Ugh... Part of me just wants to link to this old blog post and leave it at that.
But, since there's actually a bit of data to share helping to show that (at least so far) Google Instant changes less than your average algorithmic rankings update, let's share.
880,000 Search Visits Analyzed
Conductor released some nice research from anonymized data of sites on their software platform making a compelling case:
If Conductor keeps putting out this kind of stuff, they'll be a "must-read" in no time
Hmm... Looks pretty darn similar to me. A tiny increase in 4, 5 and 6 word phrases would seem to go against many of the prognostications and fears that this move would decimate the long tail (though, to be fair, plenty of savvier search folks predicted a slight increase as Google's "Suggest" function would be more obvious/visible to searchers and push them to perform more specific queries).
Data from 10,000 More Searches
Matt Bennett at MEC blogged some data from 5 sites in his purvue representing about 10,000 searches. He shared this excellent graph (similar to the one above):
That's more evidence to suggest this is a very subtle change (if there's any at all) in keyword demand.
Google Search Traffic for SEOmoz & Open Site Explorer
While I don't have as much data to share as Conductor, I can show you some tidbits from SEOmoz.
Here's SEOmoz.org's traffic from Google in the past week compared to the week prior:
And here's a similar look at OpenSiteExplorer's Google traffic:
There's a suspiciously small amount of change in the keyword demand, and although these are certainly un-representative of the broader web, we can be relatively confident that lots and lots of folks in our industry, performing queries that might lead them to these two sites, have awareness of and are using Google Instant.
One change that did catch my eye (thanks to some Tweets on the topic) is that Google's Suggest itself seems to have changed a bit:
Hard to complain about that :-)
Other Sources Worth Reading on the Topic
I was a bit dismayed to see so many in the SEO field taking this as a serious threat or even touting the massive "changes" that would be coming soon to SEO best practices or even search query demand. We're usually pretty good about shrugging off Google's pressbait around technical changes that don't have much of an impact, but this one seemed to have more legs than usual.
That said, there are a few pieces I think warrant a read-through (or at least, knowledge of):
- Google Instant was Actually a Counter to Bing's Advertising Blitz from SearchEngineLand
- Google Undertand Not Google Instant from Michael Gray
- 3,000+ Mashable Voters Prefer Google Instant (Slightly) from Mashable
- Sad to link to this PCMag Retraction Showing Their Tragic Ignorance of Search - they confused SEO and PPC, thought marketers might need to buy ads on partial words ("flow" for "flowers"), etc.
- Explanation of Ad/Query Triggering (basically it's 3 seconds of inactivity) from AdWords
- The Complete Google Instant Users Guide from SELand
- A Great Image from SEOBook showing how Instant further marginalizes the organic results
- A downloadable spreadsheet from the folks at Seer Interactive that can help you calculate your own data on keyword demand changes
Very much looking forward to the discussion, but I'm leaving for Social Media Week Milan and will be hard pressed to contribute at normal levels until my return next week. Until then - Buona notte!
p.s. If you have data to share on how Instant has or hasn't impacted your traffic-driving queries, that would be awesome. If you blog/upload it, we'll be happy to update the post with links.
Here's your real world feedback, folks.
I have a quite a few online stores that were doing very well for what they are. Not-so-sought items ranked in the top slots, but almost all of the ones that sell commonly purchased goods fall below large corporate sites.
My sales since Google instant reared it's ugly head went from about 20 to 50 orders a day during September to less than two orders a day. September used to be one of my best sales months. What little trickles in now is from other browsers and Bing/Yahoo.
I don't think Google Instant is wonderful, in fact I loathe it. It has erased ten years of my working adult life and with one stroke taken away the income my family and I have survived on. I don't have words to express how angry and outraged I am right now.
They should not even bother with showing results other than #1 and #2, as that is what people are actually clicking on and nothing else.
Hey Genius, I'm awfully sorry to hear about your loss of income. That really stinks. Are you certain it was the Instant product and not the Caffeine update that is responsible?
