Seven ways to improve enterprise search

Bad search isn’t just the fault of vendors: Here’s how to make search more effective

Bad enterprise search is a common problem in businesses, but it’s not always the vendor’s fault. Here are seven ways customers can maximize the effectiveness of enterprise search.

Bad enterprise search is a common problem in businesses – but it’s not always the vendor’s fault. Customers pay big bucks for enterprise search software but they often don’t have a strategy for deploying the program effectively.

“Many companies end up disappointed and frustrated with high-priced products that failed to live up to expectations,” Forrester analyst Matthew Brown writes in research released this week. “Yet, it’s surprising how little effort these companies typically put into creating a compelling search experience for employees – especially given the potential productivity gains effective search implies.”

Business search offers complications that don’t really affect consumer search sites such as Google, Brown notes. Business content lacks context, links and often has little text. Workers must store and access information securely, and need information of the highest quality, not information that’s the most popular.

Brown offers seven tips for enterprises looking to maximize the effectiveness of search. Here’s a summary:

1. Define objectives. Identify who is searching, what categories of information they’re looking for, and what they need to do with that information. Since you can’t account for every type of information workers need, define broad worker roles, like sales professionals and market researchers, and their most common search scenarios.

2. Set the scope. Ask yourself whether you’re searching a single repository or multiple repositories, which content must be accessed by users, which back-end repositories are required for a given search scope, and what are the unique characteristics and contents of each repository.

3. Find a good method for enriching content with descriptive text and metadata. Manual attribution can work with small sets of authoritative information, but may not be worth the effort for large amounts of data that changes rapidly, Brown writes.

4. Set your requirements, then list products and vendors to consider. Prices have fallen over the past decade and the list of vendors is long, including Autonomy, Endeca Technologies, Fast, Google Enterprise, Recommind and Vivisimo. “The key is not getting enamored with irrelevant features, but instead focusing on products that adequately meet the organizations requirements over a specified time period,” Brown writes.

5. Develop a taxonomy that organizes logical types of searches. These might include searches for people, products or customers. “It’s now possible to search and return results for virtually any logical item in an enterprise -- like orders, customers, products and places,” Brown writes. “But [information and knowledge management] teams must tap into sources like enterprise data warehouses, ERP systems, order histories, and others to create a full picture of the item that’s searched.”

6. Plan for a search experience relevant to multiple types of users. Google became a behemoth by offering a simple-to-use search box, but enterprise search often requires a more advanced interface. For example, if an employee searches for “electrical cables,” a finely-tuned enterprise search product might list cables organized by gauge, casing materials, insulation, color and length, Brown says.

7. Implement, monitor and improve. Large search projects can take three to six months, but it’s worth it. The implementation should involve at least a project manager, application architect, search product evangelist, a representative set of end users and partial involvement of content repository owners, Brown writes. A business analyst should be in charge of search reporting.

“A monthly reporting ritual that surfaces most frequent searches performed, null searches, and overall usage of the search function can help you troubleshoot existing implementations and drive future decisions on how to enhance the search experience over time,” Brown writes.

Learn more about this topic

IBM tackles “dirtier” metadata issues

06/27/07

Microsoft, Google trade barbs on enterprise search

04/12/07

You are wasting time. Find out why

01/23/07

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.

Copyright © 2007 IDG Communications, Inc.