It is no secret that higher education has become increasingly focused on efficiency. We automate messages, build self-service systems, implement artificial intelligence, track students through dashboards and create workflows designed to accomplish tasks with fewer steps and less human intervention. Much of this is good, and many of these changes are long overdue.
Students should not have to wait three days for an answer that they could receive in three minutes, and professionals should not spend hours completing administrative tasks that technology can accomplish more accurately and efficiently. Technology can remove unnecessary barriers, make information more accessible and give faculty and staff time to focus on more meaningful work. However, somewhere in our pursuit of efficiency, higher education may have adopted an assumption worth questioning, which is that every inefficiency represents a problem waiting to be solved.
Sometimes inefficiency is exactly that. A student should not need to visit three offices to answer one question, complete multiple forms requesting the same information or struggle to understand an unnecessarily complicated institutional process. An advisor should not have to enter the same information into multiple systems or spend valuable time completing repetitive administrative tasks.
Technology gives colleges and universities an opportunity to eliminate many of these frustrations for both students and employees. The problem comes when we begin treating every extra conversation, manual review, phone call or human interaction with the same suspicion. There is an important difference between an inefficient process that creates a barrier and an inefficient interaction that creates an opportunity.
When Efficiency Costs Us Insight
Academic advising provides a good example of that distinction. Imagine a student contacting an advisor with what appears to be a straightforward registration question that could probably be answered in a few minutes. From an efficiency standpoint, the ideal outcome might be resolving the question as quickly as possible, perhaps even through a chatbot, FAQ page, automated email or self-service platform, but anyone who has spent significant time working directly with students knows that the question a student asks first is not always the question they actually need answered.
A five-minute registration conversation can unexpectedly become a twenty-minute discussion about academic performance, finances, career uncertainty, family responsibilities or whether the student is considering leaving college altogether. On a spreadsheet, those additional fifteen minutes might look inefficient, but in practice, they may be the most valuable fifteen minutes of the interaction.
What Technology Cannot Always See
This point becomes especially important when supporting nontraditional learners whose experiences do not always fit neatly within traditional institutional processes. Students balancing college with full-time employment, children, financial responsibilities, military service, caregiving or a return to education after many years may encounter challenges that are difficult for any automated system to fully understand. A dashboard may show that a student has not registered for the upcoming semester, but it may not show that the student’s work schedule recently changed and made the courses they need impossible to attend. An automated message can remind someone about an approaching deadline, but it cannot always recognize hesitation in a student’s voice or ask the unexpected follow-up question that reveals why that deadline is about to be missed. These circumstances often require conversation, context, judgment and sometimes a willingness to spend more time than an efficient process would prescribe. Ultimately, data can tell us that something happened, but human interaction can help us understand why.
The same principle can extend beyond individual advising conversations and into academic program coordination. Program coordination frequently involves work that appears inefficient from the outside, including manually reviewing student records, discussing unusual situations with faculty, checking whether policies apply as intended or communicating between offices to resolve a student’s problem. Certainly, some of these processes can and should be improved, particularly when unnecessary bureaucracy creates obstacles for students.
However, repetition and manual interaction can also reveal patterns that automated systems may not immediately recognize. If one student misunderstands a requirement, that may simply be an individual misunderstanding, but if an advisor or program coordinator has the same conversation with twenty students, the problem may be the requirement, the way it is communicated or the structure of the program itself. The repeated interaction that initially appears inefficient can therefore become a valuable source of institutional knowledge.
This is where higher education’s growing need to deliver support at scale becomes more complicated. Colleges and universities understandably want student success initiatives capable of reaching hundreds or thousands of learners, but it comes with the desire to do so without requiring a proportional increase in staffing or resources. Automated nudges, early alert systems, predictive analytics, artificial intelligence and self-service platforms make that possible in ways traditional models often cannot. However, scale can also encourage institutions to measure success by how many interactions technology can eliminate or how quickly students can move through a process. Fewer appointments, fewer emails, fewer manual reviews and faster resolutions may demonstrate operational efficiency, but they do not automatically demonstrate that students are receiving better support. Instead of asking only, “Can we automate this?” leaders should also ask, “What happens during this interaction that we might lose?” If the answer is nothing meaningful, automation may be exactly the right solution, but when an interaction creates opportunities for judgment, relationship building, discovery or intervention, eliminating it deserves considerably more consideration.
Using Technology to Create Capacity for Connection
This argument does not require higher education to reject artificial intelligence, automation or data-driven approaches to student success—quite the opposite, as technology can make higher education better when it removes work that prevents people from doing the work only people can do. Students do not need a meaningful human relationship with the process that resets their password, confirms that a form was received or sends a routine reminder about an established deadline. The challenge is ensuring technology is applied intentionally, based on whether an interaction should be automated. Institutions should be especially cautious when efficiency is measured primarily through time saved, interactions eliminated or the number of students who can be served with fewer human touchpoints. A more useful measure of innovation may be whether technology increases faculty and staff capacity to recognize and respond to students as individuals rather than simply reducing the number of times they have to interact with them.
As higher education continues to serve increasingly diverse and nontraditional populations, institutions will face legitimate pressure to become more flexible, scalable, accessible and efficient. We should pursue those goals, and we should continue using technology to eliminate processes that waste students’ time or prevent professionals from focusing on more meaningful work, but we should also leave room for the phone call that takes longer than expected, the faculty and advisors who ask one more question, the coordinators who look at something twice and the conversation that wanders beyond the reason it began. Those interactions may not always fit neatly into an efficiency metric, and they may occasionally make a process slower than technology alone could make it. Not every inefficiency deserves preservation, but neither does every inefficiency deserve elimination. Sometimes the seemingly unnecessary interaction is where we discover the problem we need to solve, and sometimes the inefficiency itself becomes the intervention.