I'm assuming you are still getting at least a few orders from Google, so have you looked at your analytics to see what keywords they are now coming in on vs. the keywords they used to come in on?
Finally, assuming it is Instant, have you done competitive research using the new Instant results to try to analyze what the sites that are now receiving your old traffic have to give them the lift?
All is not lost friend, just start digging into it and reformulate your sites. I'm not implying that it's a piece of cake to do this, I'm just trying to encourage you that there is hope for you to regain some of what you lost.
I couldn't agree more with this post. If anything, Google Instant hands good keywords over to SEOs on a golden platter. My only concern is that no new keywords/phrases will develop over time because people will rely on the suggestions. I would be interested in how Google plans on handling that.
If anything changed SEO, it is localized search. I live in Fort Wayne, IN - but a trace of my IP puts me in Chicago, IL (because of my ISP). So when I go to Google and search for "web design" I receive a bunch of companies based in Chicago. Personally, I think Google should make this a feature you must be logged in and opted in to experience. Google should also request a user's actual location (for their Google profile) so that it can use that location for the search instead of inaccurate IP geolocation result.
Just my 2 cents.
Chris
The problem is that, to my knowledge, we do not have any data on the pervasiveness of Google Instant.
I have not yet had a single computer or browser in my possession have Google Instant turned on. This includes 5 browsers on my primary computer (IE, Opera, FF, Safari, Chrome), 4 browsers on my home computer (IE, FF, Opera, Chrome), 2 browsers on my laptop (Opera, FF), Safari on Ipad, my wife's netbook (IE, FF), my wife's laptop (FF, IE)...
Until we get some penetration data from Google about how many folks are seeing Google Instant, it is going to be hard to tease out the impacts.
--- Note: just started getting Google Instant on this computer moments ago! Finally.
Interesting, but I still have some questions about what Google Instant will mean for the long tail. A couple of thoughts -
It will probably take some time for a good chunk of users to adapt to searching with Google Instant. While it's definitely an in-your-face change that's hard to ignore, a lot of people are weary of changes like this and are probably still fully entering their originally intended queries. These people will come to rely on the suggested results over time, but I'm thinking that there is a lag here that is obscuring how instant results will ultimately affect user behavior.
Also, while Conductor's data shows that query length hasn't changed much, I don't think that this says much about long tail consolidation. Sure, Google Instant is serving a good mix of long tail queries and head terms, which is why this data shows no meaningful change. But how deep is the set of long tail queries they are offering up? It's probably a lot shorter than the natural long tail. That's long since been how Google has managed the AdWords keyword tool - for whatever reason, results are restricted to a small set of long tail queries.
Remember that statistic from a few years ago that 25% of the queries Google sees on a daily basis were never seen before? I'd be curious to see what that percentage is a year or two from now. Unfortunately, that's something that only Google can really measure.
All in all, it's too early to say much... but I agree with you on one thing, this isn't an SEO killer. Actually, I'm willing to go out on a limb and say that it's rife with opportunity for smart SEOs.
"Remember that statistic from a few years ago that 25% of the queries Google sees on a daily basis were never seen before?"
Yea, and how many were mis spellings. That's something most don't think of.
Its too early to judge the magnitude of the impact on SEO from Google Instant. The bigger effects will be felt over time as users change their search behavior. My hypothesis is that since users see predictions as they type, they will tend to refine their searches more often, looking only at the top results, and reducing how deep into the results list they look. If this does occur, then the importance of being in the first page of results becomes even greater. And for those sites in competitive sectors, focusing more on being at the top of a few longer tail query results should be the focus instead of being second page for more queries. But since this hypothesis is predicated on user behavioral changes, I don't see any reason to change SEO tactics in the near term.
Rand,
Your post made me think of the guy last week who accused SEOMoz of over-hyping things to gain attention for yourselves. What we've seen in the past week is a clear example of many in the community screaming "fire" in the theater with respect to Google Instant to draw some attention to themselves. To SEOMoz's credit, thank you for providing the data and evidence to say, "Move along folks, not much to see here...more important things to focus on."
