Enter a search term such as “mobile analytics” or browse our content using the filters above.
Check your spelling or try broadening your search.
Sorry about this, there is a problem with our search at the moment.
Please try again later.
As announced yesterday, Google will begin to encrypt searches made by logged in users, which also means that sites will no longer receive referral data from these searchers.
This referral data, which reveals the organics search terms which led visitors to a particular webpage, is vital in assessing the performance of keywords, optimising landing pages and more.
The decision to remove this source of data has been justified by Google on privacy grounds, citing the example of people using wi-fi in public places. Referral data from paid search ads will still be available, whether the user is logged in or not.
There is plenty of anger at this move by Google, perhaps best summed up by the headline of Ian Lurie's post: 'Dear Google: This is War."
I've been asking some of our guest bloggers and search and e-commerce experts what they think of this move...
What impact will this have for SEOs and web owners?
James Gurd, independent e-commerce and marketing consultant
The negative impact is the hit to optimisation. SEOs and web managers rely on the organic keyword data to evaluate which keywords are performing well vs. those that aren't. They then can analyse landing pages and site paths to identify what might be putting off traffic for specific keywords, come up with hypotheses for testing and test different landing page/content solutions to improve engagement and conversion.
While this is done to add value to the website, it is achieved by improving the user experience. By removing the keyword data, you are effectively blind as to what is really happening. For example, if I have a landing page that has 5,000 visits each week but really low engagement and poor conversion, I can currently dig into the keywords and find out which ones are contributing to the poor stats.
With Google's new change, all I would see is one piece of data for generic keywords with no granularity. How can I optimise the website effectively with a reduction in data quality?
Kevin Gibbons, Founder, SEOptimise:
It's difficult to say,but at the moment, assuming we won't be able to access keyword data and there's no way of working around this, it gives greater emphasis towards keyword research and rank tracking from an SEO perspective. As this is likely to be the most accurate of predicting the volume and value of traffic from individual keywords.
There are lots of questions still to be answered, such as how does this affect all analytics providers and tools such as Hitwise and SearchMetrics, which rely heavily on providing market share research at keyword level. So I think there's still a big learning curve ahead before we finally know the implications of what this means to our strategies and the best ways of working around this.
Matthew Curry, E-commerce manager at Lovehoney:
For online retailers, there are two issues:
1) Being able to track conversion rate and page path on an organic keywords basis is important, and I'm not convinced (at least, if the proportion of (not provided) entries increases), that I will be able to accurately do this.
Analysts, including me, expect this percentage to be higher than the single figures suggested by Google, and will therefore cast doubt onto the confidence of some results.
2) Not being able to see the UTM variable for keyword nixes a fair bit of innovative work, through either personalising content based on keyword or phrase (it is somewhat insulting that Google considers this practise to be primarily cloaking, rather than proving a more targeted experience), or in the case of MVT, being able to segment visitors based on inbound keyword in post-test analysis.
Teddie Cowell, Guava:
It's not good from an analytics point of view. The main loss is the referring keyphrase data. They will have a less clear view of how people arrived at their website. Things like bounce rate tracking may also be harder.
Rishi Lakhani, Search Marketing Consultant
It's downright wrong that Google is treating SEO and PPC data as two separate situations. Although SEO isn't technically 'paid for' media, businesses invest a lot in it, especially around user personalisation, and better user experiences.
At the same time, just like paid search, brands need to know what to invest in, and secluding this data isn't right. Google should either apply this broadly, to include paid search, or shouldn't block those referrers, as long as brands have taken care to protect user privacy.
Malcolm Coles, Product Manager, digital at Trinity Mirror:
It has been quoted as being single digit percentages of logged in searchers, but I suspect this will be a higher figure for searches as you might expect logged in users to be more advanced or loyal users.
This figure is only going to go up as Google pushes G+. It's reasonable to assume that as more people sign up for accounts, they'll stay logged in to Google forever, pushing the percentage who are affected higher.
Dan Barker, e-commerce and online marketing consultant:
It damages the usefulness of every web analytics package on the market. It's doubly sad as, after many years of users asking, Google had just released a great update to Google Analytics that improved site owners' ability to understand what was happening in natural search. This doesn't knock that totally on the head, but it makes it less useful.
Google's #1 organisational value is "Focus on the user and all else will follow". But if you're a website owner, this change reduces your ability to focus on the user, as it means you understand less about how users are reaching your website.
Alex Moss, Search Marketing Consultant, Pleer:
This isn't good for SEOs and may even have a detrimental effect on Google later on. If SEOs don't know individual keywords that are performing, we won't know if specific landing pages will actually serve what the user needs.
As an example, if I have a page optimised for five keywords and four of them actually need to lead to a better quality page rather than the one they're landing on, how do we know what those terms are and how to give a better experience to the user?
Combined with the fact that more and more people are joining Google+ and therefore will be more inclined to sign in, this problem will only get worse over time...
Is this just about privacy?
My view is that it's a cynical commercial decision by Google to ensure the only way you can get accurate keyword level data is by ploughing on with paid search. Yes you can get the data from Yahoo, Bing etc but what's the use when Google has over 80% market share in UK and in some countries like Holland, much higher.
For me there is no logic to the privacy claim. I am a logged-in user every time I use Google on my laptop, and the keyword searches I use contain no private data. What difference does it make to me if a web owner knows that today they had one visit from the UK on the keyword "exercise bike". Nothing. It can't identify anything about me because Google, quite rightly and because of legislation, doesn't pass on personal information. So how does moving to SSL protect privacy? It just doesn't stack up.
I think everyone's been caught by surprise a bit by this one. Normally Google's decisions, whether they are popular or not, aim to make things more relevant and easier for users, but I'm struggling to understand the reasoning behind this change.
Obviously Google will often have a need to increase revenues when it comes to algorithm updates too, but they mainly look to improve things for users at the same time, at least in theory. Continuing Google's revenue growth patterns of previous years was always going to be a huge challenge, but if this is the key reason behind removing keyword data from analytics then it's a very bold and risky move in my opinion to say the least and could quite easily backfire.
We're currently running a poll about this on SEOptimise, and based on the results so far, a resounding 82% say that this is a revenue-focused change.
I suspect it's to prevent other businesses monitoring Google's clickstream data, think about the Bing/Google spat earlier in the year. Clickstream data is a very important, often overlooked data source for search engines these days.
It doesn't offer privacy, at least not from marketers. Searchers will open natural and paid results in tabs following a search, so ultimately the search term is still being passed. it "does" protect from software on the computer, or the network, or ISP, from pulling these terms. I imagine Hitwise is not too happy.
The way this has been framed is a little unnerving, Google should at least make clear its vested interest here, and it has made me distrust the word of Google somewhat.
There have been privacy concerns raised around Google's cookie and referrer data. But what should have been a good privacy PR story just looks weird due to their decision to pass the keyword data for PPC not SEO.
I've yet to see a convincing reason for this that focuses on the customer. Paying advertisers want the data, but that's a different matter to the principle of doing something for privacy reasons for searchers.
The pitch is that this is about privacy, but there were lots of ways Google could have protected privacy to the same extent without hobbling web analytics tools.
I think the project probably started out as a pure 'privacy' thing, which is great, but the implementation of this also helps Google in other ways. For example, it makes it a little bit harder for SEOs to understand what's going on in Google. PPC was already more measurable than SEO. This further diminishes the measurability of SEO vs PPC.
Probably the biggest thing is that this hides data from its competitors. Over the last few years, Google has massively diversified outside of search. This change means Google themselves have as much data as they've always had, meanwhile their competitors (and everyone else) lose some of that knowledge.