As I have mentioned in my previous posts, the explosion in technology around us and the desire on the part of our own community members to adopt several of them is real. We, in Library and Technology Services (LTS), have an obligation to manage this process in a way that the College’s technology portfolio is manageable and sustainable. LTS, a merged Library and IT organization, is a small one and we cannot be good at supporting too many technologies at a level of depth that is necessary. For example, information security is of paramount importance and the responsibility for data grows exponentially as one adds more and more systems and services to support and we just don’t have the personnel to support them.

Our recent strategy has been focused more on a handful of principles – low code systems, minimize customizations, highly secure systems, security education of the community, satisfy the needs of the community as much as possible with existing systems and contain the portfolio sprawl. We need to constantly remind ourselves that we are here to support the academic mission of the College primarily and need to optimize our resources accordingly. One of the major steps we have taken towards these goals is to adopt systems that require less programming and change responsibilities of the programmers to make them be  business analysts. This has dramatically changed how we support administrative offices.

This change was rooted in the parallel to academic computing support. There are generally no programmers in the group. But they have subject matter experts who work with the faculty and students to adopt technology appropriately for teaching, learning and research. This is exactly what we have done on the administrative side by the move away from programmers to business analysts.

We also have been successful in consolidating services to some of our existing systems. I would like to share a major step in this direction that is taking place as we speak.

We are very proud of having constituted a data governance group 10 years ago, primarily around the governance of student data. We also implemented our first student data warehouse using a product from a company called iStrategy, which had an excellent data model for student data that was agnostic of the administrative systems. We called this system WANDA (Wellesley ANalytics Decision­ support Applications). It was built on Microsoft SQL Server and front ended by Pyramid.

Every night the system would extract relevant data from Banner, at that time, do the necessary transformations and load them into appropriate tables in SQL server. The data was optimized into OLAP (Online Analytical Processing) “cubes” that made analysis much faster. There were some technology limitations in that these worked initially much better on Windows than in MacOS, but subsequently technology improved to support them both. This has been a critical service we have provided to our institutional research colleagues as well to several academic department chairs and academic administrators.

Blackboard acquired iStrategy (and renamed it Blackboard Analytics) a few years later and their focus for this product shifted and was not in line with our expectations. In order for us to keep up with the needs of the community, essentially we branched off and continued to customize the product. You can understand how this creates problems with upgrades etc. Lack of clarity for support and costs for the Analytics product and Pyramid caused us some major headache, as a result the. infrastructure where we continue to run the components is less than ideal. Long story short, in order to make such a valuable system for the community to be secure, supportable and sustainable, we really needed a different strategy.

While engaged in a major implementation such as Workday, we were unable to look into this immediately. So, to continue WANDA, our staff did a wonderful job of extracting data from Workday and feeding it to the SQL server infrastructure. After the implementation of Workday student, we began to look at Workday Prism, which has a lot of functionality that matched our needs. As is to be expected, its terminology for presentation of data is slightly different, such as Discovery Boards, but we felt that it was functionally equivalent.

But it also had significant advantages. It is relatively low code and the configurations are far more transparent and can be easily learned by a larger group of staff than SQL server code. It obviously helps us reduce the portfolio of technology and the data we are analyzing is right there in Workday.

With some excellent work on the part of our staff and in collaboration with Workday software engineers, we have been successful in moving our WANDA to Prism. Initial reaction from some of the faculty and administrative staff has been positive and like all new technologies, there will be a transition period when we will hear about changes needed, which we are prepared to act on.

As I have explained before, our approach to such transitions is incremental. We could have waited until everything is done with data from Workday, transformations in Prism etc. Instead we have taken the following approach – extracting data from Workday to SQL server and transforming the data there is working, so let us bring the transformed data to Prism on a nightly basis and work with our users to improve the front end first. Then slowly shift this process natively to within Workday. We have done the first step and it is working great and soon we will launch the next phase.

Another technical detail. Those of us who move to Workday are always faced with the important question of “How much historical data do we need to bring to Workday?” Each of us decide differently. We decided to bring only the relevant data for active students. However, for datawarehouse reporting you need to bring all historical data. Whereas it is theoretically possible to bring them all into Workday, it is an enormous amount of work. However, we have brought them all into Prism and have been able to report on them effectively.

This is a major step for us in accomplishing several of our goals – consolidation of technology, when we finish the next phase, we will have less technical components to support, low code and sustainable solution.

This would not have been possible without the hard work of some of my colleagues in LTS and strong collaboration with our colleagues in Office of Institutional Research.