I also thought about this - but in this case it is the SEOmoz anti-hype machine. There could be an equally good infographic inserted here. Maybe I will go work on that!
Honestly, I believe it is too early to start any comparison. Although the feature is already in place, it will take a while for the average user to take advantage of it. What it is probably happening right now is that the user is typing the entire keyword (either long tail or short tail) in spite of the Google Instant suggestions and instant results.
However, it is a great benchmark for future comparisons. I would love to see the same results let's say 6 months for now.
My thoughts exactly - Too early. Instant is not rolled out fully yet. Many people use Toolbars or Firefox start pages. And its far too new for people to have got used to it yet. The graphs above comparing the week before/after are I think just too soon. It needs to bed in. What would be more useful is to make the same type of comparison for the week before with say 1 or 2 months after, if Instant is fully rolled out by then. However, I do feel that Instant will help to educate people on the use of long tail, and what other people search for, so that in time I believe long tail will become more relevant.
Look, it's all going to depend on your site. A site like SEOMoz or a site like Amazon probably won't see a lot of change as a result of Google instant. But I have some sites, let's call them directory sites, where you can be relevant to thousands of localized searches. By localized searches, I mean a search with a city and state in the query. And that's where Google Instant really matters.
Type in the word dentist, for example. Google instant will show results for 'dentist in x city y state.'
On these directory sites, I have seen total unique keywords double since Google Instant launched. I don't mean to say that overall search traffic has doubled -- in fact, it's hardly moved. But the number of keywords people have used to find me has gone up by really impressive amounts.
Whilst hardly analytically valid, out of interest I've been tracking the suggestions presented for 10 different keyphrases, ranging from 1 to 4 words, for the past 8 days (i.e. post Instant). All of the 50 (10 x 5) suggestions presented have remained 100% consistent. Obviously they'll change over time according to collective user behaviour, but in the short term they appear stable.
As a result, we are currently debating whether to go through all of our head terms, identifying the suggestions presented for each, with a view to then optimising for the terms we are not currently targeting (the suggestions are throwing up a few fairly off-the-wall terms that we've previously missed).*
One factor that suggests this might be premature (at least outside of the US) is that the suggestions appear to be based upon US search traffic. For example, a search via Google.co.uk for the term 'gift ideas' displays 'gift ideas for mom' as one of the 5 suggestions ('mom' is not common UK parlence, we use 'mum'). Presumably, as Instant rolls out, the suggestions will become more localised.
For what it's worth, figures for the site I manage are...
01-07 September - 41,987 non-paid visits from 21,208 KWs
08-14 September - 45,747 non-paid visits from 21,781 KWs
15-21 September - 50,131 non-paid visits from 22,868 KWs
...but then this could be seasonality, or the fact I've been busting my behind on content development & link building :-)
*This could be an opportunity for a savvy developer to produce an online tool to complete the task. I imagine you'd enter your head terms (perhaps pasted in bulk from the spreadsheet you use to track them), you click go, and it reports back on the suggestions presented for each/all terms combined). I'd be delighted to test it out!
Rand, do I need to be concerned about the fact 3 words search terms brings the most visits to my website, while 4 words phrase gives the best conversion rate, according to OneUpWeb? Any suggestion on how to fill the gap?
Thanks for your time! Great article! Maybe I can turn this into actionable data! :D
SuperX,
I can only offer lightweight reasoning here, but if you convert best on 3 keywords then that's fairly targeted and good work on your part. In other words, if your topic is best served through 3 keywords it's not necessarily a big deal for you if more people seem to be clicking on 4 keywords that Suggest/Instant offers.
Watch your numbers for any changes. More importantly test out queries removing personalization as well as attempting to target various data centers using the the query params.
