How complicated can it be to make things simple? If you’re considering implementing a discovery service, chances are that question has passed through your mind. If you’ve survived discovery implementation (congratulations, you pioneers!), chances are that question might make you cringe, smile smugly, or both. If you’re in the middle of discovery implementation, chances are you are hating us right now for even asking such a stupid question. We feel your pain.

Our small academic library (six full-time librarians) decided to purchase and implement a discovery service in December 2013. One year later, we’ve survived implementation (December 2013–August 2014), launch (beta in June 2014, official in August 2014), and very nearly our first semester with a discovery service. Two librarians formed an ad hoc discovery working group to guide our library through implementation. As the two librarians who worked most closely with implementation (and all that it entails), we wanted to share some of our experiences in the hope that our pain and frustration might lead to less pain and frustration on your end! While we could get into excruciatingly detailed specifics, we felt it would be more beneficial and more accessible to a wider audience to instead focus on the general ideas and practices that worked for us during our discovery implementation and pepper in the specifics when they seem useful. With that in mind, we are trying to be as platform-neutral as possible. This is what we learned and how it might help you, no matter which platform you choose. 

Getting Started

 Once we decided to try out our discovery service, we had two questionnaires to fill out. One had to do with databases, one with our catalog. Through those questionnaires, our vendor was able to create a working system for us to begin developing.

Managing Content

Our discovery vendor contracts with content providers that deliver vendor indexing and abstract information for their content to the vendor. This information is massaged, incantations said over it, and dances done around it (or something like that) until it fits within the central index of our discovery tool. This central index is what our users will search. So a key factor shoppers for discovery must investigate is the correlation between the library’s holdings and the various discovery providers’ partnerships with content providers.

So, Is It Included or Not?

The process of seeking out content partners is ongoing, and we had to learn the difference between, for example, the press release announcing a partnership with Gale and the dribble of listserv announcements of a new Gale database being available to add to discovery. It seems the consummation of the relationship takes longer than just making an announcement. Maybe that’s when incantations and dances occur.)

But having a database included did not always mean the content was as discoverable as content from other vendors. As far as we understand, the primary reason for this is the quality of the originating vendor’s indexing. After all, the search engine is still just looking for words, and relevance is, in great part, dependent on the fields in which those words appear. So if the item is not well-indexed, it is not likely to be easily discoverable. It’s worth noting that this applies not only to the database records, but to any records included in discovery, such as your catalog records.