SEO 2009: Adapt or Die
Ian Lurie Dec 29 2008
There’s a meteor headed our way. Plug your ears, sing LALALALALALA all you want. It won’t change a thing.
Google’s about to make search engine optimization (SEO) into marketing, for good. How? They’ve been slowly adding user behavior to their ranking algorithm. Bet on far bigger changes after January 1st.
I won’t go into the many painful technical details of how I’ve arrived at this conclusion. If you want to know, Twitter me and I’ll cheerfully bore you to extinction.
Here’s what’s you need to know:
Organic Click-Thru Rate Will Matter. A Lot.
The organic click-thru rate (CTR) is going to affect rankings more than before. In English, ‘organic CTR’ is the percentage of people who click on your unpaid search listing versus those who see it and don’t click. So, if 10 people search for ‘rubber baby buggy bumpers’, and 5 click on your listing, then you have a 50% organic CTR:
A higher organic CTR will tell Google you’re more relevant and authoritative for that search query.
What you need to do: Improve your organic clickthru rate
- Write a good description tag. The description tag is the summary text that shows up in your search ‘snippet’. A good, compelling description tag will encourage more folks to click.
- Write a good title tag. The title tag shows up at the top of your search result listing. A well-written title tag will also persuade more folks to click.
- Freeze. If you go and start clicking furiously on your site in the search results in an attempt to improve your rankings, you’re… well… mentally challenged. Google’s figured that one out. See the next item for that information.
Time On Site Will Matter. Even More.
Time on site after search (TOSAS – you heard it here first) is the amount of time someone spends on your web site after they click on an unpaid Google search results. The amount of time folks spend on your site after clicking a Google search result is also more and more important:
I’ve been tracking organic search traffic for client web sites over the last year. Sites with improved time on site saw improved organic search traffic within months of their improved time on site. During that time, those sites weren’t conducting any SEO and saw no improvement in link popularity. Their content wasn’t stale, but they weren’t writing their brains out, either. All 5 sites were 1-2 years old. The only thing they improved? Usability. Time on site.
What you need to do: Clarify your offer, and watch your analytics
- Make sure your description and title tag sync up with the page to which they belong. You want improved time on site? Give people what they expect when they click on your listing in the search results.
- Track your organic landing pages. Every page on your site is a potential organic landing page. Search engines pick the page they’ll list in the rankings – you don’t. Those pages are your organic landing pages. So watch the pages that draw search traffic and backtrack. Figure out the keywords that are generating traffic to those pages. Then adjust those pages so they make sense in context of those keywords.
- Use your analytics. Watch time spent on each organic landing page. Test those pages, obsessively. Find the best combination of headline, copy, layout and offer. Then do it again.
Don’t start whining that this isn’t SEO. It is now. Get used to it.
Social actions will matter, too.
This one’s been evident for a while: Search engines are also adding bookmarking, stumbles, Diggs and other information into the equation. If you can get more folks to add your site to their Google Notebook, or promote it to the top of Google Searchwiki, or bookmark your site on Del.icio.us, then yeah, your ranking’s going to probably improve a bit.
Rankings will matter. Even less.
Bruce Clay may have gone a bit far when he said rankings are dead. They’re maimed, but definitely still alive.
But rankings don’t matter. They are, at most, a means to an end. And with personalized search, rankings will be harder and harder to consistently track. Instead, track:
- Traffic from organic search. How many visitors are you getting from organic search results?
- Keywords generating traffic. What keywords are generating that traffic? Are they branded (meaning they include your brand name) or non-branded?
- Bounce rate from organic search visits. How many visitors from organic search look at one page and leave without clicking to another page? The more folks that leave, the higher your bounce rate. A high bounce rate is bad.
- Conversion rate. Hard to believe, but what really matters is sales, and leads and stuff.
It’ll be harder to spam the search engines. For a while.
It’s harder to fake behavioral triggers. You can bookmark the hell out of your site, of course, but that’s not going to be enough to tilt the algorithm in your favor. You’re going to have to (gasp) deliver real value to your audience.
However, the first company to hire 1000 people around the world to all carefully click, browse, bookmark and otherwise generate apparently legitimate traffic to a site will probably prompt the major search engines to once again tweak their algorithms, adding some new ranking factor, and ruining it for the rest of us.
