Google changes information management
The success of Google proves that if you manage content professionally, tremendous value is delivered.
Imagine you are very rich and you have a big house. One evening you have guests over. One of the guests asks you where the toilet (restroom) is. You give them instructions and off they go.
After a few turns they end up in the garden. They backtrack, apologise for their ignorance and ask you again for directions. You give them what you believe are even clearer instructions. They head off. They still end up in the garden. They are getting a bit desperate at this stage.
A neighbour is working in his garden and smiles knowingly. He shouts some instructions about how to find the toilet. The guest hurries back into your house and, following the neighbours’ instructions, quickly finds the toilet. Later that night, the guest mentions the kind neighbour that they talked to in the garden. “Oh, you met Mr Google, did you? We’d never find anything in this house without Mr Google.”
In many situations it is easier to find something on a website by using the public Google search engine than by using the website’s own search engine. This is an incredible situation. Think about it. Google, which indexes 12 billion pages, is doing a better job than, for example, a search engine that indexes 12,000.
This situation says less about the excellence of Google and more about the poor state of information management in many organizations. A great many organizations simply don’t professionally manage their information. They simply store it. All that is about to change.
It is about to change because Google is worth $150 billion. The stock market knows about value, particularly future value. The stock market is telling us that Google is one of the most valuable companies in the world. It believes Google is doing something that has great present-and particularly-future value.
What Google is doing is managing information in a way that makes it fast and easy to find what you need. Why is this important? Because, these days, before making key decisions, people are increasingly going to the Web to become better informed.
Information is becoming increasingly critical to the success of the organization. It’s just that most organizations have not recognized this, and have therefore not put proper information management structures in place. So what will the information manager of the future look like?
Their primary focus will not be on the information itself. It will not be on how to create, edit and publish content, although these are all important activities. No. The Web will radically change how we manage information, because for the first time we can systematically measure its impact.
Information management will focus on the result. It will ask the question, “What was the person trying to do and did our information help them do it easily and quickly?” They say that actions speak louder than words. Well, the future is about managing the actions that result from words.
Ian Ansdell says:
Added on May 6th, 2007 at 7:54 pm“A great many organizations simply don’t professionally manage their information. They simply store it.” A point well made, and pre-eminently applicable to Google itself, whose own website is impenetrable by means of navigation.
Cathy Jenkins says:
Added on May 6th, 2007 at 11:45 pmThe capitalist in me says “buy shares in google”. What I admire about google the most is I rarely have to use their advanced search. In fact, I’m more likely to use their main search rather than an organisations advanced search.
Mark Baartse says:
Added on May 7th, 2007 at 2:38 amA great newsletter which makes some great posts as always.
One thing I disagree with though, Google is so good because* they manage 12 billion pages. Google takes a holistic view of the web and interactions. With your 12,000 pages you can’t possibly take that broader view of your own website. I’m actually not convinced that a site search can be as good as Google. I’ve yet to find a site search which is, and I know sites that have dedicated significant resources to it. That leaves us with 2 options, either 100% of sites are doing a poor job (I would agree that 90% are, but not 100%), or that Google simply has an unfair advantage.
William Hughes says:
Added on May 7th, 2007 at 1:14 pmI say, so what if Google helps us find what we’re looking for. Does not Golden Pages (Yellow Pages) do the same for businesses? …help people find them?
Porscha says:
Added on May 7th, 2007 at 1:27 pmLots can be learnt about information management from blogging software. Search isn’t that difficult to configure or implement. Too many people get caught up in thinking of the “technology” driving it. Very few focus on writing great metadata. It can make a huge difference.
Organisations can be just as good as google in my opinion. People need to stop being lazy and focus on good metadata, being organized and managing information well. There are so many better practice guidelines out there and so few are learning from it.
Ben Anderson says:
Added on May 8th, 2007 at 2:08 pmI agree with Mark Baartse about the size of the collection helping Google’s relevance rankings. After all, what made Google different all those years ago when it first took the world by storm was the page rank algorithm. http://www.google.com/technology/
However, site search administrators also happen to have a massive advantage over Google - there is (or should be - see Peter Van Dijck’s item from 2005 http://poorbuthappy.com/ease/archives/2005/05/20/2683/enterprise-search-still-a-technology-conversation ) standard functionality called best bets/query boosting which allows the administrator to choose the “right” results for specific queries.
