I have worked with and researched more than my share of web applications that track library reference statistics. Below are my observations, which are not meant to be endorsements of any particular software package but merely a collection of possible options.
If anything below is wrong, please email me so I can correct it. I would also love to hear from librarians who use other options which I have not listed.
The “Free” Options
I’ve never used this but it appears to be very common in the library community. The biggest issue might be maintenance: while bugs have been filed in Google Code, there hasn’t been a line of code committed since 2009. It runs on top of Pear::DB which is stable but has been deprecated.
I was curious about the possibility of setting up a statistics tracking system using Google Forms so I made a simple template that collects mode, location (could be set to different branches), READ rating, & an open-ended note. Basically, you record data in a form using either your institution’s Google Apps account or another account set up for that purpose, which is then available as raw data in a Google Spreadsheet. Google Forms come with a limited, built-in summary but could also be analyzed in the spreadsheet or exported to csv.
Sunshine Carter and Thomas Ambrosi have written a nice article in Computers in Libraries which covers quickly building a statistics tracker using Google Forms.
All major CMS products, such as Drupal or Wordpress, contain form-building extensions that make it easy to hook into the CMS’s database. Constructing a form for reference statistics might be no more complicated than doing so in Google Forms. The advantage is that you can achieve a level of customization according to your own skills. Most commercial products still lack desirable features—such as form fields that conditionally reveal themselves based upon previous input (e.g. a field for “email address” that appears only when mode of communication is set to email). With a little scripting know-how it’s not tough to add features you need; if your library doesn’t have the skills in house, you could pay a web developer a one-time fee to build your reference tracking dream.
In case it isn’t obvious, free is in quotes for a reason. You have to have someone with the skills to configure and maintain the tracking system. The time commitment is typically greater than with a paid option. But there’s also incalculable value in owning your own data and having complete control over the system that collects it, too.
Paid (Subscription) Options
This list is alphabetical because I don’t want to prejudice one product over another. Please do not take any price quote as gospel. Vendors are typically not forthcoming with this information and the only way to obtain a firm figure is to ask for a quote for your particular institution.
I worked with Desk Tracker a lot in a previous position. In its defense, it’s very customizable and setting up new forms doesn’t require any specialized knowledge, so you can track just about anything. You can create open-ended text areas, drop-down menus, and radio button lists. But the data it outputs, both in terms of built-in graphs and raw csv files, are hard to use, especially for custom fields. Be prepared to invest a fair amount of time in the analysis phase if you use Desk Tracker. It is not the cheapest option.
Update May, 2016: a representative from Desk Tracker contacted me with claims that they’ve greatly improved their reporting and charting features. I used the software five years ago, so I don’t doubt it’s changed and I would recommend obtaining a trial rather than going on what I wrote above.
Allows you to create custom ticksheets to record reference interactions. It does seem to be closest to the paper ticksheet workflow, in that one can batch enter a shift’s worth of data after the fact. Comes with a built-in reporting module and csv exports. Pricing isn’t clear to me but a leaflet they sent me notes “available for small libraries for as little as $400/yr” (as of 6/2012). DeskStats integrates with Altarama’s RefTracker module, so requests handled in that system are automatically recorded.
Made by the same team who created LibStats. A clean, user-friendly product but notably less robust than other options. Your data types are basically all multiple-choice or open text fields without much customization, though the ability to tag questions is interesting. You can view question-answer pairs. With a free tier that lacks only unlimited reports, and then only $10/mo per library branch (as of 11/2012) beyond that, Gimlet is a very affordable option.
LibAnalytics / Reference Analytics
By SpringShare, makers of LibGuides. Reference Analytics is a module of the LibAnswers package; LibAnalytics is independent and offers more powerful customization options. In LibAnswers, pop-ups prompt you to enter data after answering a chat or FAQ inquiry. You can save question/answer pairs and review them, which makes creating an FAQ or identifying common questions easier. The default charts are nice and offer many filters. One built-in graph is traffic by weekday hour—i.e. precisely what you need to make informed staffing decisions—a graph I’ve spent hours making in other systems. As of November of 2012, I was given a price quote at $599/yr (for LibAnalytics alone) for five independent datasets, which makes LibAnalytics one of the pricier options. You can maintain a single dataset for free, a good opportunity to try LibAnalytics out and see if it’s worth it.