What you need to learn.
If you’re an SEO professional or The Person Who Got Stuck With SEO, you’d better add a few skills to your tool belt:
- Usability. Don’t confuse this with design. Google doesn’t give a flying poop if your site looks nice.
- Analytics. You’d better be good at analyzing how long folks spend visiting your site, and how to improve it.
- Writing. I don’t mean the act of pressing keys on a keyboard. I mean the act of making yourself understood when pressing those keys.
Does this mean content’s gone? Woo hoo! No more writing!!!
Yeah, no.
Search engines still care about content, link authority, visibility and the like. They’re just adding these behavioral hallmarks to the equation.
Plus, if you write crappy content or just never update your site, visitors will click through to the site, then bounce back to Google, lowering your TOSAS and killing your rankings.
So it’s still about the content.
Other resources
While researching this post, I spent a lot of time poring over Google patent applications. The best plain English description of the patent that caught my interest (and a few other people’s), is SEOMOZ’s nifty explanation.
I also read a whole bunch of different forum threads where lots of smart people were discussing comments made by Bruce Clay about the death of rankings. The best thread I found was on the Cre8asite Forums.
And, I plumbed the depths of my extremely messed up, travel- and kid-addled head. So don’t be a lemming: Read what I’ve written, check your own assumptions, and then let me know what you find. I’ll pull those findings together here.
This article is only about Google, but trust me, it applies to every major search engine. If you think Yahoo! and Live won’t immediately race to match Google feature-for-feature, you’re insane.
Ian Lurie
CEO
Ian Lurie is CEO and founder of Portent Inc. He's recorded training for Lynda.com, writes regularly for the Portent Blog and has been published on AllThingsD, Forbes.com and TechCrunch. Ian speaks at conferences around the world, including SearchLove, MozCon, SIC and ad:Tech. Follow him on Twitter at portentint. He also just published a book about strategy for services businesses: One Trick Ponies Get Shot, available on Kindle. Read More
40 Comments
Comments are closed.
Wow.
Seo… complex it is… but very fun!
thanks for the great post!
I’ll have to watch my search engine and title tag!
We are not worthy! Thanks for this, Ian – more essential reading for everyone with a website.
Oh that people would take heed and pay attention to what really matters instead of just bleating about their lack of results and desperately chasing useless links.
Time to really take notice and get “conversation marketing” going. Time for a sequel, perhaps?
Very interesting post – and makes a lot of sense. I think time spent on site and whether visitors choose to tell their friends and contacts about your site should be part of the natural rankings. Sure does make things hard for us webmasters though!
A great subtitle for this post could be the changing nature of SEO for the rest of us. Thanks for doing the heavy thinking and digging on this. While there are many “SEO experts” out there giving advice, many folks handling this task are the ones responsible for site content. And they frequently can’t make sense of SEO strategies. It is neither art nor science.
You make a great point that the addition of behavioral aspects to the algorithms doesn’t reduce the importance of content, but strengthens it. First, you need to make sure you have something to say. Second, you need to make sure people can find it. And finally, you need to make sure it is easy to read one they get to your site.
Jeff Cohen
http://koroberi.com/blog
Very interesting! Seems like google adwords were for testing the new algorhythm components like page visit time and ct.
How do you think they manage new sites? If they rank in the places 30-80, they will never get clicked by anyone.
If I use google analytics, google knows everything about my jump off rate and my visited time of direct entries. That could be a problem, I think – or am I chickenhearted? :D
Excellent post.
I’m interested in hearing more about how Google is collecting data for TOSAS. Data from Google toolbar users?
Could an alternate explanation account for the increase in organic traffic after usability improvements? More repeat visitors, for example?
Was the traffic from Google up significantly more than that of other engines?
Great post. My sentiments exactly. We are going to have to start making valuable content that is useful, as of course Google wants us to.
Great post! Now I’ve got to go adjust my title tags… *sigh.
@Jim I can only guess at exactly what tools they’re using, but I would assume toolbar data, analytics and, most important, IP + Cookie tracking. I was tracking unique organic visits, so repeat visitors wouldn’t have affected the numbers. I did indeed track Google separately.
Great post Ian….well thought out and very thorough.
