• The Search Blog

    Klout.com now also considers Facebook for influencer scores


    Klout.com is a relatively new site which calculates a ‘Klout’ score to represent someone’s level of influence on social networks. 

    The score is calculated using a number of factors such as number of followers and percentage of posts retweeted.  Up until now, Twitter has been the only social network which Klout has used to generate scores, however I just got an email from them notifying me that my Facebook account (which I have previously connected to Klout) is also now used to contribute towards my Klout Score…


    Facebook and Twitter social networks are very different

    This is very interesting, because most people have a completely different network of ‘friends’ on Facebook to what they do on Twitter.  The more personally and socially connected nature of Facebook (in relation to Twitter) suggests that there will be a significant number of people who may not have influence on Twitter, but DO have significance influence on Facebook through their 1st, 2nd and even 3rd degree friend connections.

    Fast Company recently reported that the amount of links shared on Facebook is WAY higher than Twitter, however the click through rate on Twitter is WAY higher than Facebook...

    Whilst users may not be ‘clicking through’ as much from Facebook, I suspect that the ‘like’ rate and ‘view’ rate (for videos/photos) is WAY higher again than Twitter click through rates.  How do the two types of influence compare?

    Facebook is more social, but is it just as influential as Twitter?

    Typically Tweeters follow each other for business, or another shared areas of interest.  Facebook on the other hand represents networks mostly formed by social connections. 

    If I share information about the latest Search Engine Optimisation technique with my Facebook network, most of my ‘friends’ will ignore it, or ask “what the heck I’m talking about?!”.  If I tweet the same message, my numerous followers interested in the web will be keen to share it with their respective network of web and search professionals. 

    This does not mean that I am less influential on Facebook.  In fact, I have a large collection of friends who live in my local town on my Facebook friends list, so my links, comments, videos and photos about local businesses and events can be very influential.

    So what does Facebook Klout look like?

    According to Klout.com, my Facebook profile does not seem to have had a huge effect on my Klout score…


    Perhaps this is because it considers my Twitter profile WAY more influential and therefore dwarfing the effect of my Facebook activity.  Or perhaps my primarily social activities on Facebook make me only an influencer within my small group of local and work related ‘friends’, so my Facebook Klout really does not pack much of a punch at all.

    It is interesting to see how it calculates some of the values based on a combination of the Twitter variables (in green) and Facebook (in blue).  Here is the breakdown for my ‘true reach’…


    So 70% of my Facebook posts are ‘liked’! That seems pretty good to me!  Much better than the 4% retweet rate from Twitter.  I guess the question here is what impact is reaching someone who has ‘liked’ my post, compared to having someone click a link I share on Twitter.  I quite often ‘like’ my friend’s facebook updates which read ‘Carrie had Sugarpuffs for breakfast’, but it doesn’t mean she influenced me…or does it?

    There are also amplification and network variables which measure the likelihood that your content will be acted on, and the influence level of your engaged audience respectively.

    What do they need to make it complete? And even then, would I trust it?

    imageWhilst there is a lot of 2 way interactions on Twitter, it is more common for people to ‘broadcast’ information and links, with the intention of getting them shared virally.  Facebook has less of a broadcast dynamic, and more of a multidimensional socially connected set of networks encouraging 1:1, 1:many and many:1 relationships focussed around friendships, relationships, hobbies and even a bit of business.

    Facebook is without a doubt the number 1 store of social connections on the web, however the professional nature of LinkedIn makes it a valuable data store to indicate how people are connected in business.  If Klout could pull in LinkedIn data and effectively merge it with FB and Twitter information, they would seem to have a relatively complete view of social interaction and influence online.

    Even if they do pull in data from all three, the secret is still in the sauce.  I would love to see the algorithms Klout are using to calculate influence (I didn’t want to use the word ‘klout’ twice in that sentence), and would loved to have been in the discussions where they were trying to figure out how to merge Facebook with Twitter data.  In fact, without knowing how they do it, I’m not sure that I completely trust the results! Now there’s a dilemma.

    It’s all about topics

    Whilst it’s very cool to look at your own analysis and see how Klout classifies you, the question is how useful actually is this data?