Perhaps these posts will help with data center targeting (keep in mind that I don't know about using proxies to target specific data centers):
https://www.seomoz.org/ugc/how-to-determine-your-true-organic-google-ranking
https://www.seomoz.org/blog/indexation-for-seo-real-numbers-in-5-easy-steps
Try out your prime 3 keyword queries and see what Instant offers at 4 keywords and beyond. Are you still there? Is it your competitors? Does 4+ keywords go in a completely random direction?
Personally, I think you're fine and some testing should give insight into whether or not you may want you want to test out optimizations on an extra keyword or two. Being the trouble maker that I am, I'll also suggest that you test out the LDA tool in labs after your testing is underway. I believe the LDA tool focuses on single keywords, but there no reason not to play.
Well... Actually I wasn't talking about my numbers, but about the fact Conductor's research on CTR was slightly different than the one OneUpWeb made a while ago, which was about conversion rate.
However, as your in-depth analysis suggest, results are always unique, depending of your audience and of what you're selling.
Thanks for your reply.
Why are people attributing the number of words in a search term to be the "long tail"
Long tail is the likelyhood of something occuring.
Whilst there is some correlation between shorter keywords and how likely someone will use that as a search query, I don't think it is possible to attribute a change in the dynamics in the simplistic way it is being currently portrayed.
Queries will change over time, possibly to more of the suggested terms, but those queries can be both longer and shorter.
What you are most likely to see disappearing first are junk queries and misspellings.
Spot on Andy. From Google's perspective it does more of their work for them. Crap sites optimised for mis-spells will suffer, relevancy of Google's results should improve.
Sure! Six of one and half dozen of the other please! :)
We already had auto-suggest to get rid of these misspells optimized websites... Instant doesn't reinvent the wheel...!
The only thing it does is perform the search for us each time we type a letter. What are true benefits of this? Because, imho, it does nothin useful, except eating bandwith with a ton of JSON queries, which cause servers to take more power, which pollute atmosphere!
Just kidding, but really, can somepoint point me pros? Please? I don't understand what this fuss is all about!
I'm with you Super X. I haven't used it extensively since my everyday browser is Opera, but I've dabbled with it on Firefox.
I don't get how working with less characters as I type would be more accurate than waiting for my full query. As a result, if I did use a browser with GI enabled, I'd be ignoring the changing results until I finished typing.
Maybe I'm missing something but it doesn't seem to be an easier search path if I have to keep looking back and forth between my typing and the ever changing results. It's certainly an interesting effect, but I've turned if off in Firefox as it's more annoying than anything else.
There seems to be a bias towards popular brands and I wonder how this will affect less popular brands.
I haven't seen the impression spike and CTR drop predicted for PPC accounts, either. I wonder, though, what the current adoption rate is. To my knowledge, Instant is still only available directly from Google.com for people with a Google account. Granted, most of us here have Google accounts and are logged in most of the time, but I'm betting that percentage is much, much smaller for the general populace.
You're totally right... and infact that was what was going me crazy while reading all the post claiming SEO is dead...
A clear sign that before writing so called SEO experts should just try to not think they use the web as the majority of the people do. And it's ironic, as one the first thing we learn is: know how people use the Search Engines...
for me seo... suggestions are "seoul", "seoul weather", "seohyun" and then "seomoz". I'm in Canada
If Google say you want to know about Korean weather, that's WHAT YOU WANT. Stop making trouble ;)
One of my clients is in the "seminar" business and if I type in "seminar...", the first option is "seminary coop". Yeah, I'm sure more people want to know about seminary coops than seminars. My father and brother went to seminary, and I don't even know what that phrase means.
Actually Eric Schmidt said in an interview with the Wall Street Journal. "I actually think most people don't want Google to answer their questions," he elaborates. "They want Google to tell them what they should be doing next." https://online.wsj.com/article/SB10001424052748704901104575423294099527212.html
Very Useful post with lots of examples..... It will help SEO guys a lot !!
Good points raised. I don`t think it will immediately effect results as you don`t have to rely on the google instant function to get a search all the time depending on which interface you use. There will be eventual results because of it though.
