SEO

SEO Under Attack – The Google Analytics Keyword Data Apocalypse

Google announced today they’re going to take a step toward user privacy by hiding keywords used by searchers signed into their Google accounts.  We’ll leave the discussion regarding their real reason for doing so for another day or someone else to blog about. Me, I’m going to talk about what I think this means for the industry and why I think so many people that are freaking out over this change are missing the bigger opportunity here.

Okay so – people are crying foul on this change to Google Analytics.  They’re saying they need keyword data in GA, because they tie those keywords to conversions – the more conversions that come from a specific keyword, the more important that phrase is.

If that keyword is getting conversions, but it’s ranked on the 2nd page of Google, the thinking is – hey – do more SEO for this phrase, get it higher in the SERP, and in turn, naturally, we should see exponential increase in conversions.

That may or may not be true.

First of all, on any given day, there could be a thousand reasons someone comes to your site after seeing a specific result in the SERPs.  Maybe that specific phrase is the golden egg, and maybe it’s not.  Maybe it only appears to be. Maybe there’s a better phrase you aren’t even considering, that could get even better results.  Maybe there isn’t.  It’s a real possibility though, and you may not even be aware that it is due to being fixated on what’s there in front of you.

And Google isn’t eliminating ALL keyword data.  Only a sub-set.  Okay – so if you have a particular site where you know for a fact that the majority of visits come from people signed in to Google, maybe in that particular scenario, you may truly be negatively impacted even from the loss of ability to look at general keyword trends.  But how many of you know that’s your particular situation?

The Bigger Problem In Our Industry

What I have found over the years I’ve performed audits on sites that others have implemented SEO for is that every single site, without exception, has had massive or at the very least, significant problems in its SEO, regardless of keyword choices, that shows me, over and over again, that most in our industry get hung up on granular issues and miss the bigger opportunities.

If a site has one or more critical big-picture SEO issues, all the granular analysis in the world is nothing more than a distraction to me.

SO what if I can move the conversion needle 5%, 15% or even 20% over time, if in reality, when proper forensic SEO is performed and the resulting changes cause a site to see 20%, 40% or even 60% increases in that same timeframe?

That’s the issue to me, in a nutshell.  If a site has structural flaws, and if you fix those structural flaws, not only will you automatically get higher rankings for whatever phrases you optimize for, you’ll get higher conversions.  You’ll get an exponential growth in the long tail without having to focus on the long tail.  You’ll get more visits and more conversions from other non-Google sources as well.

And honestly, either I can analyze keyword relevance without seeing it in GA, and tying it to specific conversions, or I can’t. If I can’t then maybe its’ because I’m not looking at other ways to get that data.

No, to me, once again, this latest apocalyptic cry is no different than any other previous panic attack due to people being hung up on shiny objects, magic bullets, and myopic SEO.  As painful as it may be to hear me say it, my best recommendation is to pause and consider whether you’ve been missing more important issues in your work than you realized.  And because most site owners don’t have an unlimited budget, it’s just my opinion (controversial as it may be), that people in this industry need to wake up and recognize that if they truly want to maximize their client’s or employers SEO money, they need to stop and learn that they have bigger fish to fry that they weren’t even aware of.

 

12bcd73262dd3dcb8597e6d4f9884119 64 SEO Under Attack   The Google Analytics Keyword Data Apocalypse
Alan Bleiweiss is a Forensic SEO audit consultant with audit client sites consisting of upwards of 50 million pages and tens of millions of visitors a month. A noted industry speaker, author and blogger, his posts are quite often as much controversial as they are thought provoking.
12bcd73262dd3dcb8597e6d4f9884119 64 SEO Under Attack   The Google Analytics Keyword Data Apocalypse

You Might Also Like

Comments are closed.

40 thoughts on “SEO Under Attack – The Google Analytics Keyword Data Apocalypse

  1. Alan, I respectfully disagree here.

    This change doesn’t just block GA’s access, but all third party tools. Because all logged in traffic off a SERP will be encrypted. So how are you going to calculate conversions then?

    Secondly, if you really want to see the big picture you need to see whats ranking. I don’t understand how anyone can see the big picture otherwise.

    1. In fact Alan, I believe that this change will result in more web masters using myopic SEO as they are forced to focus on a handful of terms versus having a clear understanding of the long tail.

  2. I don’t care about them taking away the data so much. I care about them taking away the data from everyone but their paying advertisers and doing so on false claims of privacy concerns. If they cared they’d take it away from everybody, or they’d apply the same fix they made for AdWords to organic search results.

    1. Spot on Joost!

      This is just Google taking control over more data, and using it as a competitive edge to gain more revenue. Nothing to do with privacy concerns. Come on! Google is the only one who can tie that search query back to an actual name, birthdate, and mobile phone number! The website owner would just see it as an anonymous query from within a city. This smells from the start. It will only get worse.

