September 16, 2014

Instant Insights on Google Instant

I shared an (edited) version of the following on the ZAAZ intranet today, and I wanted to repost some of the information here as well.

Ok, so by now (hopefully) you’ve heard of the Google Instant announcement yesterday and may have even played with it a little. If you haven’t, check it out: google.com/instant. I’m not going to go in depth talking about what exactly Google Instant is, as several tech blogs have already done a great job of doing that.

Instead, I want to give some quick bullets on how Google Instant affects those working in SEO or Analytics, what insights we can learn from it, and what our clients can expect.

So how does Google Instant change SEO Strategy?

  • Well, for starters most savvy internet users don’t pull up Google, they search from a toolbar – so it doesn’t affect them. It does affect less savvy users though (and mostly the general populace), and these people could be worth more (more likely to buy / click an ad / whatever).
  • Instant suggestions seem to be based off of the Google suggest feature. Google suggest works off of search frequency and volume. So it will show what the majority are searching for. I haven’t figured out if this is spamable yet but expect several people to try.
  • Instant should have the effect of funneling people toward higher searched terms, so I would expect to see a slight decrease in long tail terms and “unique keywords” sent to websites. Conversely, I expect to see an increase in traffic for the “funneled” terms that I do rank for.
  • There seems to be a lot of localization built in. When I search for “city” I get “city of Detroit.” When I search the letter S my default is “Sears” but the default for our Seattle office is “Seattle.” This should make optimizing for local search even more important.
  • The big part I don’t like is the “auto complete,” for example I can’t see results for “car” without hitting enter and actually searching. Instead it shows me “careerbuilder.” I expect this to improve over time.
  • SEOs who optimized for commonly searched terms (those who did good keyword research) will see the biggest reward. At the very least, one should look at Google suggest and see what’s being suggested– as the goal could be to try and show up for those suggestions to maximize visibility.
  • It doesn’t work for terms that would otherwise be filtered by “safesearch” and it doesn’t work for “sucks” style terms. In fact, we noticed last month that Google removed “sucks” and other negative sentiment terms from Google suggest.

What about AdWords?
You’ve probably noticed the AdWords ads change by the query. According to Google, an AdWords impression is not counted until the user stops typing for 3 seconds. Type in a query and you’ll notice the URL in your browser will change. When the URL changes you’ve “fired” an AdWords impression even if you resume typing after the pause. I expect to see impressions increase across the board – and CTR to fall as a result.

Technical Stuff
If you’re looking at referrers in your site logs, you should see several new parameters coming from Google searches. The biggest new one is the oq= parameter. When a user searches on Google Instant, the q= parameter passes the search term that Google showed results for (the auto suggest one) while the oq= parameter passes what was actually typed into the box. I would love to see webmaster tools incorporate this data so i can see the differences. There are several insights to be had from that data.

Also of note, many rank tracker programs have broken today due to Google changing the underlying HTML structure of the results. I’d wait a few days before running any ranking reports.

What About Clients?
Google instant could be a challenge to explain to clients – especially if they haven’t used it yet. If you use unique keywords as a performance metric though, you may want to spend the time explaining the change to clients. This change isn’t a bad thing, and it should yield great results for SEOs who put the time into doing good keyword research.

So who will it hurt
Hopefully nobody – but if you’ve got a poorly optimized site you could expect to see larger sites take some of your long tail traffic. If your site is positioned to capitalize on mis-spellings or common typing errors then I would expect to see a large decrease in traffic. The same goes for spammy sites who optimize solely for long tail keywords.

What’s your 140 character summary?
Google Instant doesn’t kill SEO. It rewards SEOs who did proper keyword research. It’ll be interesting to keep up with and see what happens.
(and yes, that was exactly 140 characters)

About Ryan Jones

Ryan Jones is an SEO from Detroit. By day he works as a manager of SEO & Analytics at SapientNitro where his team performs SEO for Fortune500 clients. By night he's either playing hockey or attempting to take over the world with his own websites - which he would have already succeeded in doing had it not been for those meddling kids and their dog. The views expressed here have not been paid for and belong only to Ryan, not any of his employers or clients. Follow Ryan on Twitter at: @RyanJones, add him on Google+ or visit his personal website: www.RyanMJones.com

Comments

  1. It can have some unintended consequences too. For example, when you type FAIL, it defaults to “FailBlog”

    As a result I’m seeing significantly less traffic to Failpictures.com because people appear to be selecting that failblog result and not actually finishing typing “fail pictures”

    I expect there to be similar issues for any result that returns a specific website.

  2. I noticed that we also have the ability to turn it off, next to the search bar, if we don’t like it. One thing is for sure – from an SEO perspective, this changes the way we have to think when we target keywords in the articles we write. I see the results from short words preceeding the long tail phrases. I want to grab the persons attention first with the short words… I love it! But, yah, impressions will certainly rise from slow thinkers.