    Of course knowing that I am influential (or not) on Twitter is not useful to any business unless they know what topics I am influential on.  Klout is capable for extracting out topics which relate to each users Klout score, here is mine (seems pretty accurate)…


    Right now, they do not have a function built in to the site which allows you to search for a topic and find top influencers.  This information is available through their API however, and some companies have already started to use it.  E.g. www.liveintent.com which enables publishers to connect with influential audiences identified using data from Klout…



    Author: Chris Moore is a program manager from Microsoft working on Search Engine Optimisation.  Follow him on Twitter

  • The Search Blog

    How to find things online without searching (Part 3 of 5)


    imageThis mini series of posts focuses on methods of finding information online without typing a single search phrases in to Bing, Google or any other search engine.  Part 1 focussed on check in services and part 2 discussed social voting mechanisms

    In this post I will highlight search technologies which allow people to search the web using photographs and other visuals…

    Image search without keywords

    This post series is all about how to find information on the web without typing search words, so I would be cheating if I discussed traditional image search (such as the awesome functionality provided by Bing). 

    However, one particular piece of technology which can be found in Bing image search does warrant inclusion in this blog post – that of ‘Similar images’, which utilises image recognition technology to find images which are similar to those you have selected.  So looking at these image search results for ‘Counting Crows’…


    I can see a cool looking photo of the band casually standing around, which I would like to find similar images to…


    With one click on ‘similar images’ Bing will intelligently match similar photos (but not necessarily the same) and present a page full of casual looking photos…


    Neither Bing or Google let you search using the image as a starting point (yet), however http://www.tineye.com does let you upload a photo, or provide a URL to a photo on the web and then search for similar photos – no search words required!  This is a great way to search the web for sites who have used your photographs/diagrams without asking for permission Smile

    Facial search – fantastic or scary?

    image[24]As image/facial recognition advances on the web, the power of image search becomes more and more exciting! (or perhaps worrying).  Face.com is a great example of facial recognition working well to automatically scan your facebook photo albums, and those of your friends and automatically tag you and your friends in the photos.

    If you haven’t yet tried it, I guarantee you will be impressed with the accuracy!

    Philanderers, weekend drunks and secrets agents…beware!  Whilst you may be efficient at untagging yourself from embarrassing Facebook photos and checking for undesirable mentions of your name within search results, it’s only a matter of time before every single photo/video of you will be discoverable with a couple of clicks – whether you are tagged in them or not.  In fact, I think the technology is already there – perhaps there are privacy issues preventing such a service…I will read more about it when I get the chance.

    Barcode scanners

    imageA few weeks ago I was shopping in a local electronics retailer for some travel speakers.  After being pestered by the shop assistant who was trying to demo and sell me the set of speakers which would get him most commission, I finally made up my mind and asked him to go and fetch a Logitech speakers for me which were priced at £105. 

    When he proudly walked back with the package, I pointed my iPhone at the side of the box and ‘scanned’ it using the RedLaser app.  This app uses image analysis to detect bar codes using the iPhone camera, and then automatically searches for products online which match the scanned barcode…

    photo (7) photo (6)

    As you can see, the product was available much cheaper than £105 online.  I show the shop assistant the screen of my phone, which prompted him to “go speak to his manager” and resulted in a discounted price of £80 for the speakers Smile The ‘search’ took just 3 or 4 seconds and I didn’t have to type a single search word.

    By the way, this functionality is also available free of charge in the latest Bing iPhone app, however unfortunately that app is not officially launched in the UK (where I am based).

    Connecting the pieces together

    Of course, recognising a barcode, converting it to a product code and then searching for that product code online is not making use of image search functionality, but instead just relatively simple image recognition followed by a keyword search.

    Google Goggles is a first effort to connect the pieces together in a way which allows users to search the web by simply pointing their phone’s camera at an image.  This video explains how Google Goggles send images from the phone, runs them through image search processors online and then returns relevant web results…

    Using Google Goggles still requires the user to make a ‘search’, even if they do not need to type a keyword.  As we move towards the future, we will see technology simply present images, related facts or other pieces of contextual information to users at the point of need.  Bing.com maps provides a great example of seamlessly ‘searching’ for and blending Flickr photos on top of street side imagery, using advanced image recognition and manipulation to present contextually relevant and valuable information without the user having to even think about the word ‘search’…


    Finally check out this video demo of Bing.com maps and take a few moments to imagine the possibilities of these types of contextually aware, automatic, unobtrusive visual searches which enhance any browsing experience by providing easy and instant access to any relevant data on the web…

    Author: Chris Moore is a program manager from Microsoft working on Search Engine Optimisation.  Follow him on Twitter

  • The Search Blog

    Bing + IE9 = the future of search


    This short preview video shows how the soon to be released HTML5 features of Bing, along with the superfast hardware accelerated IE9 rendering produces a truly unique search experience, moving us a giant step away from the traditional 10 blue links we have had for so long…

    Exciting Stuff!

    Author: Chris Moore is a program manager from Microsoft working on Search Engine Optimisation.  Follow him on Twitter

Page 5 of 23 (67 items) «34567»