As we fast become a Google Nation, it's apparent that the fundamental tenets of Search (with a capital 'S'!) are the things that will turn your web site into all-powerful revenue-generating machine. And at the heart of this evolutionary process is the humble librarian.I realise that librarianship may not be the most glamorous art form, but, if you examine the development of the web, then you'll see that the skill of classifying and organising information is at the root of the most successful and profitable web applications -- Google, and ebay. Each of them is in the business of data classification. Google provides you with the web links that you need, based on its ability to decode your search enquiry and present a bunch of related web pages that it has already classified. sells you things you never thought you needed by correlating your purchase behaviour with other, pre-classified items that you might find interesting. And ebay enables you to bid on weird and wonderful things through its ability to correlate what you're looking for with the items that its sellers have classified. All of these businesses are built on the power of search. More specifically, they run their businesses on sophisticated classification schemes – or to use the technical term, metadata. To be a successful web business then is to act as a successful librarian. You will ensure that more people will find you and buy your stuff if you master the art of classification. Metadata and Keywords At this point, it's worth describing the two mechanical components of web librarianship: metadata and keywords. Metadata is basically information about information. Whilst this is abstract to describe, it's simple to put into practise. Consider the last book you stumbled across and bought on Amazon. Amazon maximises your buying 'serendipity' by applying a very successful metadata scheme to each of its products. The DVD of England's recent Ashes victory is described within Amazon's system by a variety of metadata fields. One will be 'sport'; a subset of which will be 'cricket.' Another may be 'Freddie Flintoff.' Another will be 'electronics'; a subset of which will be 'DVD,' and so on. So, as a cricket fan, when I look on Amazon for 'cricket' products, my search returns are likely to produce this very DVD. Further, when I click on the link and browse the description of this product I'm also likely to see a promotion for Freddie Flintoff's recent autobiography. And so I buy both of them and end up spending more cash then I'd initially bargained for. (But no matter, I'm a very happy customer!) Basically, what Amazon has done is to employ some smart librarians to sell me more stuff. That describes things at the backend of a web system (i.e., a bunch of skilled librarians ordering content into effective metadata schemes to make things easy to find and to ensure that relevant products are presented to you at opportune times). But as a web user, I look at this process through the opposite end of the lens. I'm concerned with 'keywords,' i.e., what are the most effective words I can use to find stuff I'm interested in (and quickly!)? Keywords can be thought of as metadata in reverse. They are the descriptions that I apply to information or products when I'm not quite so sure what I'm looking for. When I Google, I'm guessing ...but my guess is usually an educated one. When I tried to find information to support an important Jimmy Choo purchase decision this Xmas (I know ziltch about ladies shoes!), I typed "Jimmy Choo, shoes, prices, discounts, UK" into Google. I know the type of information I need (and at this point I should say that I'm buying for my wife), but I have no clue about the best place to find it. However, I love Google because it does most of the hard work for me. It basically scours the web and gives me back its pick of the best Jimmy Choo librarians that are out there. But note that Google is an advocate of librarians, not web sites. It won't always find the cheapest shoes for me, or even the best web sites that stock them. But it always tells me which web sites are best able to describe the fact that they sell them or talk about them. Google's main search engine service, like Amazon's, isn't subjective. It doesn't make value decisions based on concepts like 'best service.' What it does is make decisions on a set of pre-determined relevance metrics. It's basically an uber-librarian that classifies stuff for you. Therefore, your job in achieving a successful Google ranking is to match the way you describe your products and services (your metadata -- the information you use to describe your information) to the way in which web users are likely to search for them (i.e., their keywords). And this is why you need to act like a librarian. Innovation: Smart Librarians! To make a bold statement, this concept is driving every marketing and communications innovation that we see on the web today. Aside from the Google's, Amazon's and ebay's, others such as Flickr (acquired by Yahoo!) and Technorati (the Google of blogs) are taking the best classification practices of the web and creating new applications and services. Each of them is basing their businesses around the successful employment of metadata schemes. Flickr is a public photo gallery that enables users to apply their own metadata schemes to their pictures and share them with the world (it's free!). By doing so, Flickr is able to create clusters of images around any given classication tag. For example, I posted my holiday snaps from Barcelona on Flickr last year, and was able to see what other people had been doing (or were doing) in Barcelona at the same time. I tagged my pictures as "Barcelona", and when I viewed them I was able to browse through a bunch of other pictures that had been tagged with the same word. Even better, I could subscribe to this tag through my RSS reader, meaning that as new "Barcelona" pictures were posted, I got an alert immediately. I learned a lot about Barcelona this way (or at least I learned of the places that I really should have visited!). Technorati enables me to search the blogosphere by using a Google-like search interface. When I search for ' Barcelona , bars,' I get back a bunch of blog postings that have been categorised by their blog owners using these same terms. Even better (with a bit of playing around), I can also find the most influential blog posts about the subject -- like Google, Technorati orders its links by a number of relevance metrics such as "number of people who have read", "number of blogs also linking to this web page", etc. It's a great way of discovering information and opinions (try a Technorati search on your firm and you may get some interesting surprises -- "Disgruntled of Tunbridge Wells" may well have something to say that you haven't heard yet!). Both of these services have huge value in the commercial world. Flickr is a creative director's wet dream -- new images, ideas and creative associations on tap. Technorati is like a wiretap for PR professionals -- unfiltered opinions from real people on any given subject. Essentially they are both librarian services. But their driving principles should be applied to your business as soon as possible. You need to tell the world what you do, therefore you need to classify your information so that people can find you and make profitable correlations (i.e., 'because I buy this service, I'd also be interested in that service). What this entails is a) an understanding of how people search for you (their keywords) and b) the application of an effective metadata and keyword scheme to every piece of content, and every product and/or service that you publish on the web. Good librarianship: Talking the Customer's Language I won't dwell on how to construct a metadata scheme -- that belongs to a whole other book; but in brief, to improve your external search engine rankings and to improve cross-selling of products or information within your site, you should use a content management system that enables you to publish information about your information. When this is done successfully, your selling or marketing 'serendipity' will increase radically. Like Amazon, you ought to be able to push new content (or services, or products) to people who, when browsing your site, are interested in related content (or services, or products). And Google will show you new levels of respect -- the uber-librarian will appreciate your librarianship. In terms of keywords, you need to take on board the quirks of how people expect to find you and apply this to the way you describe yourself -- both at the metadata level and in the body of your web pages. This is a great exercise in self discovery. When I walk into a good library and ask a librarian to help me look for a book on pie and mash shops, the librarian understands me and points me in the right direction. What s/he doesn't say to me is "Look, you should be thinking about low end fast food franchises in the business section ...go away, think about that and ask me again later." To improve the way in which Google classifies you, you need to look at your web stats and your Google-rated competitors, check out the words that people use to search for you/them, and then appropriate this language for yourself. A good librarian knows that the best way to file a book is to understand how a reader would classify it themselves, but we in the world of marketing have neglected the importance of talking the customer's language. If you want to maximise your chances of being found on the web, and optimise the way in which you can sell and cross-sell, then you need to think like a librarian. (See Related Article: The Two Fundamental Skills of Web Writing) ---- Roger Warner is the Director of Squiz UK Ltd., a commercial open source Web CMS solutions vendor.