SEO Blog - Internet marketing news and views  

Making sense of real-time social search

Written by David Harry   
Wednesday, 08 July 2009 08:48

Going beyond the catch phrases

Ok, since I wrote the post last week about ‘real time’ search engines I’ve had more than a few interesting convos and was even contacted by one engine purporting to be a ‘social search engine’ which is a term we’ve not heard much of since last year. One thing is for sure… the terminology is just whacked out there… there is no real semblance of order to any of it.

So WTF is going on with all of this? Has last year’s catch phrase ‘social search’ become this year’s, ‘real-time search’? Seems that way… so let me give you my perspective;

 

Types of Search Engines

Traditional search engine – has discovery elements and makes (algorithmic) valuations on indexation and ranking while employing spam detection/reduction methods.

Vertical search engine – deals with specific areas (such as video, blogs, topical) and generally has ranking mechanisms and the usual traditional search engines. We could even include such areas as ‘people search’ etc…

Social search engine – does the same as a traditional engine, but provides explicit user feedback tools as well as sharing (network/book marking etc...). I had a great chat with Bill Slawski last year on social search engines… well worth a read.

Recommendation engine – provides potential advice through implicit/explicit avenues (such as Google social profiling or Amazon recommendations etc..). To an extent even personalization and review sites could be here.

Word of Mouth engine – this would be approaches such as Twitter reliant tools or Ardvark, (we could also include book marking sites). In this instance it is your social connections providing answers to your queries. Danny has called this a ‘help engine’… either way.

You get the idea... a general categorization...


 

Now what about the current madness?

Real time search – these are really not usually much more than social aggregators that spit back results as they find them. Discovery is limited, ranking non-existent (in most cases) and spam detection, not really all that great.

Social Search – this tends to once more be aggregators of social mentions or meta-search engines with social voting elements. Any true social search engines lack the critical mass needed to truly be effective (to date).

These are more buzz words and not really adding much to what we’ve seen in the past. There really aren’t any large scale adoption of any of these which means we’re no closer to cracking the nut of combining social and search.

And so the question remains; what is a social search engine?

 

A framework for a social search engine

Personally I can certainly advocate what Google’s Marissa Mayer has said about it,

“We believe social search is any search aided by a social interaction or a social connection” – via Venture Beat

The recipe for such an approach can include;

  • 1 part traditional search engine (such as Google)
  • 1 part explicit user feedback (trusted sources, voting elements, annotations etc..)
  • 1 part social network profiling (and implicit feedback)
  • 1 part recommendation engine (based on trusted user data)

At the end of the day there really does need to be some form of explicit interactions from the end user that denotes which users/sources they want to have influencing their search results as purely social approaches tend to be lacking so far. To me most of the so-called ‘social search’ or ‘real time’ search engines tend to be little more than social regurgitations or buzz monitors.

A social search engine would want to identify like-minded user types or at least connect the dots within my social network to better produce relevant results. And they’d need to have a vast set of resources to draw from… not merely spitting back random Tweets (which many RT search apps do).

 

A grain of salt

And so my friends, let’s be cautious of what we see out there. It seems the catch phrases are more important than the actual realities. Some of these applications are simply social aggregators, others are meta-search with social elements and there are WOM connectors. None though have effectively created a truly social search engine.

There is nothing easy about nailing social/real-time search and to date no one has effectively done so to the extent that makes social search a reality. Sure, those drinking the social media kool-aid even go to the extent of saying Twitter search is the answer, to that I say it is little more than a search function on a social site…it is NOT a social, (nor real-time) search engine….

..but hey, I could be wrong… so feel free to speak up… (look for more on this soon, a real pet peeve)

 

 

Comments  

 
0 # Joe Hall 2009-07-09 10:58
Excellent work! I think its also important to point out the difference between an engine and well everything else. I think (correct me if I am wrong) the term "engine" refers mostly to a complex systematic process of indexing and ranking content/data to better aid Information Retrieval.

Some of the more recent search mechanisms introduce difference filters and ranking systems, but lack the overall technical and algorithmic features that would normally be associated with the term "engine". This would be where I would classify my search tool WhosTalkin.com. With WhosTalkin there is a systematic filter and ranking methodology for the initial results for each query, but it isn't based on quantitative metrics like linkback or PageRank that requires a sizeable index of data to calculate. Therefore, I would never call it an "engine".

Anyways, great post!
Reply | Reply with quote | Quote
 
 
0 # Bajki dla Dzieci 2009-07-14 06:25
Real time search could be huge. With the emphasis on "could". If it's not perfected soon, people will start dismissing it. So we need to either get it to work perfectly or get rid of it alltogether.
Reply | Reply with quote | Quote
 
 
0 # Nick Stamoulis 2009-07-15 09:08
I think social search is something that is going to be very powerful in the future. Right now it is in its infancy stage but evolving quickly.
Reply | Reply with quote | Quote
 
 
0 # Jerry 2009-12-16 06:06
Well, I appreciate Google's wonder wheel new feature that shows more related keywords based on your query.
Reply | Reply with quote | Quote
 

Add comment


Security code
Refresh

Search the Site

SEO Training

Tools of the Trade

Banner
Banner
Banner

On Twitter

Follow me on Twitter

Site Designed by Verve Developments.