Stop Thinking So Much Like a Damn Librarian (or how I started liking discovery layers)


My library just implemented a discovery layer – Primo from Ex Libris (branded as Search@UW since most campuses in the University of Wisconsin System are using it) – to combine catalog records, plus articles and other resources from our databases. Frankly, I wasn’t excited about it at first. It had nothing to do with the product itself. It just seemed like we were getting something that we weren’t asking for.

As an instruction librarian, I approach things from a pedagogical standpoint: How will students use it? What will it do for them? I ask a lot of “what ifs.”

While the discovery layer was being tested, I happened to be teaching a semester-long senior-level information science class. As one of our projects, we did some usability testing on Primo. Guess what? The students loved the discovery layer.

We compared finding information it in versus searching the online catalog and databases separately. The discovery layer won hands down in terms of speed and ease of use. My biggest worry: “Were students finding relevant information?” was assuaged.

And this is when I had to teach myself to STOP THINKING SO MUCH LIKE A DAMN LIBRARIAN!

The discovery layer makes perfect sense to students:

  • A seamless experience for finding information.
  • A simple search interface, a la Google.
  • Start with a broad search and then narrow it to particular types of items (books, articles, etc.).
  • An element of exploration.

Screen Shot 2013-09-10 at 12.05.20 PM

Don’t Box Me In!

Silo-ing information – which libraries are REALLY good at (…and which is NOT a compliment by the way) – does not make sense to students. A catalog to search for books? The databases to search for articles? It’s a holdover from the olden days of libraries: “real” card catalogs with endless drawers of records to locate books in the stacks, and volume after volume of print indexes to find articles in a periodical.

While we were testing the discovery layer, there were a few things I didn’t like. Case in point: In our “old” catalog, I had a drop down box to limit my searching to our Reference Collection. The discovery layer did not have that option from the main menu. But there I went again: thinking like a librarian! Stop. Examine what your users need to do with the tool at hand. Do my students need an option on the main menu to search for reference books? The answer is a resounding NO! Searching for reference books is simply NOT a priority. It’s OK to re-evaluate those sacred cows.

At the same time, I recognize that if you’re doing heavy duty research in a particular subject area, then a subject-specific database is your best bet. So, I created a guide for students: What Tool Should I Use to Find Information? to direct them to the appropriate tools.

Permanent Beta is OK

We rolled out our discovery layer in a not-quite-perfected state. Each class and group I’ve shown it to has loved it. Librarians get too pre-occupied with perfecting everything before rolling it out. This causes delays for your users and dwindling interest as well. Seize your patrons’ needs and desires and then deliver in a timely manner. Get feedback, re-tool, adapt, and grow from there.

The User is #1

So I came around on the discovery layer. While I always like to think I have my patrons’ needs in mind, you really need to step out of those librarian loafers and examine them. There are services that patrons would like, and probably some that they couldn’t even imagine. Harness this information and then deliver it for your patrons.

22 thoughts on “Stop Thinking So Much Like a Damn Librarian (or how I started liking discovery layers)

  1. Great post and especially great guide on which tool to use! It’s so hard to step outside the librarian worldview when evaluating services, products, marketing efforts, etc. and I applaud your effort. I agree that if students are still meeting their information needs using the library, then we are meeting our goal as academic librarians.

  2. I was studying my library course and working as a librarian at a community college when my university introduced a discovery layer (also Primo), and from day one I LOVED it. From the student perspective it’s so much easier – you start big and narrow down results through the options on the side and suddenly I was using information from more than one database (I had the hang of EBSCO so I tended to just use that).

    Sadly we don’t have plans to introduce one at work anytime soon (and if you think databases are overwhelming for undergraduates, imagine trying to grasp it as a mature student who wasn’t particularly academic in high school) but we’re using search forms and customised links where possible in our libguides to try to streamline the search process

  3. Pingback: Stop Thinking So Much Like a Damn Librarian (or...

  4. Can you share what your Primo consists of? I ask this because while the search layer is fairly common across all implementations there are a lot of options re content that (I am hypothesizing) effect how useful it is – i.e., whether relevant information comes to the top of the relevancy ranking.

