Workflows or How do we tame Higher Ed inefficiency

One of the main causes of low organizational efficiency is that universities do not routinely review their procedures. It does not seem to be anyone’s job. Various people create procedures all the time, but no one seems to be in charge of periodically reviewing them and making dure they still do what they are supposed to do. For example, in the College of Education, there are 47 annual processes that require some college-level coordination. Some of them repeat every semester. In addition, there are dozens within each of our branches that only their respective staff have. Some of these are fully redundant. For example, we report reassigned time for our chairs and coordinators in a special form, but we also report the same on faculty workload modules. Or here is another beautiful thing: The University has a storage for course evaluations by students. But because they wipe the data out every so often, for completely obscure reasons, our staff download hundreds of them on a network drive, and then upload a significant portion of them again to faculty RTP files, this time on OneDrive. What should have been one link, becomes three downloads and two uploads, for hundreds of files. I could go on and on with a list. A fundamental problem is that someone creates workflows, but no one reviews and improves them.

To conduct a workflow review, these are the basic steps:

1. Is the work process needed at all? What happens if it is simply abandoned? In most cases, the answer is no, but asking this question really helps tog et at the purposes. Some procedures have been designed to prevent a very rare undesired outcome. But their cost is often disproportionally high. I worked at a place, where the President had to sign all travel requests over $250, because many years ago someone had an embarrassing trip to Hawaii that got media coverage. Well, such trips never occurred again, but the procedure stuck for years.

2. Work procedures are flows of information. It is helpful to start considering them from the first point to the last one and try to create the shortest path possible. For example, some processes are applications, where everyone is eligible. There is no point in approving applications, so that step should be removed. In other cases, a university unit would send something to deans with a request to distribute to faculty. However, that unit has access to all faculty emails and can the information directly, without bothering deans. Sometimes extra steps appear just in case, because in similar processes, there have been such extra steps. The shortest is the chain, the better.

3. Similarly, instead of approvals, consider informing. If the risk of inappropriate decision is very low, consider self-attestation instead of active controls. For example, we routinely approve student requests for late class drops. And these requests require approvals of their instructor, the chair and the dean. I guess the hope was that 3-level approval would give the process some rigor. But it does not, for none of the three have any way of verifying student stories of hardship. In these cases, it is enough to say something “I swear the explanation is true, and if not, this would be a violation of honor’s code.” This way, three people would not be involved in processing a decision none of them actually made. Examples of extraneous approvals are a many; they all distract us from doing our jobs.

4. Universities have many databases that often do not speak to each other. What we do not want is a process, where someone reads data and them manually enters it into another database. Not only it is insanely time-consuming, but also prone to errors. If at all possible, a procedure should stay within one database.

In the end, a workflow review is not a set of specific steps. It is just an attempt to take a look at what we are doing and why we are doing it. There is no magic solution, just a drive toward simplicity, sanity, and automation.