The Weekly Bellows: Someone on the Other End (10/14/26)

*Lena, Jo and Ellis are fictional characters created for this episode. Their campus is imagined; the research and policy developments are real. Evidence checked October 13, 2026, Pacific time.* The question under the question Lena has two windows open on her laptop and a piece of toast going cold beside it. One window says her courses transfer. The other lists the requirements for the degree she wants. She has been switching between them long enough that the laptop has asked whether she is still there. She would also like to know. Lena is twenty-nine. In our fictional campus story, she is trying to arrange college around a job whose schedule arrives one week at a time. She can manage another form. What she cannot manage is choosing classes on the assumption that somebody, somewhere, has checked whether they move her toward graduation. She types a question to the transfer office. Will these courses count toward my major? The reply arrives almost immediately. It is friendly. It includes a link to the page she already has open. Across campus, Jo is clearing the morning queue. She stops at Lena’s reply: I’ve read this. I’m asking what it means for me. Jo picks up the phone. This week, we’re looking at what happens between an answer being sent and a student being able to use it. A new transfer platform, recent re-enrollment findings, and a fresh reminder about trust in artificial intelligence all bring us back to the same workbench. Somewhere in the process, a person needs to take responsibility for the next step. And for those of us building the forms, messages and dashboards, that responsibility ought to be visible. When credits travel but the student stalls A September report from the University of Washington’s Community College Research Initiatives and the Washington Student Achievement Council gives us a useful place to start. Researchers interviewed twenty Washington students at different stages of transfer, including people who had paused their education. Students described credits that were accepted by a receiving institution but did not fulfill the requirements they needed for their intended major or graduation. They also described advising arriving late, or without enough connection to their circumstances. This is a small qualitative study, not a statewide estimate of how often those problems occur. Its strength is in exposing where a process can become difficult to use.1 The report recommends earlier, program-specific advising and contact with prospective transfer students before admission.1 That is a practical invitation to look at the boundary between recruitment and advising. Who is allowed to answer a substantive question while a student is still deciding whether to apply? In our story, Jo opens Lena’s message and asks which degree she has in mind. Then she asks when Lena can actually attend classes. Lena had expected to repeat her student identification number. She had not expected anyone to ask about Tuesday evenings. For a workshop exercise, take a transfer page and read it as someone trying to make one decision. Can I finish this program on the schedule I can afford? Circle the parts that answer it. Circle, in another color, the parts that simply send the reader elsewhere. You may discover that the page is beautifully written and the answer is still missing. A new map in North Carolina On October 9, the North Carolina Community College System and North Carolina Independent Colleges and Universities announced the public launch of NCICU Transfer Pathways, following a September soft launch. It connects all fifty-eight community colleges with twenty-nine participating independent campuses. That total includes twenty-eight four-year institutions and the two-year Louisburg College. The platform brings course matching, degree planning and campus connections together. Reverse-transfer capability was described as future work, not something already delivered.2 There is an important qualification in the platform’s own FAQ: its evaluation is a planning guide, and the receiving institution makes the final official transfer decision.3 That qualification belongs beside the exciting feature. If we let an early estimate sound like a final decision, we can create a new surprise precisely where we hoped to remove one. My recommendation for a transfer team is to pair every preliminary result with a clear next step: who can verify it, what they will need, and how the student will hear back. The registrar and academic program need to agree on what can be promised. Admissions can then make that promise understandable. For analytics colleagues, a useful measure might be the time between a student’s first credit question and a confirmed answer that resolves it. Treat that as a proposed local measure. We do not have a universal target to stamp onto every campus. And please keep the difficult cases in view. An average that looks reassuring can conceal the people who are still waiting. The handoff that disappears from the dashboard Ellis, our fictional CRM analyst, joins Jo at her desk. His screen says Lena’s inquiry is complete. The first response was sent. The workflow did exactly what it was built to do. Jo slides the conversation across to him. Ellis reads it, then looks back at the green status on his screen. He changes the label in his notes from answered to reply sent. The green looks a little less impressive now. This is where a small build can be more useful than a grand redesign. In Slate or another CRM, choose one high-friction question and make the handoff inspectable. Record the question, the office receiving it, whether someone has accepted it, and what the student has been told to expect. Agree on what counts as resolved before adding another completion flag. These are design recommendations, not newly announced Slate features. Use the permissions and tools your institution has approved. Keep sensitive documents in their proper system; a routing