I’m not one of those people who claims that social media doesn’t do anything for your search results. On the contrary, I think it absolutely helps SEO.
For some, any notion that Twitter could influence your search rankings was blown into smithereens when it was revealed that Google pays no attention to links from Twitter. Call the cops!
That kind of thinking totally misses the point, because Twitter is a network of networks. Yours and mine, for starters. And I bet you that they’re overlapping right now, if you’re reading this and also active on Twitter. People talk beyond the virtual walls of Twitter. The network effect is an almighty thing when it’s in full swing. Raising awareness in a meaningful and relevant way is what Twitter is all about, as far as our business is concerned.
In any event, Twitter can also directly affect Google rankings for you in a positive way, starting with universal search...
Twitter helps you claim your universal search placements
It is crucial to own / control as many first page search results for your own company / brand name. Sites like YouTube, Slideshare and LinkedIn are easy wins, as far as universal search goes. Twitter is another way of claiming one of these slots. Twitter is always in the first five results for a Google search on 'Econsultancy'.
Twitter can do amazing things for generic keyphrases
Seriously amazing things. Read my analysis of how Mahalo came out of nowhere to rank highly for ‘answers’, to support Mahalo Answers (which only launched in December 2008). It did this by setting up www.twitter.com/answers. There are half a billion results for the word ‘answers’, and the Mahalo Twitter account sits at the top of the second page.
Frequency matters
Mahalo made Twitter a part of its business: you can ask questions via Twitter, and it posts answers via Twitter. This was a smart move, not least because it encourages a lot more activity. In eight months @answers has posted almost 54,000 tweets. Ask yourself whether 54,000 tweets and a third placed result on Google (for a ridiculously popular keyword) is a coincidence? Personally, I don’t think so.
- Consider that there are 54,000 individual pages on Twitter all pointing at @answers, as well as to Mahalo Answers.
- One tweet = one page; one page = one link to the homepage, and one link to the source of the tweet client (e.g. ‘Tweetdeck, or in this case ‘Mahalo Answers’).
More followers help
Points make prizes. More followers help create more internal linklove to your Twitter homepage, as well as to individual tweets. Frequency and reach!
Some outbound links are Nofollowed…
Big deal. Step back and see the bigger picture. Twitter drives reach and awareness in what you’re up to. It’s as much about encouraging your followers to spread word – and create new links - on other sites that do not bother with Nofollow. Andy Beal has some good ideas on why Twitter and Google need to dump the Nofollow protocol.
Retweets create linklove and awareness
This follows on from the last point. By creating compelling / useful / valuable content you can generate a lot of retweets, which – for link-based tweets – can mean lots of visitors. Traffic is one thing, but again it’s about the bigger picture. We live in a world of bookmarklets, and there is a large blogger community on Twitter. Links, links, links. One of my posts on measuring social media success generated more than 1,000 retweets, which in turn delivered 30,000+ visitors. Those people also wrote about it and bookmarked it on sites like Delicious (about 600 times), StumbleUpon, Sphinn, Digg and FriendFeed, which in turn generated yet more traffic, links, interest, and Econsultancy subscriptions / registered users.
Don’t panic if you failed to claim your brand on Twitter
Your Twitter URL (e.g @lakey) is important, but it’s not make or break. Hopefully you’ll have claimed your first choice name by now (note that I didn’t grab mine, alas). This matters a lot more if it’s for your company or brand name, but it’s not the end of the world in SEO terms because…
Twitter usernames are optimised for SEO
Your username (e.g. ‘Chris Lake’) is where the money's at, so to speak. Your brand name / keywords should live here. Twitter recently optimised the username to display the username first, so in Google you’ll see ‘Chris Lake (lakey) on Twitter’. Very helpful.
Some URL shorteners are better than others
301 redirects are what you're looking for. I use Bit.ly, which works very well and provides permanent 301 redirection on its links. Others aren’t so generous, willing, or able. Search Engine Land has a great chart to help you choose one.