    2. I wonder what logic they used to say regular users deserve privacy, but those who click on ads don’t deserve any.”…

      What logic was employed that decided that advertisers and anybody with the tools, no matter what their motive, should have access to this information? Are you saying that you didn’t know that advertisers were using this data, or is it that it’s all or nothing…either Google shares the data they have spent so much time and money collecting with everybody and blow off attempts at privacy, or block everybody?? I mean, if you all can make a good argument why they should cut into their revenue stream simply because it inconveniences yours, I’ll agree…otherwise I think there is still room for you at one of the “Occupy” protests…

  3. Hi Alan,

    I agree with a lot of what you are saying, but this is a huge deal to smaller sites that don’t have large structural flaws. Not all SEOs work on old sites with poor architecture and 250k pages. Google’s organic keyword data gives SEOs a great resource for on-page optimization, new content creation link building opportunities and more.

    Is this change the end of the world for SEOs? Of course not. Is the outrage on Twitter a little much? Possibly. SEOs will adapt and make due with the data and tools we have. The real problem comes from the spin of Google’s “privacy” slant. Making this data available for paid searches buy not organic searches is hypocrisy at its finest.

  4. While I don’t think its the apocalypse of SEO, and there are always issues to work on, ignoring information that is often actionable by other teams or resources is foolish. I also agree with Joost that leaving the data in for paid advertisers while claiming privacy issues is a bit much even for a PR powerhouse like Matt Cutts to spin well.

    Is this data going to be available in the premium GA product and they are positioning things to make that price tag more attractive?

    Your position seems to be that don’t worry if this info isn’t available to you there are other things to work on, and I just don’t agree with that type of thinking. Other than that, nice rant as always :)

  5. Alan,
    I agree the impact likely will be small (we’ll know soon enough). Like Joost says though it seems poorly thought out, and not a true commitment to privacy. Though how then do they provide accountability for clicks people pay for on the paid searches? I didn’t really know there was a big outcry from users for this feature either.

    From an analytics stand point I think the impact will be small as any data people act on should be driven based on actionable volumes anyways.

    As for the overall article I think there’s 2 tings here, general site wide SEO, fixing the big issues, but there is also very targeted SEO where you need to identify if people are converting on your site, and what pages they arrive on from specific keywords. Search keywords provide intent. I can get a better frame work of my users intent from search if I know the sets of words. I can then ensure we have pages that focus on conversions and converting users. Otherwise I’m guessing at this and it’s no different then of optimizing for display ads, or other sources that don’t provide specific intent.

  6. Sorry Alan, but what you are saying misses the big opportunities right out. If making a small myopic change for a single keyword will significantly improve sales for a client, and has a better ROI than a big site wide high investment change, I will recommend that make that myopic change as a higher priority than anything else.

    My clients prioritize getting the biggest return for the least investment, and if that means making small profitable changes over larger ‘big picture’ changes, then that’s what my clients deserve. It’s not about winning the award for ‘best SEO’d site’ it’s about making money at the end of the day

    1. I agree with Sarah here. While structural changes may be the big problem plaguing most websites, it is also one of the toughest recommendations to implement. The lowest hanging fruit may be a handful of keywords that rank decently, but convert like gangbusters and would drive sales if ranked higher. If that has a higher ROI than structural changes, why wouldn’t you address those first?

      The reason for the outcry on Twitter is because Google is taking something away from webmasters. Something webmasters used to make better business decisions. If Google had never provided keyword referral data in the first place than it wouldn’t be an issue. But they did, and now it’s a punch in the gut for those who use keyword level data to build stronger strategies.

  7. Devil’s advocate: let’s say 3 months from now, a site gets a 200% lift in visits and revenue from organic search, and that traffic consisted of 100% branded traffic from a TV campaign blitz. The SEO campaign gets all the credit because, conveniently, all that traffic was scrubbed and listed under “query not available.”

    Obviously not a win for the SEO campaign, but all signs point to the contrary. Granted, this scrubbed traffic will only make up a subset of traffic so it would be possible to make correlations, but correlations != causation, and that’s a bad thing for data/performance-driven decision making.

    On the plus side, we can focus more on ROI-based, higher-level SEO as CMOs are forced to not pay attention to specific keyword rankings and traffic growth, but less data is never a good thing in my book.

  8. Alan,

    I foresee the issue being reporting to clients. It is valuable to maintain a client when you can show an ROI for them. Some clients tend to have narrow focus and some have a hard time attributing any success to the seo campaign that they are paying for. Without an ability to point them to their analytics data that shows the amount of revenue different keywords are bringing in and the increase in that revenue from the time the campaign started, it may be harder for seos to prove their worth.

    I do agree that it is too easy to become focused on smaller issues when larger ones exist.