record does not need a copy of everything a student has disclosed. Then test the awkward case. The named adviser is out. The student replies after the queue closes. Two offices each believe the other owns the decision. Does the system expose that uncertainty, or does it quietly mark the work done? A well-made joint has to hold when somebody leans on it. A handoff deserves the same test. An invitation with support behind it We also have stronger evidence for reaching students who have already left college. MDRC released new REACT findings on October 6, and they reached our reading queue on October 7. We are carrying that evidence into this week’s discussion rather than presenting it as a new October 14 release.4 The study used random assignment at four Florida community colleges. In the full sample of 4,226 eligible students, about twenty-four percent offered the program re-enrolled, compared with fifteen percent in the comparison group. The package combined outreach, a designated adviser, and tuition and institutional-fee waivers for up to fifteen remaining credits toward an associate’s degree. Eligibility was specific; the study was not an invitation to every former student.4 Those results support the package in that setting. They cannot tell us that a particular subject line, an adviser alone, or a tuition waiver alone produced the difference. The early findings also do not establish long-term completion effects.4 That distinction matters when a result becomes a strategy slide. It would be tempting to write a better return-to-college email, put the study in the footnote, and expect the same improvement. The support behind the message is part of what was tested. Before launching a local re-enrollment campaign, I would put enrollment, advising and financial aid in the same conversation. Which students can we genuinely help this term? Can we confirm what remains in their program? Is there assistance available, and can someone explain the conditions before the student commits? This is also a place to slow down our language. A record may say stopped out. A person may say their work hours changed, their transportation failed, or they could no longer make the schedule fit. Our fictional Lena has not lost interest in education. She is trying to establish whether this version of college has room for her actual week. Re-engagement gives enrollment teams another population to consider alongside the incoming first-year class. It does not, by itself, tell us how large the reachable population is at any particular institution. That is work for our own carefully defined data. The bot and the person behind it EAB’s October 12 briefing reported findings from its 2026 survey of 11,700 students and recent graduates. It said eighty-six percent had used AI tools, while twenty-five percent mostly or completely trusted them. This is vendor research reported in a received newsletter. The email does not supply enough sampling, weighting or questionnaire detail to establish how representative those figures are. We should treat the finding as a prompt for investigation, not a benchmark for our own applicant pool.5 The useful question for a campus is what earns confidence after someone has tried a tool. A Washington example offers a design to inspect. A July brief described an OtterBot customization pilot with five regional partners. It added recognizable local branding, local information, messages from human counselors and escalation to trusted people in the community. The brief anticipated later findings on engagement and aid applications; it did not demonstrate an enrollment gain. It also does not establish that the chatbot is a generative-AI system. This is older implementation context, not a new product launch or proof of AI effectiveness.6 My practical takeaway is to test the moment when an automated answer reaches its limits. Can the student recognize that limit? Can they reach someone who can help? Does that person receive enough context to avoid making the student begin again? Ellis drafts a message for Lena. He starts with three paragraphs about the institution’s commitment to her success. Jo points to the last sentence, where he finally explains who will review the courses. They move that sentence to the top. The message gets shorter. It also gets more useful. If you want to try this with your team, take one real template and remove the branding for a moment. Ask a colleague to identify the next action, the person responsible, and the promise being made. If those answers take a debate, the student is being asked to do the same interpretive work alone. You can use AI to assist with drafting or checking consistency, subject to your institution’s rules. The named owner still needs to verify the facts, the conditions and the route to help. The test is whether the final communication can be trusted and acted on. Two items for the operations desk First, financial aid reporting. Federal Student Aid updated its Financial Value Transparency and Gainful Employment guidance, usually shortened to FVT/GE, on October 8. The update identifies January 15, 2027, as the final extension for required reporting across the 2024, 2025 and 2026 cycles, including missing or underreported data. The 2026 cycle originally had an October 1 deadline. Institutions may also revise or correct submitted data through January 15. The reduced reporting option associated with early implementation of the Student Tuition and Transparency System, or STATS, applies to the 2026 cycle; it does not erase the prior-cycle requirements.7 The reporting owner, institutional research and financial aid should reconcile the applicable cycles, files and unresolved errors together. A file having been submitted does not establish that it is complete. FSA explicitly makes that distinction.7 We are not making a finding about any particular institution’s compliance. Second, FSA’s October 9 update says borrower-defense functionality is available again in Partner Connect