I wrote much more in my FIRST post here, but I forgot to add my email, hence all is lost!
I agree with this post though the question I raise is where Google will gain most of this intelligence.
Click-backs to the serps I get, but for those not using GA or Chrome, how will G gain knowledge of a sites bounce rates, tos, or other usability metrics??
Thanks again for a great post…keep them coming!
(i am c and p’ing this post prior to submitting) ;-)
-MB
Thanks Ian
The people just looking for a quick answer, in my site’s case, usually find it really quick. Like an alternate guitar tuning for example. A lot of times they’ll get right to the page, see the tuning and probably be gone in less than a minute. So then should I kick myself for providing straightforward easy to find information?
If Google uses time spent on the site to gauge listings, how can that directly correspond to a user finding the relevant information they need? Because relevant information can really be connected with time spent. Am I right?
By the way, Ian, I love the ‘flying poop’ part ;)
There seems to be one big thing missing that I’ve been playing with from Google labs and search. User voted content on Google.
Yes, it’s coming, and if you want to see it now, just go to Google labs and check it out. I think this will play a huge factor in the rankings, as actual users will be able to tell Google if it’s good or not.
@Mike I think they’ll get the vast majority of their data on click-backs. After that, aggregate toolbar data would be my next guess, but that’s only a guess.
@Jason Google won’t base their rankings purely on time on site. Don’t forget keyword relevance, clickthru, bounce rate, link authority, trustrank, etc.. It all adds up to quality sites like yours getting better and better exposure.
Thanks, Ian, for a great article and for taking the time to fill us all in on the changes coming our way. To all of those folks who are planning a redesign on their websites in 2009, read Ian’s article first…several times! Instead of spending hours worrying about colors and images, work on your title tags, descriptions and content. You probably would be better off not redesigning your website at all. Most website redesign projects are started because folks believe it will help them draw more traffic. Very often it has the opposite result. I am not saying that your website should be ugly and I don’t believe Ian is either. Simply put, if no one can find it, it doesn’t matter what it looks like. Roll up your sleeves and get to work on organic search engine optimization instead. You will be glad you did. Thanks again, Ian!
Hmmm… isn’t this how SEO myths are born? As someone that has been writing (studying and testing) all things relating to behavioral metrics for more than 18 months I can say that calling this a reality is at best supositional and at worst, SEO FUD or glory hounding.
As I have mentioned in articles disputing Search Engine Watch and Search Engine Land’s collums on ‘bounce rates’ – these are noisy signals and unlikely to get any serious consideration in a non-personalized environment. To that end bot (search quality lead) John Mueller and Matt Cutts both have poo poo’d them as ‘noisy and spammable’. I have talked with Barry and hopefully the gang at SEL will be doing an update post in early 09.
You are making statements without ANY proof, evidence nor statements from Google (beyond looking at query analysis which Marrissa confirmed). That is simply bad form… and to what end? What is your motivation? Potentially suspect if you really care about your readers and the community at large.
If you’re interested, get in touch and I can explain the problems relating to BMs and even pass along further papers/patents.. the (very old) one on HRFs is not even close to the full level of interest Google has shown in this area, but it is problematic still.
Hunches and reality are different beasts and one thing I learned from Bill when talking about patents is to use phrases like ‘Google might be’ ‘Google has been looking at’ as opposed to definative supositions as U have taken here…
Sorry to bitch, but as someone that has been writing about this for well over a year… it makes me sad when peeps ‘all of a sudden’ start calling things a reality which may not be…
Wow great post! Things are going to change a lot. I wonder how Google will get this all done…
I think it’s a progression that Google will depend rankings on the visitors itself instead of what the Google bot thinks about your site. It’s done with making friend with the bots and the spiders. Seo future might be hard…
Google is trying to change its ads habit to benefit advertisers. Think that’s part of Google recession proof technique, good work!
But no matter how, content still the king.
Bravo!
You mean we have to revert to the old days? Straight up SEO with REAL CONTENT!?!?! Call me kooky, or old school but i was doing this with sites for webcrawler, infoseek, and lycos back in ’97. It’s about time.
How would Google calculate Time on Site into their algorithm? Once a user leaves google.com (clicks through to a website) – Google will have no idea how long a user spends – i.e. Bounce Rate on the landing page and subsequent page of the website itself.