Thanks for the data; interesting to see you haven't noticed a change :-)
I echo Genius Goods Inc's views though in that I've noticed many people (especially those running e-commerce sites) who have seen a massive downturn in sales since Google Instant. I'm not sure how prominent this has been, although Webmaster World has been discussing this a fair amount.
Also for what it's worth, here's my views on Google Instant:
https://www.computerlover.com/general/google-instant-amazing-or-mistake.php
In summation (in-case people don't want to read all 1,100+ words :)), I think that Google Instant could be amazing although I honestly feel that it's been a little bit rushed out since I think it's got poor usability (whatever happened to "Keep It Simple, Stupid"?!?) and Google Suggest isn't always very relevant (Kudos on the amazing suggest 'ranking' for seoMoz though :-))
Thanks! Tristan Perry
i havent noticed any changes from an seo point of view but googles co2 output is through the roof! https://gpssystems.net/google-instant-bad-10-million-hamburgers/
A great post and some fascinating results there. I'm sure a lot of people will be surprised to see the long tail doing as well as ever after Instant. I just wonder how the general public will react. Taking off my SEO hat for a minute, I get a bit miffed by Google second guessing me and trying to steer me in any particular direction. I wonder if some users will be annoyed enough to switch to another search engine. Or will they even notice? I suspect the majority won't notice or won't be bothered because as searchers, we're a lazy bunch !
I really like this post. I'm tired of the "doom & gloom" as well as soon as any one thing changes. Do I think we'll see some long term changes with Instant? Absolutely. But that's to be expected with SEO. If you're an "SEO" but can't change with the times. Then you're not really an SEO at all. As long as people search for content on the internet. There will be SEOs.
We have seen a big increase in long tail searches for our site, also some of our big spending competition who have massive link building campagns have seen big drops, does this point to google seeing bought links?
hello, sorry for this comment.. i dont received any email response from your group..
i was charged 2 months aug 14 and sept 14 but i already end my subscription during the trial period that was July 20, 2010 and confirmed by your staff and also my member status changed to free member.
to be honest iam really amazed to your seomoz tools.. but i decided not to have it because i find it same as my current tools..please bring my money back, i dont have money to pay my bills..
I hope from this comment i get the attention of SEOMOZ..and i am hoping SEOMOZ is not a scam..
Rand please help me.. i dont know to contact you guys directly, i will try to contact you by all means"sorry for this comment post..
I'm a little late to the comments here, but I wanted to gather a bit more data before making any assumptions about the death/life of the long tail, etc. So far, from what I've seen (across 3+ weeks of data from 24 sites, with 3,182,487 visits from 1,197,374 keywords) there is only a very slight increase in search queries that contain either 1 or 5+ words. Queries that contain either 2, 3 or 4 words, on the other hand, have seen a very slight decrease.
It seems that Google Instant is putting the "chunky middle" on a diet. Here's the data that I've collected, complete with all the pretty charts and graphs.
This new instant search feature is way more powerful than Google Suggest. With suggest you still had to choose a phrase you wanted, hit search and be taken to the results page to find what you are looking for. Now results are displayed dynamically as you type, and searchers may stop to find something that interests them mid – search.
Hey Rand. Nice, calming post pal. I look forward to more and more hard data re: Instant as I just don't "get" it. But as I'm usually the one to color outside the lines, I assume it's my fault that I'm missing the "why" we need Google Instant.
I hope you are enjoying Milan and I assume you've got Geraldine with you. If so I hope she sneaks you off amongst the locals so you can truly experience Milan.
Bit of a relief to hear that longtail keywords have had little to no effect.
Wicked to see that Google Suggest actually shows SEOmoz has part of it's 'SEO' keyword suggestion range.
Thanks Rand for the calming pat on the back.
I can't help but think your comments are a little dismissive, Rand. I appreciate that not all websites are going to see a significant decrease in traffic but the real damage is going to come to smaller E-Commerce sites that relied on long-tail keywords to compete with the likes of Amazon.
It's not massively surprising that SEOmoz has seen little change in overall traffic levels and I highly doubt it's going to have a huge affect on the bottom line, but for many websites owners it will.
