New research finds that colleges have automated student communications and data systems, but transcript evaluations, identity checks, and complex admissions processing still consume significant staff time.

WHAT’S HAPPENING

Colleges and universities are investing in automation to improve admissions operations, but new research suggests that much of their most time-consuming administrative work remains manual.

A 2026 report from education technology company EDMO examines how higher education institutions are using automation and where significant operational gaps remain.

The study surveyed 40 higher education leaders at colleges and universities across the United States.

Its findings reveal a substantial difference between automating routine administrative activities and improving the more complicated processes involved in admitting students.

The difference is significant.

Colleges have made progress automating activities such as sending routine messages, transferring information between systems, and maintaining applicant records.

But evaluating academic transcripts, verifying applicant information, reviewing documentation, and resolving discrepancies frequently require substantial staff involvement.

These activities can be complicated because applicants come from different schools, countries, and educational systems.

Transcripts may use different grading scales, course descriptions, credit systems, and academic requirements.

Admissions personnel must determine whether applicants meet institutional standards while ensuring that important decisions are accurate and fair.

The report also identifies staffing pressures.

Approximately 70% of surveyed leaders cited staffing and workload challenges, suggesting that automation investments have not eliminated the operational burden on admissions departments.

Applicant support presents another challenge.

About 72.5% of surveyed institutions reported lacking around-the-clock applicant support.

Nearly half indicated that 20% to 40% of student inquiries still required staff intervention.

The research also identifies growing concern about applicant fraud, with 65% of respondents reporting an increase over the previous two years.

That creates additional pressure to verify identities, academic records, and supporting documentation.

Importantly, the study addresses admissions automation broadly. Its findings should not be interpreted as evidence that every automated function uses advanced artificial intelligence.

Some processes rely on conventional software rules, customer relationship management systems, or established workflow automation.

The central finding is that colleges have automated many routine activities while leaving some of the most complicated and potentially expensive admissions responsibilities largely dependent on human employees.

WHY IT MATTERS

Artificial intelligence is frequently promoted as a way to reduce operating expenses and improve productivity.

However, the financial benefits depend on which activities are automated.

A college may save time by automatically sending thousands of admissions emails.

But if staff members still spend hours reviewing transcripts, resolving documentation problems, and verifying applicant information, the largest operational bottlenecks may remain.

Automating a process does not necessarily mean eliminating the work that makes the process expensive.

That distinction is particularly important in higher education, where admissions departments manage complex decisions involving applicants’ academic records, qualifications, and eligibility.

Simple communication tasks are often easier to automate because they follow predictable patterns.

An admissions system can automatically remind applicants about deadlines or notify them that documents are missing.

Transcript evaluation is different.

A single applicant may have completed coursework at multiple institutions using different credit systems.

The college must determine whether those courses satisfy its academic standards.

Incorrect evaluations can affect admission decisions, transfer credits, and students’ educational progress.

As a result, automating these activities requires more than generating text or moving information between software systems.

It may involve document recognition, data extraction, institutional policies, academic equivalency rules, verification procedures, and human review.

AI could help with portions of this work, but consequential decisions still require appropriate safeguards.

The same principle applies to identity verification and fraud detection.

Automated systems may identify inconsistencies or suspicious information.

However, suspicious patterns should trigger appropriate review rather than automatically establish that an applicant committed fraud.

False positives can create serious consequences for legitimate students.

The report therefore illustrates an important challenge in adopting AI responsibly.

The greatest opportunity may involve improving difficult operational processes rather than simply automating the activities that are easiest to automate.

This has implications well beyond higher education.

Mortgage lenders, insurance companies, financial institutions, legal offices, and other document-intensive businesses face similar problems.

Many organizations have automated routine communications while continuing to depend on employees for complicated verification, exception handling, and decision-making.

For example, an insurance company may automatically acknowledge receipt of a claim while still requiring extensive manual review before determining whether to pay it.

A mortgage lender may automatically request financial documents while still needing qualified staff to evaluate complicated borrower circumstances.

A business may therefore appear highly automated without achieving major reductions in the work required to complete its core operations.

There is also an opportunity involving applicant communication.

Admissions departments receive numerous repetitive questions about deadlines, application requirements, financial aid procedures, academic programs, and document submission.

An AI assistant configured with approved institutional information could help answer many routine questions outside normal office hours.

More complicated requests could then be directed to employees.

However, such systems must accurately represent institutional policies and avoid making unauthorized admissions decisions.

Accessing an applicant’s individual status would also require secure connections to approved institutional systems.

The objective should be to reduce unnecessary administrative work while maintaining accuracy, privacy, fairness, and human accountability.

WHO BENEFITS

– Colleges and universities: Could improve admissions efficiency by identifying and automating repetitive tasks while reducing bottlenecks in more complicated processes.

– Admissions personnel: May spend less time answering routine questions and transferring information, allowing greater attention to complex applications.