And some 'URL shorteners' are horrible
There is a horrifying trend towards 'framebars', which sit at the top of the page and make your world slightly worse. They can have all manner of implications on SEO, not to mention the user experience. The Digg URL shortener is one that has put Digg firmly into ‘jump the shark’ territory for me, and created controversy a few weeks ago. It’s like a nuclear arms race, when everybody involved should know better.
Customising URL shortener extensions is only good for users
Many of the URL shortening services allow you to customise the link (e.g. ‘Bit.ly/customisethisbit’) but it’s not going to help with SEO specifically (due to the nature of the redirection). It might help persuade humans to click the link, but that’s the long and short of it, as far as I can tell.
Contextual tweeting helps
Go niche for the best results. This is about attracting the right audience, as much as anything. When you’ve got them try not to alienate them with too many off-topic tweets. Help people quickly figure out what you’re all about, when they browse your Twitter feed. And it’s the same with Googlebot, right? If context matters on blogs, websites and web pages then I’d wager it matters for Twitter too. A liberal sprinkling of keywords and phrases won’t do you any harm...
Individual tweets can rank highly
Twitter is about sub-folders (e.g. 'twitter.com/lakey') rather than sub-directories (e.g. 'lakey.twitter.com'), so Google love is passed down through to tweets. As such we’re seeing individual tweets ranking relatively highly in Google, typically for long tail phrases.
Integrate Twitter with your website and beyond
Prop it up. Give your Twitter homepage some linklove from your site. Let your audience see it and help them embrace it. Our own Twitter experiment, rolled out on a whim one Friday in March, was based around the notion that we’ll let people talk about us on Twitter and then feature their tweets on our homepage. We then tweaked the code to pull in mentions of specific blog articles (these sit in the blog sidebar, labeled ‘Twitter Buzz’). Further integration is planned. It is about driving engagement and participation, on our site and on Twitter, and also acts as a great feedback channel. All of this helps feed some of the other things I’ve talked about.
Add your account to Twitter directories
A no-brainer. Start with WeFollow and then check out some others…
Optimise your bio
The Twitter API allows third party websites / apps / directories to pull through this information. And this is precisely what Google uses as the description alongside your search listing. Here’s mine, and this is exactly what it looks like in Google: ‘Editor in chief at Econsultancy, blogger and entrepreneur. I always do my best work when listening to The Fall.’
Nanocontent FTW
What do we mean by nanocontent? We mean the first 11 characters in your tweet / headline. Consider that Google will be able to index part of your tweet in its results headings (the first part). This is a likely SEO ranking factor and it’s persuasive, from a human perspective. Front load those keywords where possible. More on nanocontent and online copywriting here.
Agree? Disagree? Let me know what I missed...
Chris Lake is Director of Product Development at Econsultancy, an entrepreneur and a long-term internet fiend. Follow him on Twitter or connect via Linkedin.
2:49PM on 10th August 2009
Good day. Since Econsultancy does not use Custom Named URI Shorts, how can you come to the above conclusion? What about keyword rich anchor text? What about the relationship of that keyword rich anchor text to the surrounding text? Sure, there's a 301 there but, there is quite a bit that happens prior to that 301 being processed.
I think if you utilized the Custom Name option yourselves, you'd probably have a different view on their effectiveness. You have 140 characters to work with, make every single one of them count in the overall equation. Random Base 36 makes for butt ugly URIs. :)
Director of Product Development at Econsultancy
4:12PM on 10th August 2009
Agree about the butt ugly URLs, though as previously mentioned we don't customise due to the way we track buzz on Twitter. We can join it up but it's not right at the top of the agenda right now. We've filed this under 'nice to have' rather than top priority.
Aesthetics aside, I was under the impression that the 301 redirects *totally* ignored the shortened URL redirection link / URL text, passing on PageRank and anchor text from the host page (as opposed to the link itself). You have any evidence to the contrary? Happy to stand corrected if it works...