XAdd Comment
Interesting data. What do you think about the theory that fewer keywords will now draw significantly more traffic due to Google Instant "focusing the user's attention on more popular search phrases" as stated here: https://techcrunch.com/2010/09/21/guest-post-how-google-instant-can-help-and-hurt-seo/
This data is interesting in that it's not showing much yet. I think many folks have been waiting for data based on the availability of enough samples to look at how Instant is affecting things, but that's a bit over-eager. The searching public at large still has no idea what Google Instant is, and they are probably mostly still either not noticing at all, or confused as to why their screen is changing while they type. Like with many things, we've got to wait a few months or a year to give people a chance to learn how to trust the feature, and only then will it alter the way they type or whether they stop short on their query.
We have had a 20% increase in our longntail traffic after instant was launched!!
Thanks for sharing this information. Happy to see some more numbers coming out with some before and after shots to give factual insights to the Google Instant craziness. Especially silencing some of pretty crazy predictions on what it will impact.
Great post, I have been waiting to hear more about instant. I noticed right away that results are different before and after you press enter, this seemed to be the biggest thing to me, however it also makes me think I need to pay a lot more attention to the suggestions as well. Great information though, thanks a lot.
What though of the new premium on online brand management and maintenance/protection? By this I mean now that the suggestions are near universally immediate surely XYZ Company will become ultra concerned with not having terms such as "complaints", "rip off" or "scam" pop into the searcher/potential client's line of vision (independent of whether or not they actually click on the option - at the very least the subconscious effect will be damaging to brand equity). For this reason I feel Google Instant is a potential and even probable boon for online marketing and SEO. Thoughts, Rand?
Thanks for the stats before and after. This is a big change on Google and the fear for me is that the bigger sites will get stronger making it harder for the smaller companies to compete.
I'm going out on a limb here and state that the 2 compared weeks of data here are statistically incomparable and cannot be measured against each other because the first contains the holiday weekend for Labor Day that is observer in the US and Canada. Maybe elsewhere, I'm not sure. Regardless, search patterns vary too much during holiday weeks that include at least 1 day where government offices and most businesses are closed to be regarded as statistically viable.
What might be a better comparison is to segment keyword and traffic stats by a keyword pattern that has a few common elements and then compare that to a previous non-holiday week. A common keyword pattern should not contain keywords that produce large volume of visits that can be attributed to recent posts. A better pattern might even to be to compare traffic stats generated from non-blog landing pages as those are probably less influenced by trending searches.
I guess that being all said I would have a hard time placing faith in Conductors data for some of the same reasons.
To me Google Instant can be summed up in the same way as any of the engines Search Suggest: a "Keyword Funnel". It is not a coincidence that the majority of "Suggested" Instant searches are for keywords that contain AdWords ads in the top 1-3 center results...
Hi James, The data MEC used was purely based on searchers using G dot co dot uk and in the UK this didn't include any public holidays (just missed one in August) otherwise you're right, this will have had an effect. They intend to do a follow up to the post in another 8 weeks when they should have a lot more data. Hopefully in that time they'll see a full UK roll out to none logged in users as well.
Good input. I will be looking forward to future posts. Identifying data sets of statistical significance is tricky for measuring GI and forthcoming results are going to be very interesting... at least to me:)
Is there any data on how many people opt-out of using instant?
Its early days but its becoming obvious lots of users simply hate it (results appearing after one letter is typed is useless!!!) so it would be interesting to see some figures on people who return to the simple 'old days'
Marvalous Explntion of google....I realy like it.
to display with the chart it is helpful to me.....
thanks sharing with us
It seems to me that this is just a larger version of the suggestions Google has always given people. I don't see where it will impact search that much. You have to remember that people looking to buy something are typically a little more detailed in their search query and so I would say 3 and 4 keyword queries are just fine. Although, I might be a poor example when I have been using Google I have tended to completely ignore the suggestions of Google Instant, type my full phrase and then click search.