    We’re taking a look at logs of hundreds/thousands of searches a day (so the ones done by library users – not examples made up by librarians for usability testing) and the performance just isn’t there. In many many cases, better results can be seen in EBSCO academic (FYI – not EDS, just the plain vanilla database). So – very different than what it seems you are finding. But, I’m cognizant that Primo isn’t Primo isn’t Primo – content variations could be having a significant impact … especially with all of the documented issues with phrase and author name searching.

    BTW, no argument that users like the interface in usability testing. In fact, using the convenient pull-down for the “scope to the OPAC” – it is my favorite OPAC interface. If only our consortia holdings were in there too I could never use webVoyage or vuFind again! 🙂

    • Hi Lisa–most of our campuses in the Univ. of Wisconsin are/or will be using Primo. It includes books & media at each campus, books & media at all of the other campuses, most of the content in our databases, and digital collections (e.g., Hathi Trust). By default, materials from each of these scopes show up in a general search, and then users can limit to a specific scope from there (e.g., just books & media at our campus). Our tech people are able to play with relevance rankings and can push specific kinds of content to the “top.”

      For intro students, we’ve only gotten positive feedback–but we just rolled it out 3 weeks ago–we need to do a more in-depth study. The big downside with Primo (at least for us) is that it’s not the best when finding known physical items (a specific book), so that’s the big stumbling block we’re dealing with now. For advanced students, we also want to be mindful of over-relying on Primo when a subject-specific databases is probably a better source.

      For finding articles, students have been ecstatic about it. Although we have mostly EBSCO databases and Primo (Ex Libris) does not play well with EBSCO (or is it vice versa!?), 90% of our database content is indexed elsewhere, so the results for students appear to be pretty seamless.

      It’s a work in progress and frankly I didn’t like that it was rolled out the week before school started, but we had a university system timeline to follow.

      • Thanks. It will be interesting to see how your experience develops. We’ve had Primo for about a year so I’m definitely aware of the known item searching problems as well as the EBSCO. When we did a pilot analysis of user searches and the results comparatively in Primo and Academic Search (EBSCO) – for me, I’d definitely recommend the second to a student wanting a better chance of meeting their professor’s expectations for quality sources (plus easier access from citation/abstract to full-text). Will be interested in your results as you do your studies!

  5. One thing I like about a discovery tool is that sometimes students are looking for information and think they want articles because they don’t want to do too much reading, but the type of information they need is more likely to be found in a book. If this is the case, their original search in the discovery tool will give them lots of book results and few article results and the student can stop and think about what they really need.

    In our experience there have been so many glitches with Primo that we frequently tell students not to use it. We will be switching to the EBSCO discovery tool sometime next year.

    Yes, students love it, but as our own “internal customers”, we should be able to have a product that librarians are also happy with.

    • I agree with you that librarians need to be happy too. My biggest concern right now are problems searching for known physical materials (e.g., a particular book title), whereas just the general “discovery” search seems to be bringing back useful items for students.

  6. Hi Joe – I agree, the tool you designed for students is very nice, very clear. I’m interested in the comment you made
    …”if you’re doing heavy duty research in a particular subject area, then a subject-specific database is your best bet.” I think a lot of us instinctively agree, but do you have any specific reasons, or are you aware of any research done that backs that up? Often times the subject specific databases are at risk of cancellation to fund expensive multidisciplinary resources – so some kind of specific rationale for retention of topic specific resources might be helful

    • Hi Keith…thanks for the comment. I tend to focus upper-level students to a particular database (or set of databases) if they need to do more focused searching. For example, a 300-400 level Psychology student needs to be using the PsycINFO database as opposed to the discovery layer–or at least needs to get into PsycINFO after exploring around in the discovery layer. Of course, if we cancelled all our subject-specific databases, then that content would no longer appear in our discovery layer! 🙂

  7. Wow, am I glad I did my discount double-check and looked at the chart before commenting. The chart provides nuance and shading absent from the blog post itself and changes one’s perception of the post. I, too, would like permission to borrow/modify the chart for my campus. It is a fine piece of work.
    I do wonder, though, why your own catalog is not linked, nor Google Books.