after the earlier outage. It extends each pending case-response deadline by seven calendar days, with the updated dates shown in the system.8 If your team has relevant cases, use those case-specific dates. Do not leave an old outage instruction on a checklist simply because it was correct when somebody printed it. The practical thread is familiar: a notice has to reach the person who can act on it, and somebody has to check that the action happened. The international watch For colleagues following the proposed OPT fees, the October 8 published Federal Register notice confirms November 9, 2026, for comments on the rule, and December 7 for the separate information-collection comments. Online comments are due by 11:59 p.m. Eastern on the applicable final day. Those are comment deadlines. The proposed fees are not currently effective.9 Our Breaking Bellows episode explains the proposal in more detail. For this week’s campus work, keep student-facing guidance dated, link the official notice, and route individual immigration questions to the designated school official or qualified adviser. A proposed change can require careful communication well before anyone knows the final outcome. Put one handoff on the bench Back in Jo’s office, Lena now has a preliminary plan, a clearly marked question for the academic program, and a name attached to the follow-up. She has not been promised that everything will work. She knows what is being checked. Before she leaves, she asks whether she should keep refreshing the portal. Jo tells her when to expect a reply. Lena closes the laptop. It is the first task she has completed all morning that does not create another tab. Our suggested workshop this week is deliberately small. Choose one student question that crosses an office boundary. Follow it from arrival through accepted ownership to a usable answer. Write down where the student has to repeat themselves, where a promise has no owner, and where a status changes before the underlying work is complete. Bring the people who do that work into the review. Ask them which exception causes the most trouble and what they already do to keep it from falling apart. There may be considerable craftsmanship hiding behind a process diagram that looks very simple. Then make one repair and test it again. Record whether the answer arrived, whether it resolved the question, and what still needs a person. Keep the results modest and local until you have enough evidence to say more. That is this week’s work at the Forge. Give one handoff a name, a next step and a way to tell whether it held. References 1\. Davis, S. C., Reed, J. A., Nuñez Martinez, M., Wetzstein, L., and Chien, A. Navigating Transfer in Washington: Student Experiences, Barriers, and Supports. Executive summary. University of Washington Community College Research Initiatives and Washington Student Achievement Council, September 2026. Qualitative interviews with 20 students. https://wsac.wa.gov/sites/default/files/2026.09-Navigating.Transfer.in\_.WA-Executive.Summary.pdf 2\. North Carolina Community College System. NC Community College System, NCICU Launch Statewide Transfer Pathways Platform to Streamline Degree Completion. October 9, 2026. https://www.nccommunitycolleges.edu/news/ncicu-transfer-pathways/ 3\. North Carolina Independent Colleges and Universities. NCICU Transfer Pathways, public FAQ, especially the question about an evaluation that looks wrong. Accessed October 13, 2026. https://ncicu.org/transfer-pathways/ 4\. Sommo, C., Slaughter, A., Rubin, J., Woo, J., and Ortagus, J. C. Boosting Reenrollment and Graduation: Early Findings from the Re-Enrollment and Completion Team (REACT) Intervention. MDRC, October 2026, released October 6. See pp. 2–5 and 10–12 for the model, eligibility, sample and outcomes. https://www.mdrc.org/work/publications/boosting-reenrollment-and-graduation ; full report https://www.mdrc.org/work/publications/boosting-reenrollment-and-graduation/file-full 5\. EAB Briefing. AI is reshaping students’ college decisions—here’s how. Email newsletter dated October 12, 2026, reporting EAB’s 2026 survey of 11,700 students and recent graduates. Received-email evidence; methodology and item denominators not independently verified. 6\. Kukreja, A., and Deane, K. Piloting a Customized Chatbot to Increase College and Financial Aid Awareness. Washington Student Achievement Council, July 2026. Pilot design, not an outcome evaluation. https://wsac.wa.gov/sites/default/files/OtterBot.Pilot-Piloting.a.Customized.Chatbot.pdf 7\. Federal Student Aid. GENERAL-26-49, Guidance on FVT/GE Data Reporting, STATS Early Implementation, and Next Steps for Publication. Originally August 11, 2026; controlling update October 8, 2026. https://fsapartners.ed.gov/knowledge-center/library/electronic-announcements/2026-08-11/guidance-fvt/ge-data-reporting-stats-early-implementation-and-next-steps-publication-updated-oct-8-2026 8\. Federal Student Aid. GENERAL-26-62, Borrower Defense to Repayment Functionality Temporarily Unavailable for Schools. Originally October 7, 2026; restoration and deadline update October 9, 2026. https://fsapartners.ed.gov/knowledge-center/library/electronic-announcements/2026-10-07/borrower-defense-repayment-functionality-temporarily-unavailable-schools-updated-oct-9-2026 9\. Department of Homeland Security, U.S. Immigration and Customs Enforcement. Optional Practical Training Fees. Notice of proposed rulemaking, 91 FR 64566–64600, October 8, 2026, document 2026-20660, docket ICEB-2026-0100. Dates and submission cutoff: p. 64566. https://www.govinfo.gov/content/pkg/FR-2026-10-08/pdf/2026-20660.pdf *Practical exercises, fictional scenes and proposed local measures are original editorial recommendations, not validated interventions or legal advice. Readers should consult the linked primary material and the appropriate campus specialists for their own circumstances.* Get full access to DysArtisanal Innovations at dysartisanalinnovations.substack.com/subscribe