The might be able to do some crappy calculation of the same user (Which Google has cookied) returning back to Google.com? Not sure that’s enough to really add CTR as a major factor.
Plus, while Google has done some things to flag “click-fraud” in the paid side, transferring that risk factor into their natural algo is a can of worms I think they’ve thought about….
Don’t get me wrong – I think CTR would be a great addition to the algo.. It’s logical, I just struggle with the technical, privacy, and fraud issues that would unfold.
We’ll know soon enough…
Thanks for the great article and sources regarding Time On Site as a factor for Google ranking. @lookadoo also gave me a good source regarding bounce rate playing a role that can add to this conversation, hope it helps: http://www.webpronews.com/topnews/2008/11/21/is-bounce-rate-a-google-ranking-factor
Hi
Agreed! And here i like to add a single line;
Seo is all about patience & persistence with a combination of virtue to subdue search engines.
Cheers
Nawaz Shahzad
THANK YOU THANK YOU THANK YOU THANK YOU THANK YOU. Where do I send the box of chocolates?
@Mandy Oh, just leave them somewhere in downtown Seattle. I’ll find ’em. Remember, if I’m wrong, it’s all SEOMOZ’s fault!
This was a very informative article. I will read your blog often.
Excellent article. Thanks for your hard work. SEO 2009 – just what I was looking for.
Ian,
Well written, I must say. Having 10 years in the SEO industry, I’ve worried these days would be upon us, and this is really going to complicate things to a level that is going to see a lot of SEO experts pulling out their hair.
Hi, This is really an interesting post. I was also noticing some drastic changes in Google algorithms for the last few months. We really need to spend more time on site analytics for better site performance.
Correct me if I am wrong but if time on site is going to matter that much, won’t people just be sticking a bunch of video’s, and java games, or stuff on a site and get a better ratio?
So now is the time to take SEO seriously; learning to adapt to the times in order to make a good impression online. Well good to note that there will still be a focus on usability and content. And for those who are looking at ranking in a high regard; then this content will wake them up since ranking’s reputation will be affected..well slightly.
The SEO Moz you reference is an article written about the patent filed by Google in 2005…. is this old news or are you saying they will begin implementing those changes soon?
@Brian Definitely not old news. Figure if they filed that patent in 2005, they’ve spent time since then working on adding those benchmarks to their algorithm. Plus, changes in the rankings, and testing we’ve been doing, point to Google accelerating and launching more and more algo changes in the last year.
Have you shared any of this info with some of the SEO elite (Danny Sullivan, Rand Fishkin, etc. etc.)? I have been looking for someone to do a decent treatment on these and other search engine changes for a while now and I have yet to see a useful panel at a conference, etc. on it.
I just got back from SMX West hoping that someone would be giving some detail on some of the things that you have outlined in your post as I too have observed many of the same things that you have and so far, I’m seeing very little.
I think your post should be “sphunn” (and so far I haven’t seen it there) so I’m going to create a Sphinn account to get it there, I hope it will spark a useful conversation on the many changes happening in the engines of late.
Wait, are you saying I’m not part of the SEO Elite?
Kidding.
I haven’t, but they’re all aware of it. Rand wrote one of the patent analyses I used to figure some of this out.
Call me crazy but I’m not 100% convinced this is something to really worry about. It seems content and inbound links are still king. But good info nonetheless.
Cheers!
Mark
Great read. I do tons and tons of adwords and still find SEO a complex maze of twists and turns. This article helps but still wish I had more confidence in what my money was going towards when I pay an SEO organic expert.
Excellent: more evidence to backup my insistence that businesses invest in content and not just graphics.
You know – this approach actually makes sense and REWARDS those people who have actually made a genuine effort to create:
1) Novel and informative content
2) A clear, usable easy to read/navigate website
Thanks, mate. A good read.
Very good information. In terms of Google is looking at click thru rate, do they look at Click thru from many different IP as well?
Very interesting. I think 2010 will actually bring more changes than we have had in SEO, because with Caffeine, Google finally has the flexibility they needed to be faster than us (=SEOs) and tweak the algorithm before we spam it with our SEO optimised pages