by Antonio Alonso Espadas (Library IT Specialist, Universidad Europea de Madrid)
and Javier Martín Rodríguez (Head of Technologies at LRC Library Dulce Chacón – Universidad Europea de Madrid)
Please Note: Some of the images in this article may be difficult to read. To access the high-res versions, please contact the article’s authors.
Once NASA succeeded in getting men on the moon, the American voter began wondering why it was necessary to allocate such a big amount of money to something that seemed already routine when there were other incoming and urgent issues, such as the oil crisis or the Vietnam War. At that time, the budget of the institution was reduced, and the good times did not come back.
NASA understood then the need to convince the public that every invested dollar in space exploration was well spent, and it developed a marketing program conveying the benefits that exploring and experimenting in space provided to society, which had been proven effective.
Once the objective of reaching the moon was almost assured, the budget of the agency was drastically reduced.
Something similar happens to us libraries; we are highly praised in terms of academic and cultural reputation and everybody agrees on the role we exercise in our communities. But when we refer to the resources we need to deliver on it, there is not always such unanimity.
If there is anything that characterizes a library, it is its capacity to gather vast amounts of data in a systematic way, thanks to the intensive use of technology, such as the ILS and other information systems. But, what do we do with all that information? To a large extent, we use it to demonstrate how good we are in certain aspects when we compare ourselves to other libraries (number of visits, online resources downloads, borrowings, etc.)
The chart above shows the indicators provided by our library to REBIUN (Red Española de Bibliotecas Universitarias). The data and statistics related to the institutions that participate in REBIUN can be consulted in: https://rebiun.um.es/rebiun/admin/ManageIndicatorsPage.
We all know that data help library directors explain strategic decisions, such as why we need to increase our budget or how our bibliographic collection needs to grow.
But data should also help the Heads of Service know how many librarians must be at the reference desk at certain times, or the impact the digital resources used in teaching students has on the level of degree achieved.
If we go a little further, data should also assist all librarians to make decisions, like determining which parts of the collection are more in demand so they can prioritize accordingly; or calling the maintenance department if a particular place of the library has issues with the temperature; or if it would be advisable not to schedule meetings at times when most users visit the library heavily.
This all sounds great, but it’s not easily achieved. Even though libraries usually count on various systems for collecting information, such us our ILS, COUNTER reports of digital resources, temperature, humidity and fullness data gathered by the “Internet of Things” (IoT) sensors, or the borrowing of equipment and work rooms of the library, unfortunately, there was no system capable of collecting data from different sources, grouping and connecting them, and providing us with the necessary evidence to make decisions; we needed to create it.
We needed the system to be flexible, capable of analyzing raw data, like already processed reports from different sources and formats, as well as being always available so as not to have to run reports every time a certain fact is needed; we wanted it to be ready to go instead.
Three years ago, we started testing Microsoft Power BI, a business intelligence system which is simple to parametrize, intuitive and creates reports that allow us to zoom into the data series and interact with them in a wide variety of graphic representations.
We started processing data from our own reports, mostly in Excel. It was easy because both programs belong to the same producer; therefore, we only needed to open the Excel files in a OneDrive account connected to our MS Power BI account to have our data updated in the reports constantly.
We then decided to go even further; two years ago, we moved to ILS Koha, an Open Source system that uses MySQL’s database engine, and we integrated it directly to our MS Power BI system. After all, why get data from a system to make a report in Excel when you can obtain it directly from the database charts from the ILS? Thanks to this initiative, all librarians have real-time information regarding circulation transactions (borrowing, renewals, use within the library or devolutions) through branch and collection or how digital resources with MARC 856 field are being used (books, digital magazines and databases). Depending on this and other information, we can make decisions about which subscriptions we should renew and which ones we should cancel.
In the first image above, there is a capture of the data from Koha database in the statistics chart, which is followed by a graphic representation of the data collected in this chart. The report allows you to limit the search regarding date, kind of transaction, branch and collection. All data are recalculated automatically and once you click on the elements, they appear in different colors.
This report shows the copies and statistics charts of circulation. The dots represent the letters of the LC classification; those indicated in the higher positions and on the left of the graphic show the thematic areas of preferent growth (through smaller number of copies and higher number of borrowings), while the lower ones are areas of potential to purge. The report has options of classification, collection, type of material and year of publication.
Last year, we started to introduce IoT systems in our library to simplify data gathering regarding the number of user visits, peaks of occupancy per hour and measures of temperature and humidity. We also started an inventory of materials with an RFID system with WIFI connection.
Those systems depended on their own platforms and measuring reports, but once again we wondered, why not integrate them as well? Thanks to this, the task of inventory of materials is done in real-time, without the need to overturn data from the reading device in the ILS.
We have also been able to detect air conditioning failures in certain areas, which has allowed us to provide maintenance services with the precise information to fix it.
This report provides information about the environmental conditions and the occupancy levels of the LRC Library (it collects information from eleven sensors, whose information is shown in two simple graphs). https://web-uem.bibliocrai.universidadeuropea.es/index.php/es/nivel-de-ocupaci%C3%B3n-de-las-instalaciones-y-condiciones-ambientales
Furthermore, we also have updated information on the occupancy level of the Library LRC.
The first picture above shows the information provided by a person-counting camera in real-time. The report below that shows the same information, but processed to know the occupancy level of the LRC Library in one-hour intervals, with a big enough sample to predict the occupancy peaks and to assets the number of times the LRC Library has reached its maximum occupancy level.
We now have a system that integrates practically all the reports of the library, updates them with required regularity and offers interactive and analytical tools for the decision-making process based on evidence.
It is worth mentioning that by the time this article is published, we will have started a new website for the LRC Library, which incorporates some reports and tools managed through this system, such as occupancy level and environmental conditions of our branches in real time, the description of the bibliographic collection of each of our schools, the programming of training workshops from the library and the level of accomplishment of our services (more information can be retrieved from: https://web-uem.bibliocrai.universidadeuropea.es/index.php/es).
In conclusion, sometimes you will need to make improvements to satisfy the user’s needs, but if you cannot measure the needs with objective data you will hardly be able to justify the budget and resources needed to start them.
What are the main ways to take this path? The first one is allowing your team to make better decisions based on evidence, and the second is to be trustworthy and transparent for managers and community, offering accurate and updated data about the status of collections, facilities, use of resources, etc.
We have been speaking about technology and making decisions, but what happens to us, librarians? What will be the role we’ll play in this context? Let us finish with a few words about this aspect.
All the systems we mentioned in this article are impressive and help to minimize the risk of making wrong decisions, but we should not fall to the temptation of forgetting about our own responsibilities. When we make a decision, there are different dimensions we must consider, and data analysis is only one of them; there are also qualitative aspects these systems are still not capable of understanding.
Librarians are the key element in any decision-making process; data analysis systems are helpful tools for us, but what really matters are the decisions you make.