    Just my 2 cents :)

    1. I totally agree. If a client sells for instance “blue buttons” and decides to invest in organic SEO – Let’s say I identify “blue buttons” as a valuable organic search term.

      How can I prove it was my SEO work for that term that sold them more blue buttons?

  9. Alan,
    Thanks for the timely and informative article. I am not completely sure how hiding keywords will help with privacy concerns, but fine.
    What is more important is the point you make – people are not focusing. Obviously, there is still work to be done on a website after the change. Usability, conversion optimization, on-page SEO still can be improved with much more impact than tinkering with one phrase. I see it every day – people not looking at the bigger picture.

  10. wow. a lot of great comments, so thanks to everyone for being willing to at least discuss different view points. Okay so I can see where some people would be harmed by this – to the degree the lack of this specific source data skews the overall accuracy when it comes to evaluating whether a particular phrase is really generating conversion value.

    And I can see where for much smaller companies, the impact would be greater when conversion data on individual keywords can otherwise help focus the very limited resources available.

    I still stand behind my position though overall. If a company only has a very small budget, I prefer to put more money in getting more work done based on the most obvious areas – more content, more social, more links, around phrases that don’t take a rocket scientist or analytics whore (waves to Annie Cushing) time to figure out. But that’s just a personal preference. Because either path will get positive results when done properly. It’s not an either -or thing. honest.

    The REAL issue is, if this is so harmful, what is everyone who believes it is, going to do? give up? tell clients “we can’t do our work now”? No. You’re going to figure out other ways to get results for your clients. And if you don’t, don’t be shocked beyond recovery down the road when all that data is gone. Not just for “signed in” users.

  11. Let say if 20% of keywords traffic came from signed-in users and it marked as “Query not available” but it contributed to 40-60% of conversions, this will absolutely impact how to analyse the data, prioritize the major converting keywords and optimize it. Matt Cutts said that it is only single digit for signed-in users , but i think it would be more significant especially growth from Google+ and Mobile. People always stays on the signed-in session and conduct search when they are using Google+.

  12. Hi Alan,

    “I Must Say, I Have Been a Fan of Yours For Years!”

    Thank you!’- for the ongoing gems of information
    that make sense for a vast number of fortunate
    web creators, new and seasoned) Many people
    were able to move ahead beyond their learning
    curves, and continue to do so successfully today.

    Best wishes,

    D:-)

  13. Hi Alan, i agree with Yoast regarding the privacy issue as this just doesnt cut the mustard in my opinion however i agree that site structure/structure is a big area where fixing issue can bring big gains so yes I agree to an extend that some seos may miss the bigger picture. Should google announce in the future that the GA premium product will provide this missing data I think smaller sites will suffer/those with not big bugets to invest in this tool and bigger sites/busnesses will invest in GA premium This data is important to have and Google have for a long time given away a lot for free so it is no big surprise that they are going down this road.

  14. I am curious to see how/if this change manifests itself in the Google keyword research tool (and other keyword research tools for that matter).

    If the data that is hidden in Analytics is omitted from the research tool’s results then it could create a scenario where significant misjudgements follow, particularly if anyone assumes that the percentage of logged in users searching will be even across all search terms.

    1. Iain It shouldn’t impact the keyword tool at all. First, the keyword tool is part of the AdWords system, and since that’s not being affected, the tool should be fine. Second, and more important however, is the reality that relying on any data as if it is factual across the board, let alone truly accurate, has always been a misguided belief practiced in relation to SEO.

      Since NO data tracking solution is perfect (Google Analytics is NOT an exception here – it’s always been flawed and imperfect just as all others are), data should always and only be used for trend analysis. And only when held up against other supporting data.

      So when the day comes that ALL referring traffic goes away, or even if the Google Keyword Tool goes away, people will need to wake up and find alternate methods for making decisions. Or they can prepare for that day now, and not be caught with their pants down as many have been this week.

  15. lol..Google: the mastah of establishing fare-of-need, give em what they need to build Google up, then flip the switch.

  16. What ways are there to get the data without GA? Another analytics tool like Omniture? I’m truly open to this discussion. Or without the keywords what would you look at differently instead? Because you know that one day logged in Google users will be the majority not the minority.

    I’m lost when it comes to knowing what customers are asking for without the keywords they clicked on. I’ll still see what they are buying but how will I match the pages to the right keywords. Perhaps Google really wants to be in complete control and squash us SEM people because they know best what should rank without any of our help.

    1. What pages get conversions? those pages have been optimized for various phrases. Look at the context of those to other similar pages. There’s patterns that can be found in that data. And you control that data. Those are the pages to consider getting more optimization for, driving more traffic to. Pages that have a lot of visits but lower conversions can be considered for either overhauling or redirecting to other pages.