    Given all that is necessary to refine a search down to what used to be called a working bibliography, I have never been on board with the “beauty of the Google-like single box search,” and the ubiquity of that phrase makes me wonder whether these vendors are developing web-scale discovery tools (WSDs) because they truly are a “better mousetrap” our because Google’s market share has them worried. And I also wonder, if students had had continuous incremental instruction in information literacy — the use of catalogs, databases, and other information sources and tools — in k-12, whether they would flock to Google. Or is Google just the devil they know, and it fills the void? (And this still begs the question of whether students truly know *how* to use Google, or just use it.)

    With your permission, and with an eye to the length of the other comments, i will split my comment into two parts.

    • Hi Harry–thanks for the comment. Yes, please feel free to use/modify the chart. Our catalog IS the discovery layer. The records are integrated into the discovery layer. Users can narrow just to our library’s holdings or the holdings of the other campuses in the university system. I also agree: incremental instruction would be wonderful, but as school librarians disappear and with schools more likely to just “teach to the test” – I don’t see it happening soon, unfortunately.

      • Thanks for permission to use the chart, Joe; I plan to forward the link to my liaison faculty — my University’s English Department — even before we try to make any changes. I think it is that useful.

        I neglected to mention in my earlier post that like you, we had our discovery service “gifted” to us without our asking for it. I am afraid you are right about the school librarians. I would have hoped they would be all over early, incremental ILI if for no other reason than the job security it would provide. But I don’t see it happening. Both Alabama and Georgia for more than a decade have had database packages provided by their legislatures,
        but the informal queries I make with the students I deal with indicate that most have never heard of them and none of the students have been taught them. I think this failure is more on the superintendents and boards of education than it is the teachers or librarians.

  8. Part the second.

    My library’s experience with WSD largely mirrors yours, but the conclusions of our public services staff to some degee are quite different. Our WSD (not Primo) was rolled out in January, three weeks *after* classes had begun, and it was far from stable. It is better now, but still not to the point of being the first search option (mainly because of results quantity, relevance, and ranking); it is just one more club in the bag. And it has no monopoly in making sense to students. Your four bulleted points are just as doable in a silo’d system or even a catalog and single database search strategy.

    (I would LOVE to have a Reference included in our catalog’s drop-down. The floors that house my library’s humanities collection have extensive print reference holdings, and one of our history assignments, used in dozens of sections, has print reference as a central component. And it still is hard to beat reference for getting background and overview on a topic to help with
    initial focus. That is what specialized reference is for.)

    I agree that for tightly-focused research a LibGuide likely is the best way to go.
    But for broad-based underclassman research a WSD is the last place I’d go: too many results and a ranking system that is inscrutable. If WSDs offer us the best of Google, IMO they also offer us the worst.

    Unlike you, I have had far better success with known single-item searches than with a broad bread-on-the-waters search. Searching for things that likely have received little ink, like academic journal articles on international films or obscure battles — in my opinion this is what WSD does best. Huge results numbers and ranking issues are minimized.

    I think we should not be shackled by our professional training and mindsets, but we should not abandon them. If we stop thinking like librarians then we defraud students of our experience and expertise — the things that enabled you to create that fine chart. We just need to know how much influence to give these things in a given situation. Would you want your attorney to stop thinking like a lawyer?

    Thanks for giving me the space to opine.

  9. Pingback: Cheating on This Blog | Digital Carrie

  10. Pingback: Lonely, Lonely Links | Catherine Odson

Comments are closed.