this.

So regardless of the rights and wrongs or motivations behind Google's decision, is this good news or bad news for the search marketing industry?

over 4 years ago

Avatar-blank-50x50

Andy Atkins-Krueger

Matthew, the problem with this change is that you have no idea how far Google will go with it.

Secondly, I would consider those that are "signed in" in some way as a different persona or segment which behaves in a different way - so the sample which is left is no longer representative. They may be less frequent users and less experienced web users, for instance and behave quite differently. You may find your best customers are always in the signed in segment, but have no data on them.

This change is criminal. Google is saying that websites may not have the data (even annonimised) for users which are arriving at their own websites.

Google, however, still has the data and, as we know from leaks such as the AOL case, can readily track it back to individual users.

The privacy situation is worsened by this - not improved.

Here we go with big brother Google - as petitions are so popular perhaps eConsultancy would like to set up a petition for website owners to ask Google to step back from this dangerous attack on web freedoms.

over 4 years ago

Avatar-blank-50x50

Sally Kavanagh

This must be amount money - it certainly isn't about privacy or Adwords data would be affected too. I don't have a problem with GA being monetised so long as it is affordable to small sites but how can Google start charging for data which it has now said it is withholding to protect searchers' privacy? Google must have a more subtle plan, or it has backed itself into a hole - and I can't believe that.

over 4 years ago

Gerry White

Gerry White, Techical SEO Director at SiteVisibility

The Adwords part is a bit different - it isn't taking what the user typed in, so much as it is what you bid on - I think the differentiation here is that this can be carried in a parameter to any analytics package, so there isn't the same inherent risk - it doesn't bring in referral information so it doesn't show any additional information that isn't carried in the parameter (I could be wrong here, if someone can post the same sort of information that Peter Young posted on hollistic search (the PPC referral information) ... from logged in.

over 4 years ago

Avatar-blank-50x50

Dan Lindop, Director at Surefire Media

Just checked some of our client sites and most are showing just under 1% of keyword data for November isn't provided.

I'm also getting some searches showing as (content targeting) without keyword info. Can anyone shed some light on this?

over 4 years ago

Avatar-blank-50x50

Lukas Meijer

Devil's advocate:
In a way there can be good in this measure benefitting the internet information as a whole. So far, the situation was as follows:

-> Company-A has product-X for sale. To boost sales it optimises its site for key-1 and key-2.
-> Company-B has similar product-Y for sale. To boost sales it optimises its site for key-1 and key-3.
-> People search for key-2, but it turns out they also search for key-3.
-> Company-A and Company-B now change their optimisation to key-2 and key-3, to optimise results.

The result is that the difference between product-X and product-Y has become less obvious for people. You can even ask the question whether the search result still reflects the best result for them.

From this point of view, you may conclude that removing the tools to optimise websites will mean that companies have to resort to promoting the specific qualities of their product (once again). This may well lead to more textual differentiation on the web producting more specific search results in search engines.

Curious about your opinion on the subject!

over 4 years ago

Avatar-blank-50x50

Leigh

This isn't Google being evil, or anything like that. This happens because it is part of the HTTP Specification.
http://tools.ietf.org/html/rfc2616#section-15.1.3

"Clients SHOULD NOT include a Referer header field in a (non-secure) HTTP request if the referring page was transferred with a secure protocol."

So if your page is HTTP you can't get the keywords because the http referer is not sent. Doing so would be a breach of security. The simple solution to this is to make your site https as well.

over 4 years ago

Avatar-blank-50x50

Ad

It makes me chuckle how many people assume there are secret, ulterior motives behind this kind of change. It's almost certainly an unintentional or unavoidable by-product of some other Google policy or technology change.

The interesting question for me is, how successful will Google be in attracting users to take up its social media offerings over the coming couple of years. That will determine how many users are logged in to Google, and consequently how many search results are encrypted.

Also, somebody should carry out a test by referring some traffic to their own site while logged in and while not logged in - to establish whether this is really the cause and whether behaviour can be predicted.

over 4 years ago

Doug Kessler

Doug Kessler, Director at VelocitySmall Business Multi-user

Google Enters Sinister Phase Two of World Domination Plan.

Ouch.

over 4 years ago

Avatar-blank-50x50

Nick Armstead, SEO + PPC Consultant at Orantec

I dont understand how they think people wont go to others. If Bing start offering better search result data everyone will be forced to use thier data to optimise for, which will mean we would promote them over Google as it would be the only way we know how to rank.

I do totally agree with some above comments, this reekes of some bigger plan.

over 4 years ago

Avatar-blank-50x50

Duncan Smith

30% is nothing, try 90%!

An ill thought through implementation of the new legal requirements set out in the The Privacy and Electronic Communications (EC Directive) (Amendment) Regulations 2011 a.k.a. "Cookie Law" could see your GA tracking fall to just 10% of pre-opt-in rates.