My thought about this subject are rather in that direction, too. Why should they have a separate algorithm for Google Instant? I suppose they don't. One will only show up in the suggestions if you are a real big brand or if the search term is very specific. Of course you have to submit that those suggestions can definitifely change the usual search behaviours.
Another tought which came to my mind regarding Google Instant. As far as I know it won't work in the Google Toolbar. Does anyone have any numbers how many persons search with the Google Toolbar or opens the Google window first? I use the toolbar ... I prefer the fast option.
There doesn't seem to be any really great numbers for this but I found this article where the author was looking for the same numbers - https://gigaom.com/2010/09/08/google-instant-a-mobile-app-approach-to-search/ . It led me to Chitika who had some numbers regarding Firefox toolbar / start page searches. https://chitika.com/research/2010/firefox-2nd-most-influential-player-in-search/ - they said about 9% of all searches came from Firefox. I think Google has more accurate numbers but shh.. they are a secret..
My experience has been similar--no noticeable change in Google traffic when separated before/after.
I think that not much will happen to long tail. Some of the already strong companies ranking for head terms might get more traffic but I think the whole issue is more related to UI.
Don't forget that Google can't read people minds. It is especially true for localized queries. How good is Google predicting exactly what user are searching for.
I find that typing Richard Getz without hitting return shows results that one might not expect. Richard Getzoff with OFF greyed out. Frustrating when someone is looking for me LOL (enter the you are a geek joke here)
Well it seemed like this would be the inevitable result. People will generaly gravitate to whatever gives the most immediate answers to their queries. Now of course this doesn't mean that if you are an online retailer that going product specific will necessarily yeild you more traffic. Before people get too excited and start filling their campaigns with pointless keywords it'd be best to see what is really converting for your site.
Hey, thanks for the info I have been waiting for seomoz's opinion and data on this!
Great article!
BTW whats up with Miami Hurricanes and Dolphins colors in the charts.
I agree - I'm not going to share my logs, but I'm really not seeing any noticeable difference.
Rand - Thanks for sharing the data. Glad to see some more numbers coming out with some before and after shots to give factual insights to the Google Instant madness that swept the industry. Especially squelching some of pretty crazy predictions on what it will impact.
But, I still can't help but have the take away from the whole Google Instant thing be - Eh [shoulder shrug], life goes on.
Good to know that you indeed shared this to us. I should have to say that I also noticed that long tailed keywords get more attention now but it doesn't mean it is more weighted. Although we cannot deny the fact that one way or another Google can actually index them more clearly primarily because long tail keywords become more specific than 2 word searches or 1 word search. Well, let's just work hard on our site because there's a lot of work to be done for sure.
There was a really good article today on TechCrunch about the subject:
https://techcrunch.com/2010/09/21/guest-post-how-google-instant-can-help-and-hurt-seo/
My main takeaways:
It doesn't really change the way SEO is done (i.e. linkbuilding) but it definitely affects what impacts our efforts will have.
We've just done a similar bit of research and post on the effects on UK traffic, here - https://www.mecmanchester.co.uk/blog/google-instant-data-after-12-days.html. We used some very high traffic sites across various verticals and only used data from logged in users (we don't have full roll out on g dot co dot uk.
The only area where we found a large enough change to warrant discussion was in the 7+ keyword category, which saw a noticeable increase, but for us the real analysis will come when G push this out to none logged in searchers.
IMO (personal) this will make SEO easier as it's educating the user in how to search more efficiently.
This is great. It partly confirms what I was talking about here https://www.danielduckworth.com.au/blog/how-google-instant-changes-seo/
However I think that the longtail keywords will continue to increase in volume and searchers become more comfortable using the search suggestions.
I figured that this would be the result, that the longer tailed keywords would get recognition. As so, SEO optimizers will likely have more work. This is definitely key to follow as more results flow in, and I'm willing to predict that the long tails will get even more exposure.
I'll be linking back to you from this blog post: Google Instant Search: Good or Bad?