7:42PM on 10th August 2009
Understood. That is how most entities would handle the situation. That's one of the issues I run into with Corporate America all the time too. They tend to low prioritize the things that have long term beneficial impact.
You're talking about the "immediate" effects. There are long term benefits at play here. At the same time that 301 is being processed, there are other textual processing routines taking place.
You mention above about NanoContent. If you have a URI at the beginning of the Tweet that uses a random Base 36 string, how do your Base 36 URIs stand up to the first 11 test? By the way, there are no hard numbers here. The Twitter title before truncation can range between 48-61 characters depending on the username length and construct of the Tweet.
http://www.SEOConsultants.com/Twitter/Titles/#CharacterCounts
What type of evidence would you like to see? I could point you to ALL of the guidelines on how UAs handle contextual content, there are well over a 100 pages for you to read. We're discussing anchor text here, there is nothing to prove and/or test. One of the biggest benefits of using the Custom Naming conventions comes from being able to include a few more keywords in the anchor, not to mention the visual that is presented to the user. People tend to distrust randomly generated URIs these days. The fact that I can negate some of that distrust by using a Custom Name is enough reason for me to utilize them diligently.
Now, let's talk about all the third party scraping of Twitter Statuses. If the Custom Name can be maintained during its Tweet Lifecycle which I estimate to be 7 days (normally), you're getting the benefit of that keyword anchor text no matter what happens in the process. There's a lot going on and I'm not going to be the one to point out ALL of the evidence. With your volume of visitors, you should be able to easily test this out.
By the way, Google have indexed 247,000,000 results from Twitter /status/ in the past 24 hours. If you only have 140 characters to work with, don't you think using every character from a 'NanoContent" perspective is to your advantage? Including the Custom Named URI Shorts?
Director of Product Development at Econsultancy
8:07PM on 10th August 2009
There are other things that have both immediate and longer term benefits that we are prioritising. Whatever marginal uplift in Google we'll get from this is less of a priority than optimising other aspects of our website, well outside of the scope of SEO. If I could have everything done overnight then we'd run an experiment and integrate the Bit.ly API, but as is, it will have to wait before we schedule in the time to look at it. We will get into this somewhere down the line. Corporate America has far greater resources than we do, so the comparison isn't entirely valid.
I never place links at the front of my tweets, nor do we do this via Twitterfeed. Links in tweets need labels, and I label first, and link later. As such Google isn't going to be highlighting the link in the results, and if it did it would be a wasted opportunity ('http://bit.l' as the first 11 characters doesn't do much for me).
First I've heard of it. People tend to trust the referral source. It's the referrer on a social media site like Twitter that makes it a credible link, as opposed to the URI itself.
8:41PM on 10th August 2009
<meta content="18 observations on Twitter and SEO http://bit.ly/3pGEP8" name="description" />
The above meta description is from the Tweet that @Econsultancy posted earlier that drew me to this topic. There are 55 characters in that Tweet, 20 of them are used in the bit.ly URI. That leaves 35 characters that really make sense in the overall equation.
I took that same URI and Custom Named it and am using it in my Tweets pertaining to this topic. I came up with a Tweet construct that looks like this...
<meta content="http://bit.ly/18-Observations on Twitter and SEO" name="description" />
I trimmed the Tweet to 48 characters. I can't ask you which one you find more appealing, the visual, as I know you'll say it doesn't matter. But, for the rest of those reading, which one of these three would you choose?
Now, I may come back a little later and Retweet again. This time I'm going to mix things up and repurpose the original Tweet...
And, there's a chance it may even get a 3rd or 4th Retweet at which time I'll mix em up a bit again...
If I really thought hard about this one, which I didn't, I would have used a http://bit.ly/Twitter-and-SEO Custom Name instead of the 18 Observations version.
There are many ways to approach this. We've exchanged Tweets about the Custom Names in the past and I'm sure you know my views on this. If you don't use them, I believe there are many missed opportunities that are below the surface of it all. http://Search.Twitter.com/ would be a prime example.