Tracy Brock's FOI request to the ICO shows 'hard numbers' demonstrating that their attemp to gain visitor consent for the placement and access to non-essential data (GA Cookies et al) has failed.

Negotiating privacy with your web visitor is going to become THE most important aspect of your web presence. If THEY don't perceive a benefit from letting YOU gather GA data, then they may not let you have any!

Please keep your eye on the W3C Tracking Protection Workgroup http://www.w3.org/2011/tracking-protection/ as they attempt to define how the TRACK in Do-Not-Track will work.

Have a look at my blog post here http://tinyurl.com/bsf87ed for a bit more info.on the DNT problem and why this really does matter for all our analytics.

over 4 years ago

Avatar-blank-50x50

Craig Mackay, eCommerce Optimisation & Conversion Manager at Dnata Travel part of Emirates Group

Hello, we have not noticed anything like this % being affected, perhaps as we have a dominant UK market as you mentioned in the article (< 5%).

Regards

over 4 years ago

Avatar-blank-50x50

Enrico

First, thanks for sharing both the correct data and the data related to a specific web page of your web site.

Unfortunately, there is a problem of transparency, here.

The correct result of the analyses conducted in this article is the one provided in the section "The big picture": 10% of keyword data has the "not provided" label.

Then, why did you put the 33% figure (the one related to a specific page of the website) in the title of the post?

As anyone can read from your article, the 33% figure is related to a specific web page of your website and not to your entire website.

So how it is possible that you produced a title like "Google now encrypts up to 33% of search referral data", a generic statement that even does not apply to your own website?

over 4 years ago

Avatar-blank-50x50

Lawrence

The figure I'm seeing is around the 4% mark for a website that attracts 30% of traffic from the US.

So not a concern at the moment, but a worry.

@Enrico, the title refers to a specific Econsultancy product - a version of the site tailored for the US audience. In that context (which I'm assuming is a very important one from a business persepctive) the title is correct.

over 4 years ago

Sushant Ajmani

Sushant Ajmani, Managing Partner at 3TouchPoints, LLC

This is just ridiculous. What kind of show Google is putting up here? It's completely insane and they are taking advantage of their monopolistic situation in the digital world.

over 4 years ago

Avatar-blank-50x50

Chris

seems very illogical for google to be doing this.

over 4 years ago

Avatar-blank-50x50

Jalal

How can Google claim that signed in users' privacy is the reason for taking such decision when they've just created a social network platform that expose people's documented life to anyone online (Google+) We all know Google+ has not even given us the option of approving the people that circle us!??

over 4 years ago

Avatar-blank-50x50

Martiena van der Meer, Digital Marketing Manager at St John Ambulance

The google corporate moto is "dont be evil"
well you could of fooled me!

Obviously percentages will vary depending on specific sites and target audience but it's bound to get worse and ultimatley its a bad move from Google and the back lash is justified.

Their comments relating to protecting privacy is all smoke and mirrors seeing as i've seen them in the papers breaking privacy grounds more times than i have hot dinners.

There is a hidden agenda here, and we will wait and see if its in their best interest or ours. I know where my money is going.

over 4 years ago

Avatar-blank-50x50

Daivd

I know this is radical thinking but maybe some people don't want us to know how they got to the site. :-0

As an industry, I think we are going to have to face the cold, hard fact that privacy IS going to become more of a front-line issue and we are going to have to adapt accordingly.

Imagine a world where the majority of search activity is via SSL and every site must get explicit permission to track usage. It may not be that far away. What do we do then?

over 4 years ago

Tom Stuart

Tom Stuart, Chief Architect at Econsultancy Enterprise Guest Access

A few clarifications for the above discussion:

* Serving pages over SSL is hugely important for protecting the privacy and security of users. For example, non-SSL sites are vulnerable to trivial session sidejacking attacks: when you log into a site on a cafe's free wifi, someone else in the cafe can steal your cookie and log into the site as you. It is easy to not care about this sort of thing until you get bitten by it; it is absolutely responsible behaviour for companies like Google to worry about this so you don't have to. On the scale of the internet, this kind of concern is several orders of magnitude more important than your company's SEO strategy.

* Nobody has ever been entitled to search referral information; its erstwhile availability is largely an accident of history and technical circumstance, not a fundamental right. As an earlier comment has said: maybe it's none of your business what someone typed into another web site before they came to yours. You are also not entitled to know their age, gender, sexual preference and exact geographical location, but of course those pieces of information weren't available in the first place.

* It's the user's browser, not Google, which sends the search referral data to your site. Google have not "encrypted" or "stopped sending" this information.