      I’ve been using this technique for two years. And the results are stunning. It takes less time and effort than getting caught up on long tail conversions, and gets better results on every site I’ve used it on. The work inevitably leads to sites being found for even more long tail (without spending time focused on the long tail!) as well.

  17. So, this discussion leads me to this question – Does anyone have a reference on some sort of statistic on how many people surf through Google logged into their account? In my experience, most people open their computer, log into their email to check them (gmail mostly) and then they would do a google search, which would mean they’re logged in. However, I realize – based on the people that I am around all of the time – that what is the “norm” for me and those around me is not necessarily the “norm” I am curious to know what % of users are logged in when they do searches.

  18. This will have an impact on smaller sites with less traffic to analyze. On bigger sites with 100k+ visits per month, not so much. But those bigger sites are the minority. So I would say that this has a huge impact and that many people are not realizing this and are trivializing the change.

    I agree that most people have things that they SHOULD be working on in the site and that most people tend to not do these things and rather focus on other distractions. But that really has nothing to do with having a major data source taken away from you. For those of us that are tied to data in order to make a real business case in order to get things done, this sucks. Maybe, Alan, you are lucky enough to be able to go off of your personal brand and reputation and are able to hand select who you work with and you can just “tell” people that what they should be doing SEO wise? But the rest of the world, us peons in the trenches, we don’t get that luxury. I spend 60% of my time building my cases with data. I can’t just “tell” my clients what I think they should be doing. Tried that, doesn’t work most of the time. I need keyword level data to prove to them that the long tail is better than the 1-3 vanity terms that they are in love with.

  19. Privacy? My baloney detecting tool is going off the charts. Its one thing to disregard your customers, and its another to insult their intelligence.

    1. You are being silly Edward. It doesn’t matter what analytics program you use. No getting that data. We all know GA is free. The point I made above is so what are we going to do to do things like brand and non brand comparison searches when the day hits 50% or more logged in searches. It’s not like Google auto logs you off or their % of users are going down.

  20. Great article mate …….but how i am going to know by which keyword visitors came to my site if GA encrypt it all
    It would make SEO more difficult then

  21. Good point but your whole premise is based on there being an existing bigger problem with a website. Whether this is true or not does not address the fact that a certain portion of data will no longer be available from GA. You mention “. . . If I can’t then maybe its’ because I’m not looking at other ways to get that data.”

    Seems like those “other ways” might have been very much on topic.

    Thanks for the article.

  22. Google is forcing whenever possible everyone to use a single Google Account so they can compile information across sites. They are also begging for telephone numbers as a means to tie accounts together. We use a lot of different Google properties and for our own reasons do not want to tie them together.

    In my view Google Analytics has run their course and feeding the machine may actually come back to bite you. There is nothing to prevent Google from useing the Data they feed you against you. We’ve pulled Google Analytics and are now using something else. Sometimes free isn’t free.

    I totally agree with Joost de Valk this is about something else entirely. Google is making it harder and harder to use any of their services without paying. But that is afterall their busines model… scrape content from others and figure out ways to monitize the content they scrape. What exactly is Google without other people’s content?

  23. I agree this isn’t a cause for panic, but if in the long-term Google succeed in getting more and more users using an account it would be a great pity to lose the majority of keyword search data. Not something that is going to happen any time soon, but it may do potentially in the future.

    And truthfully, even if we lost it all you’re right about this not being a disaster. Even if it was I’m not so worried about what phrases are ‘most important’, and more about being able to see which phrases lead to a higher conversion rate and which perhaps lead to more bounces (indicating a problem that can then be solved). That kind of refinement and trouble-shooting can be very helpful, so I’d certainly not want to lose that.

  24. Thank you very much for another very informative post.

    I’m not sure I really trust anything that comes from google anyway. Their keyword tool for example does often seem to be inaccurate and unreliable.

    They seem more focused on bleeding as much money from advertisers while giving very little real info to help people properly optimize for best results to cost.

  25. I agree this isn’t a cause for panic, but if in the long-term Google succeed in getting more and more users using an account it would be a great pity to lose the majority of keyword search data… I agree this isn’t a cause for panic, but if in the long-term Google succeed in getting more and more users using an account it would be a great pity to lose the majority of keyword search data. Not something that is going to happen any time soon, but it may do potentially in the future…. Not something that is going to happen any time soon, but it may do potentially in the future.

  26. All of this re-arrange wont mererly block GA’s connection, but every 3rd party solutions. Due to the fact that all logged web traffic off a SERP will certainly be encoded. For that reason exactly how you are going calculate conversions?

    1. Hi Peter

      Visitor referer sources won’t be blocked, only the keywords they used. So you’ll still be able to track conversions, just not associate them with the exact keywords they used. It’s not ideal, yet you can still know they came through Google organic, the page they came in on, and the topical focus (primary phrase(s) that page is already optimized for.