If you look at most search query logs you’ll see that a very small queries make up a disproportionate number of the total queries. Tanya Rabourn talks about it here, referring to work by Richard Wiggins, but her nice graph was missing last time I checked…. http://www.pixelcharmer.com/fieldnotes/2003/best-bets/
Marcia Morante says:
Added on May 11th, 2007 at 3:05 pmI agree that Google is a great search engine. I use it quite successfully alll the time for searching the billions of pages that they index.
But, it’s going too far to say that they are managing content. A good search algorithmn makes the odds better than people will find what they’re looking for, but it doesn’t help at all if the content that is being retrieved is out-of-date or inaccurate or poorly written. The solution to these problems is more complex and definitely not Google.
Tea says:
Added on May 13th, 2007 at 8:24 amThis is all fantastic info. on how great google has become. But, what about the huge issue of privatecy? It has been portrayed in the news that goodle has given out info. on a users search pattern to att= govn. How do you explain that? If a customer does not have privatecy and protection what good are you? How much are you payed anyway?
Gerry McGovern (blog author) says:
Added on May 13th, 2007 at 7:20 pmI agree with basically everything that’s said above. After re-reading my piece I think I didn’t do a good job in getting across a core point I wanted to make.
The point was not so much about Google itself, but that Google has shown that search is now a very valuable activity that is hugely valued by the marketplace. The stock market is saying that search has major value, yet most organizations I come across see search on their websites as something not worth putting much resources behind. Why is that?
Mark makes a good point about Google perhaps having an unfair advantage because of linking/ranking. But Ben’s point is definitely valid too. The top 200-400 searches usually represent a substantial quantity of overall search. If we can point people in the right direction for those searches, then that would be major progress.
Paul Biggins says:
Added on May 14th, 2007 at 9:16 amAll thought-provoking and valuable, especially ’storing information, not managing it’…guilty as charged! And the value of ’search’.
But surely, Google search has value over a site search because it searches millions of pages on millions of sites, and nothing else (with some exceptions) does that with comparable ease and reliability. A site search represents a failure of the navigation on that site? And using Google to search a site must be the ultimate failure of site navigation!
If information is logically and sensibly arranged, then a ‘user’ should be able to find it?
Unfortunately, that approach fails as other people’s logic and sense aren’t the same as an author’s, and even if one undertakes usability trials (always illuminating!), there are still people who will simply not understand and follow your menu and page structure. In a big intranet with multiple authors, enforcing the menu and page logic is difficult and unreliable, even with a rigourous CMS that enforces metadata rules. You’d think that if a page is worth publishing, then it would be worth indexing properly…but if you make it too onorous to publish a page, then people won’t bother.
An A to Z can be a useful alternative as a site cross-index, but it requires maintenance, and again, if the user is looking for the ‘wrong’ word (i.e. the word that ‘they’ use to describe the activity, rather than the word the author or indexer used) they will be disappointed and frustrated. That’s a failure on the part of the author/indexer to consider how the potential user will search for their information, but at least they tried.
So a site Search function serves to fill the gap left by inadequate navigation (despite our best efforts!) and incomplete A-Z indexing (ditto), but then we have equally disappointing and frustrating searches caused by poor use of meta tags, and illogical page indexing using - again - the ‘wrong’ or inappropriate words or phrases.
And the fact that a search may take the search terms out of context, and present inappropriate pages simply because they contain those unconnected terms.
Hardly a scientific approach to managing - and retrieving - information.
All we can do is our best…
Bob Johnson says:
Added on May 14th, 2007 at 8:33 pmWhen I review websites for colleges and universities, I always test the resident search feature for “scholarships” and “premed” and similar popular terms. Very few return meaningful information, although many of them report increased use of their “search” functions.
The point that nobody has a vested interest in the success of search is valid. Nobody seems to know what they are missing when search fails and so the idea that there might be positive “ROI” for time taken to improve search results has little weight.
In future work, I think I’ll routine ask for the top 10 percent of search queries submitted… and see how many people can answer the question and how long it takes them to do it. That indeed will tell me something useful to turn around in future reports.