* As has already been explained in this comment thread, the HTTP specification mandates that browsers must not send referrer information from a secure site to an insecure one; if a site doesn't use SSL, your browser decides correctly to not send referrer information to that site when you follow a link from an SSL site like Google. Again, this behaviour is correct and necessary to protect against a variety of security and privacy threats. Google didn't make this rule, nor are they the ones enforcing it.

* If you want to get the referrer information, serve your site's pages over SSL so that browsers can send you that information securely. This is better for your users anyway.

* Google's unofficial motto is "don't be evil", not "do no evil", which is an important difference. (See http://lists.canonical.org/pipermail/kragen-tol/2011-August/000936.html for more.) In any case, it's misguided to characterise a sensible technical decision as "evil" just because it inconveniences you.

over 4 years ago

Avatar-blank-50x50

GrumpySEO

I'm absolutely staggered that most people think Google's moneygrabbing is anything new. I just read a recent comment from Matt Cutts on this referral encryption where he says Google is doing this because "its what our users want" LMFAO !!!

over 4 years ago

Avatar-blank-50x50

Everfluxx

@Tom Stuart: Reading what you wrote in your comment above regarding serving pages over SSL in order to "get referrer information", I wonder whether any of you guys actually took some time to test https://www.google.com and see how it works (VS. https://encrypted.google.com).

I have to assume none of you did; otherwise, you would know by now that what the SSL version of www does is *strip* the query from the Referer even when the target URI is on the HTTPS protocol (whereas encrypted.google.com preserves it), and you would have not written "Google didn't make this rule", implying what Google did here is just by design in the HTTP protocol --which is *not* the case (details here: http://www.everfluxx.com/google-ssl-referer-test-results/).

Finally, about this post's title, I totally agree with Enrico: quite misleading!

over 4 years ago

Tom Stuart

Tom Stuart, Chief Architect at Econsultancy Enterprise Guest Access

@Everfluxx: Sorry, you're right, I was looking at https://encrypted.google.com/, not https://www.google.com/ — I didn't realise they were different.

So my fifth point is incorrect: you won't get search referral information from https://www.google.com/ even if your site's pages are served over SSL, because the referring URL doesn't contain the search terms. Thanks for the correction.

Arguably this information belongs to the user, not to Google, so as a user I'm glad they're not communicating it to third parties without my permission.

over 4 years ago

Avatar-blank-50x50

Everfluxx

@Tom Stuart: Thank you for your kind acknowledgment. :)

I totally agree with you that the Referer is a piece of information that belongs to the user, but how many users even know what a Referer is? On the other hand, the vast majority of us search marketers have been relying on search engines to keep disclosing this personal information to us forever, which is just... childish, if you think about it. Because we all know that the keyword-centric approach to search marketing is going to blow up in our face sooner or later, don't we?

over 4 years ago

Talha Fazlani

Talha Fazlani, Online Marketing at Language Connect

There are people out there who will think that they are getting traffic from (not provided).

I think that further security measures will be introduced where search data to marketers will slowly diminish.

I hope someone is developing an alternative to finding search referral data.

over 4 years ago

Avatar-blank-50x50

Alex Mearns

It's a problem for a site I work with too, it is however a good reason to implement an onsite search in order to take advantage of local search terms.

Then create a report to show onsite search terms for all those in the "not provided" portion of your traffic.

over 4 years ago

Avatar-blank-50x50

GrumpySEO

That's got me thinking, Alex. Here's one solution to the problem:

1.Detect Google visitor landing on your site
2.Display message thanking them for their enquiry but stating you can't direct them to the right content without knowing what they searched for (explain that Google has kept this little secret to themselves).
3. Offer to give them the page/information they were looking for by having them use the internal search box provided.
4. Serve up the relevant landing page.

Yes I know nobody will do this, but it would be funny :-)

over 4 years ago

Avatar-blank-50x50

Alex Mearns

I actually had the same thought, but again all that would serve to do is annoy any visitors :)

There could be the potential for making the site search box more prominent to all unknown visitors, or even trying out a search sidebar.

I really liked the solution on the econsultancy blog that suggests viewing the landing pages of all unknown visitors. This incredibly easy in Omniture and provides much better insight than just "Keyword Unavailable".

over 4 years ago

Avatar-blank-50x50

John Ball

Don't be evil huh?

Google new motto should rather be:
"Take it like man"

Seriously!

Google has became the most arrogant company on earth and I got the feeling that they might regret rather sooner than later...

J.

over 4 years ago

Liz Gliddon

Liz Gliddon, Client Service Manager at Andep Investment Consultancy

Almost three years after this article was written and around 90% of the keywords used when accessing my website are "(not provided)"- incredibly frustrating.

about 2 years ago

Comment
No-profile-pic
Save or Cancel
Daily_pulse_signup_wide

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Daily Pulse newsletter. Each weekday, you ll receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.