Skip to main content

Experience Management

How to Pick Your Enterprise Search Bracket

photo by slgckgc on flickr

photo by slgckgc on flickr


March Madness is in full swing. Completing your bracket probably involved some tough decisions. Did you pick the favorites over the underdogs? Big schools vs. small? How about taking into account injuries or the age of the players?
Evaluating enterprise search platforms can be complicated, too. There are numerous products in the marketplace with a range of architectures and features, and each customer has specific requirements to consider. The result is a complex decision tree that will hopefully result in picking the best solution.
While this post is not about addressing specific vendors or products, it will describe the major factors to consider when evaluating your enterprise search bracket. What should you consider when selecting one product over another?
1. Architecture: On-Premise or Cloud? Software or Appliance? Enterprise Search has traditionally been an on-premise solution (like most enterprise software). But as software and content move to the cloud, so do search engines. If you are only considering on-premise solutions, you must weigh the pros and cons of software vs. appliance-based platforms. You must decide what architecture makes sense for your situation. Consider the following factors:

  • Where does the majority of your content live — on-premise or in cloud-based repositories like Salesforce.com or Box.com or Google Drive, etc.?
  • Do you have adequate upstream bandwidth to index on-premise content with a cloud-based search engine?
  • Does your IT department continue to support on-premise hardware and software, or would they prefer a hosted solution?
  • Do you have the resources to configure, maintain, or patch enterprise software, or would an appliance-model be more suitable?
  • Does your industry regulate the transmission, storage or use of data outside of your network? Does the cloud search engine you are considering provide adequate certifications or audits?


2. Basic Search Features: All birds have wings, and all search engines have features. Some are extremely common, like spell checking or faceted navigation or result biasing and manipulation. Others are emerging, like full-document previews or optical character recognition or built-in reports and analytics. I have seen my fair share of RFP’s that list out every conceivable search engine feature — but to what end? To keep everyone sane, I would recommend focusing on a short list of must-have features, and let the rest come along for the ride. Don’t waste time sweating over features that only a small fraction of your audience will ever use. Avoid analysis paralysis.
3. Connectors, Integration and User Interface: Some enterprise search platforms include dozens of pre-built connectors to popular applications and content sources, and some do not. Do you have the skills or budget to build custom connectors or buy third-party ones? If not, you need to make sure that the search platform includes the connectors you need. If you plan to integrate the search engine with a single-sign-on tool or wrap the search results inside of a custom application, make sure that it has the necessary APIs or is compatible with the other applications you use.
4. Analytics, Machine-Learning, and Intelligence: As enterprise search tools advance, they are starting to include features once part of business intelligence or content analytics systems. Is it important for you to be able to visualize your search traffic in various ways or build complex entity recognition rules? What about machine-learning algorithms that improve search relevancy over time based on user behavior? Many search engines are beginning to include these features and the results seem positive, if you spend the time to properly set them up. If you would rather have a fire-and-forget solution, there are platforms that provide machine-learning and intelligence, but with less knobs and dials. Consider how much time you can devote to tuning your search engine or adjusting the intelligent feedback modules.
5. Vendor Profile and Roadmap – This factor might be the real elephant in the room. As much as you might want to ignore it, you must consider the size, makeup, and character of the vendor. Can you tolerate a startup that only has email-based support, or do you need a global company with a 24/7 support organization. Is the vendor large enough that you don’t have to worry about the product being developed and supported for many years? Does the platform have a roadmap that is enticing and moving in the same direction as your needs?
Stay tuned for future articles, where I plan to dig deeper into some of these features, particularly newer or more novel ones. You can subscribe for weekly updates using the form below.
 

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Chad Johnson

Chad is a Principal of Search and Knowledge Discovery at Perficient. He was previously the Director of Perficient's national Google for Work practice.

More from this Author

Follow Us