The following are the steps involved in professionally managing search...


1. Identify the top 200-400 search terms on your website. These top 200-400 terms will usually represent a very significant percentage of search volume.
2. Identify the correct search result for each of these search terms.
3. Test each of these top search terms in your search engine.
4. Measure how often the correct result appears in the first three results. (Research shows that 60 percent of people will click on one of the first three results.)
5. Compile a success rate. You will then be able to say something like: Our search has a 45 percent success rate. Then you can plan for how you will increase that success rate.
This is just the first stage in professionally managing search. The next is to identify the search component in key organization tasks and to show how improved search improves task completion.
I once asked a senior manager in an insurance company why management in his organization didn't care about search. Their search was pretty awful but there were no plans to do anything about it.
His reply was that search was this vague thing that nobody was really responsible for. In his opinion, until improvement in search was directly linked to improvements within the key functions of the business, it would continue to be ignored.
He gave an example. He said an important business metric was the annual cost of managing existing customers. If improved search could be directly linked with reducing this cost, then that would surely get attention.
For certain organizations it may even be the case that search does not add enough value. There are many websites where navigation is much more important than search. For example, Gap.com doesn't have a search engine.
What is certainly true is that you don't need a poorly performing search engine. If you can't allocate the resources to manage search professionally, then it is a much better management decision to get rid of your search engine.
You may also find that as you improve the quality of task completion on your website, your search activity will decline. We once worked with an intranet where it was very difficult to find correct office and building location information.
Nobody really owned location information and, as a result, a huge variety of maps and directional information had been created over the years. Much of this information was out-of-date and/or poorly designed and confusing.
This location information was cleaned up. A new office locator was launched, which was much more accurate and easier to use. What was interesting to note was that as the office locator became more popular, the number of location-type searches in the general search engine showed a significant decline.
We are only at the beginning of the information revolution. Search, by definition, is a frustrating and unproductive activity. As we design better websites, and as people become more familiar with the Web, search activity will decline.
Search in itself is never the objective. You search for McDonald's not to find McDonald's, but because you're hungry and you want to eat a Big Mac and fries.