Spending whole days listening to talks on search has highlighted to me that there are really two types of search:

• Broad and shallow: this is the typical intranet or website search, which supports a large number of users conducting relatively simple search tasks.
• Narrow and deep: this is the search for specialist groups, whether they be staff in the intelligence community or pharmaceutical researchers.

The “broad and shallow” search is the 80/20 solution: with relatively modest technology it meets the day-to-day needs of the majority of staff in most organisations. For this search, the focus should be on simplicity, usability, and ongoing tuning. This search is hugely important, but can be hard to measure in terms of direct ROI.

In comparison, the “narrow and deep” search is the 20/80 rule. While it only targets a fairly small number of specialist users, the benefits gained can be huge, and
often directly measured. This search focuses on functionality and capability (also hopefully usability).

The challenge is that these two cases are not well articulated in the marketplace. Customers need to be clear about what kind of search they are looking for,
and select technologies accordingly.

Similarly, vendors need to better target their  materials and marketing (and ultimately their products) to these one of these two cases. This is particularly important in the “enterprise” space, where organisation-wide is often confused with
powerful (but complex) search solutions.

Both of these situations are valid, and very important. In the work that I do, my focus is more on the first (broad and shallow) situation, with the aim of improving the search experience for the largest number of users. In contrast, a number of the search vendors
are validly focusing on the latter (narrow and deep), recognising that this is where they can really differentiate themselves.

In summary, language is important. If nothing else, it gives us some common ground to understand the problems that we are trying to solve, and to identify how best to design a solution.

PS. like any black-and-white categorisation, there are exceptions to the rule. Online yellow pages sites, for example, are both broad and deep. This is what makes them so hard to get right…