Social Media Search

Just came across this. Thought some of you may find it useful? And it prompts me to go build me own custom version!

Searching together: The biggest challenge yet to SEO?

This video, from Techcrunch, shows an experimental version of google which includes digg-like social search functions.
Quite a challenge for traditional SEO? It seems like a huge one to me.
Would love to hear some thoughts from anyone in SEO - and from anyone else!
It's clearly a huge step forward from google's previous efforts in social search.

F**K the algorithm! It's time for the human touch

Since the rise of the search engines we've surrendered the nuancing of our experience of culture to the algorithm. In other words, the machine has been in charge of editing our choice.

Google's bots, maths, hard science has shaped what we all get to know about. Increasingly so. Google accounted for 87% of searches in the UK in May, 2008.

When humans get involved, then we start to discover things we didn't know we needed to know.

Blogrolls write this large; very human recommendation lists which extend and share our networks of trust. Trust me from the metadata I express on this blog? You're likely to consider my blogroll less advert and more recommendation.

The likes of Mahalo, Powersets, StumbleUpon and Ditto (disclosure, one of our own) take advantage of this, too. The human element of discovery. Tools for us to use rather than tools that use us?

The robots have played the role of editors of choice. Page 1 or die! In many fields they are very good at it.
But now it's time for the humans to take back a little control - particularly and effectively where things have a human, emotional dimension.

Traditional maths-led search is a little like someone who thinks they are quite well-enough socially connected, thank you, if they have the phone numbers of the people they know.
My question, aimed at tempting them to dip a toe in social media is: "How do you discover new people to call?"
Human-powered search recommends relevant alternative people to call.

We're playing with a ditto.net list at the moment - aiming to rank the Rock Stars of Web2.0.

We could just google appropriate terms and see who comes out on top.
But somehow, asking you what you think seems more fun - and more appropriate. How do YOU define what makes someone important in this context. Maybe humans are better at doing this than machines?

I'd rather know how human beings are ranked by other human beings - we're complex, more-than-the-sum beasts.

In that context I don't want to be a slave to the algorithm.

Google doesn't know what you are looking for

Stowe Boyd got me thinking about search.
His post points out that his /message blog appears higher than the yahoo message boards... on google's returns (wonder what happens on yahoo?).
So I tried the search term 'faster future' on google.
And ff comes in at No4 on the list. Hurrah!
Sounds good, dunnit?
But it's a bit rubbish really. I know what I'm looking for. Google doesn't. That's why the result I actually want comes out at number 4. The result I'm actually looking for should be number 1 - in fact the only result - shouldn't it?
Yeah, but google couldn't possibly know what I was thinking as I entered those search terms, could it?
Not yet - but it is working on it. And the answer - as with most things - relies on unleashing the power of networked communities.
Social search engines strive to understand, from my previous interactions, ratings, and recommendations, what I mean when I type a search term.
What a social search engine does is give context. And as we've often said before - without context there is no meaning.
So with standard old google search Stowe is right to let out a whoop of joy on appearing on page 1 of google's returns. But it's also why when I type Bike Test - and a mean motorcycle - I get a load of returns about bicycles.
I wish social search engines, such as google's custom search well - and salute their illustration of the power of networked communities of interest.

How NOT to do search

I liked this post for the real life example (poor Nike!) it offers.

AND it's got a great link to this SEO tool to help with keyword research.

Anyway - the mistakes big brands make with search are (according to writer Wil Reynolds, and detailed in his article):

1. Missing out on the long tail: Search daddy Danny Sullivan says: "Typically, big brands want to target the big unbranded terms like "tennis rackets," "golf clubs," or "running shoes." I do recommend that they target such terms as a way to position their brand in the minds of people who are searching. But they often miss terms like "golf club reviews" or "women's trail running shoes." Typically, these long-tail terms are the ones that convert best.

2. All-Flash sites with no alternative.

3. Not reinforcing search query results on landing page.

4. Not developing a descriptive meta description tag.

5. Making things 'cool' rather than easy to find.

6. Driving a user with a very specific query to the homepage - a waste of pay-per-click budgets

7. News flash—you can't pay to be No. 1 on Google anymore. An algorithm that decides how well your landing page matches with the user's search mission will affect where you rank (along with your cost per click, and other factors). Yahoo is apparently going this route as well.

8. Flash sites adversely impact natural search rankings

9. Flash sites do poorly on froogle

10. Great tools are a mistake without investment to drive traffic to them

Google Customized Search - search for social networks?

And while we're talking about the explosion in social networking - is google customized search (just launched, and see an example in the left hand column) the google response?
Anyone know if you can get it on your mobile yet?
I left the one on this blog open so anyone can contribute sites to help hone the returns. Have a try!