If we were to have a contest on who could construct an optimized 140 character Tweet and you were a participant, would you use the Custom URI Naming conventions?
11:46PM on 10th August 2009
Mahalo Answers and Twitter.com/Answers are on page 2 for me right now (signed out, in US). Just sayin...
Director of Product Development at Econsultancy
12:51AM on 11th August 2009
@Joesph - yep, schoolboy error on my part. They're at the top of the second page. Still a result given the timeframe and popularity, but no significant movement in the past few months. Will update the post. Thanks for the spot.
12:26PM on 11th August 2009
I like to claim the name Google Ranking Account on Twitter. Just wondering if I should put a hyphen in between the words like this: google-ranking-account or can the search engines also read googlerankingaccount?
6:27PM on 11th August 2009
Really nice article but I think you've missed one thing that can improve SEO with Twitter: The link to the website of the client you have used to post your Twitter update.
That link does not have the nofollow attribute and therefore passes linkjuice.
There's a script that allows customization of that link (including anchor text) to whatever website or blog you want and pass Twitter.com link juice.
Some may consider it kind of a black hat exploit and it may not work for ages but for the moment, it does.
If you want more information about the script check my blog post on the subject http://bit.ly/12BjP
Director of Product Development at Econsultancy
6:31PM on 11th August 2009
Hey Gaetan,
I spotted your post on Twitter earlier and retweeted it. Great catch. Is 'Mahalo Answers' doing this I wonder? ; )
10:40AM on 13th August 2009
Personally, I don't like Twitter.
To me it seems to really be a PR tool for those whom are already in the public eye and has little to offer in terms of SEO.
Not my Cup of Tea but its popular enough- it must be doing somethign right.
1:41PM on 13th August 2009
I notice that google does take a massive account of twitter links. One of our keyphrases is Banksy Prints. I noticed a twitter account had this in the naturals only after a few days and a few thousand tweets.
5:03AM on 1st September 2009
Thanks for the post.The points you have mentioned out here are really very valuable.Thanks again.
5:42AM on 25th September 2009
Search engine Optimization uses keywords in text , heading to generate links to your websites , they build maximum search given various options .
5:43AM on 2nd October 2009
@Chris Lake: "Customising URL shortener extensions is only good for users"
Many of the URL shortening services allow you to customise the link (e.g. ‘Bit.ly/customisethisbit’) but it’s not going to help with SEO specifically (due to the nature of the redirection). It might help persuade humans to click the link, but that’s the long and short of it, as far as I can tell.
"Besides that Mrs Lincoln, how was the play?"
To think that making URLs more usable for users is unimportant is like saying content is only there for search engines. It's a short sighted view which I am constantly amazed that people still have. One of Twitter's most compelling features, the @replies, was empowered by their use of well designed URLs.
11:48AM on 16th December 2009
Your tips about twitter are useful, and I agree with you that one has to have a passion about what he writes and avoid marketing hypy as much as possible.
2:36PM on 14th April 2010
Great info Chris. I know this is an older post, but it's relevant to my post from yesterday.
http://www.localsearchpilot.com/local-search/twitter-seo I'd like to know if anyone else has started to see twitter accounts ranking high for keywords
8:03AM on 30th December 2010
Its great info; I am new to all this SEO and Twitter thing and trying to understand what reaction I receive from using utilizing Twitter. I have been trying to send "thank you" and "how are you sir/mis" direct messages and somehow, many of them visit my website !!!
Just posting something on wall may be good if you are having a huge amount of FOLLOWERs. I have two accoutns, one with more than 1,300 followers and other with only 300 followers.
I tried to put some attractive terms at wall of my account with 1,335 followers and sent private messages through other account, which has only 309 followers but amazingly, analytic showed huge different in traffic trends from both twitter accounts at my websites.
Attracting vistiors through twitter is good for nothing...until your product match their mind and most of people at twitter are not there for gathering,,rather, they are only engaged in doing SEO stuff. Its useless to attract irrelevant traffic, as in my case.