Getting New Sites Crawled Using Twitter

[kidder]

I setup and installed a new blog yesterday and I thought rather than link it to any of my sites it would be worth Tweeting out the new URL to see how quickly it was picked up. Today its already in the index and its showing about 16 hours of age. Now I can't say it was 100% Twitter but given the amount of bot activity tweets seem to generate its a pretty fair bet my good old twitter account has at least one use...

[tedster]

Thanks for that report, kidder. I can't say I'm surprised since Google has a powerful passion - even obnsession - for URL discovery.

I just heard today about a similar test that went even further. The site involved got a page indexed and RANKING with only Twitter links. I doubt that could be sustained for any period of time, but apparently it can work short term.

[kidder]

To be more specific my domain name was a 3 word term and its already ranking #1 for that term. As you suggest its not going to last but it was an interesting test. I will also add that until the last toolbar update my twitter page had a page rank of 4, now its back to zero. Google obviously has a specific set of rules for Twitter so we will have to do some more testing just to see what we can prove. One of our classifieds sites records page hits on each advert so its interesting to note that each time a new ad gets posted it feeds to twitter in our RSS as a post and almost instantly show between 25 - 30 impressions.

[elchenuk]

Interesting stuff, links on Twitter are nofollow so to think that you can get indexed let alone rankings is an eye opener.

[mevans05]

We saw this happen with the launch of our site < sorry, no links to your own properties >. Check out the Google cache for it and you'll notice there is a WebTrends tracking code appended to the URL proving that it came from our Twitter account.

[edited by: tedster at 4:17 pm (utc) on May 14, 2010]

[milosevic]

mevans05, that link will be censored in about 10 seconds but what you describe is not a good situation - you're creating duplicate content before even the original version of the page has been indexed.

As for links being nofollowed on Twitter, Google can easily handle links from particular sites in a different way to the rest of the net and they already do this with Tweets. The nofollow is just a suggested directive anyway and may be ignored in certain situations on ANY site.

[pageoneresults]

Interesting stuff, links on Twitter are nofollow so to think that you can get indexed let alone rankings is an eye opener.

This is all surface stuff. One you Tweet a link on Twitter, depending on the makeup of your Network (Following/Followers), it could be like sending out 1,000 Link Requests!

I've been testing all sorts of stuff on Twitter for almost two years now. I can tell you that NoFollow is just there for the first part of the equation. Once you Tweet, and the API bots pick it up, that NoFollow get's stripped and away she goes! I know, I've watched it over and over again. My logfiles are a clear sign of what happens immediately after Tweeting a link.

I've had stuff show up in less than a couple of hours depending on time of day, how many people Retweeted, etc. There are all sorts of influences here in the process. Twitter's API is one big pipe of links! And Google love Tweets that are well constructed, use Anchor Text (Custom Named URI Shorts), are Front Loaded, and are Retweeted by quality folks within your network. It's one big happy family of sorts. :)

[kidder]

So where does this leave us with nofollow? I notice speedy indexing with some of the other big social media sites as well, nofollow sites..

[misterjinx]

Hello I did some scientific tests in italian language from October 2009.

A single retweet from a trusted Twitter account make indexed the site within few hours (less than 16, also 2-3 hours at max).

Please also note that Google in real time search snippet expands the url shortened and displays the title of the linked page.

When I did tests I also noticed that for the first 7-10 days after indexing you had the tweet and linked page in SERP.
I think they were considered as a single entity because the tweet and the linked page were coupled (for example in #13 you had the tweet and in #14 the linked paged, and so on).

After the 7-10 days the tweet and the linked page started a different live in SERP: they separated and changed indipendently their position.

[directwheels]

links to your new domain is definitely a good way to get Google to start indexing your site quickly.

i have been doing the same thing since probably August of 2009. it seems to index my new domains quickly everytime. as to ranking, i am not too sure since i don't rely on it to rank my new sites.

i have a couple of twitter accounts i use solely for this purpose. those accounts have no followers and it doesn't seem to matter.

[wallarookiller]

Even without twitter I can have domains indexed and ranking fast. It has nothing to do with it. nofollow, and redirected links are what they are. It's just how it is. Why would Google or any other search engine do anything different?

[kidder]

i have a couple of twitter accounts i use solely for this purpose. those accounts have no followers and it doesn't seem to matter.

Yeah I have new Twitter account that appears to generate a lot of bot activity, as long as Google plans to serve up to date data pulled from Twitter its not going to change. I wonder how much of a drain it must be to monitor and serve updates from Twitter...

[Renaud_Joly]

Tweet generates automated link because tweets are repeated in directories / mashup.

[pageoneresults]

Here's are a few tips on optimizing your Tweets for search...

1. Front Loaded Tweets - Think of your Tweet like a META Description, you have 140 characters to make your mark. Be sure to put the most important part of your message at the beginning of your Tweet, Front Load it.

2. Anchor Text - Pay careful attention to how you construct your URIs for Tweeting. If using a shortening service like bit.ly or j.mp, use the Custom Naming option and get your anchor text in there. Use hyphens to separate words. Don't overdo it.

3. Custom Named URIs - I like to use URIs that blend in with sentences. When using bit.ly I might take the first three words of title, if appropriate, and use them as the custom named URI. I'll then start off my Tweet with the URI and lead into the rest of the title and/or sentence, like a precursor. I'll also do them where they reside in the middle or end of a sentence, it all depends on the document title.

4. Retweets - I Front Load Retweets. I'll place credit at the end of the Tweet instead of at the beginning like most others. I treat Retweets exactly the same as a Tweet.

5. Retweet Scrubbing - I'll scrub Retweets fixing case issues and typos. I'll use an appropriate title if the originator changed title based on destination document. I see this happen all the time, it fragments the meaning of the Tweet in some ways. The new RT feature prevents this from happening but doesn't allow scrubbing and/or commentary.

Those are just a handful of quick tips to get you started. Even though you only have 140 characters to work with, you'd be surprised at what you can do. Use those to your advantage, I have been and I think it works, common sense tells me it would. :)

Sidenote: I also use trademark symbols e.g. ® or ™ where appropriate. Don't ask me why, something just tells me that they enforce branding at some level. Twitter support most HTML Character Entities.

[Josun]

Thank you pageoneresults for this nice post! Your recommendation of blending URI is something I didn't think about it before. It makes more sense.

[mevans05]

milosevic - canonical tags mitigate the duplicate content issue. the tag was just missed in implementation.