– Students and applicants: Could benefit from faster responses, clearer instructions, and improved access to application information.

– Educational technology providers: May gain opportunities to develop systems addressing transcript processing, application verification, and admissions workflow management.

– AI assistant developers: Could provide institution-specific systems for answering approved admissions questions and directing complicated requests to qualified staff.

– Document-processing software companies: May benefit from demand for tools that extract, classify, and organize information from academic records.

– Institutions with limited administrative staffing: Could gain additional capacity if automation reduces repetitive work without introducing excessive new responsibilities.

– Businesses in other document-intensive industries: Can apply lessons from higher education when evaluating their own automation investments.

WHO LOSES

– Institutions automating only simple tasks: May continue facing substantial administrative expenses if their biggest operational bottlenecks remain unresolved.

– Colleges relying on disconnected software systems: Could experience duplication, inconsistent records, and additional manual work.

– Technology vendors offering superficial automation: May face greater scrutiny from customers demanding measurable improvements in complete workflows.

– Organizations deploying unreliable document-processing systems: Risk incorrect evaluations, delays, and additional correction expenses.

– Applicants affected by inaccurate automated decisions: Could face unfair outcomes if schools fail to establish effective human review procedures.

– Institutions with inadequate fraud controls: May remain exposed to fraudulent applications and questionable documentation.

– Businesses assuming automation automatically reduces staffing needs: Risk creating operational problems if they eliminate personnel before understanding the remaining workload.

– Employees performing repetitive administrative tasks: May face changing responsibilities as institutions introduce more sophisticated automation.

WHAT HAPPENS NEXT

The next stage of admissions automation will depend on whether colleges can move beyond routine administrative tasks and improve the processes that create the greatest workload.

Several developments will determine whether that happens.

More sophisticated document processing

Colleges may increasingly evaluate technology that can extract information from transcripts, identify missing records, and organize application materials.

The challenge will be maintaining accuracy across different institutions and academic systems.

Improved transcript evaluation

AI-assisted systems could help admissions personnel compare academic records with institutional requirements.

However, complex decisions may still require human evaluation, particularly when records are incomplete or difficult to interpret.

Better fraud detection

Institutions may invest in systems that flag suspicious documentation or inconsistent applicant information.

Effective procedures will need to balance security with fairness and provide appropriate opportunities for applicants to resolve problems.

Expanded applicant assistance

AI-powered admissions assistants may become more common as colleges seek to answer routine questions outside normal office hours.

These tools could provide information about deadlines, required documents, program descriptions, and application procedures.

Any access to personal application records would require appropriate authorization, privacy safeguards, and technical integration.

Integration of disconnected systems

Institutions will need to determine whether admissions software, student information systems, document platforms, and communication tools can exchange information reliably.

A process that requires staff to repeatedly enter the same information into different systems remains inefficient even when individual components are automated.

Measuring complete operating costs

Colleges should evaluate the full amount of time and money required to process applications, including automated activities, human review, exception handling, and corrections.

That will provide a more accurate measure of whether technology investments are reducing operational expenses.

Maintaining institutional accountability

Admissions decisions affect students’ opportunities and educational futures.

Schools must establish appropriate standards for accuracy, fairness, privacy, and human oversight.

Automated systems should support qualified decision-makers rather than make unsupported determinations about applicants.

The findings also create opportunities for technology providers serving education.

However, not every institution needs the same solution.

A smaller college may primarily need help answering routine questions and organizing applicant communications.

A larger university processing complex domestic and international applications may face more substantial challenges involving transcript evaluation and verification.

The appropriate approach depends on the institution’s actual workload.

This distinction is important for businesses selling AI services.

A basic informational assistant may be relatively straightforward to configure using approved institutional knowledge.

A system that evaluates academic transcripts or verifies identities requires significantly more specialized capabilities, data access, testing, security, and oversight.

Those services should not be presented as interchangeable.

For colleges considering additional automation, a practical starting point is to measure how employees spend their time.

Which activities consume the greatest number of hours?

Which processes repeatedly create delays?

Where are errors most expensive?

Which tasks can be safely automated, and which require professional judgment?

Answering those questions can help institutions invest in technology that addresses real operational problems.

Additional research involving larger samples will also be important.

The EDMO survey provides useful directional evidence, but 40 respondents cannot establish the exact state of admissions automation across all U.S. higher education institutions.

Future research should examine whether more advanced automation actually reduces processing times, improves accuracy, controls expenses, and benefits applicants.

The larger shift is clear: Artificial intelligence and automation are forcing organizations to distinguish between digitizing routine activities and solving their most expensive operational problems. EDMO’s research suggests that colleges have made progress automating communications and data management while leaving complicated admissions responsibilities largely manual. The next opportunity is to build reliable systems that address those bottlenecks without compromising accuracy, fairness, or human oversight. For colleges and businesses alike, the real measure of successful automation is not how many tasks use technology, but how much the entire organization improves as a result.