{"resourceId":"uh-shasta-chatbot-expansion-20260825","versions":[{"version":"external-a01afccbc88eff0da41d5194d7de5f4475ff07ebc722a13fc863222957c61bb9","resource":{"id":"uh-shasta-chatbot-expansion-20260825","title":"UH Shasta chatbot doubles department coverage with measured after-hours use and call diversion","organization":"University of Houston","sector":"Higher education / campus administration and IT","geography":"United States (Texas)","publishedAt":"August 25, 2026","publicationDate":"2026-08-25","eventDate":null,"sourceName":"University of Houston Administration and Finance","sourceLabel":"UH Administration and Finance press release","sourceUrl":"https://www.uh.edu/af/news/press-releases/releases-articles/2026/aug-26/shasta-chatbot.php","evidenceClass":"independent-reporting","outcomeClass":"effective","topics":["knowledge-work","operating-model","accessibility-workforce"],"finding":"UH reports Shasta expanded from 11 to 23 departments, handled 72,433 conversations with an 83% resolution rate, and that nearly 40% of conversations occur outside office hours. Student Business Services calls fell from 3,417 (Aug 2024) to 2,807 (Aug 2025). Leaders frame the bot as complementing staff, not replacing them, and UIT added web-content review to onboarding so crawls stay accurate.","sledRelevance":"Direct campus-operations evidence for multi-department service chatbots spanning bursar, card services, parking, libraries, HR and finance touchpoints, with local volume and after-hours access metrics.","evidence":"Institutional press release quoting UIT AI solutions manager and unit leaders; cites conversation counts, resolution rate, after-hours share, and year-over-year August call counts for Student Business Services. No independent audit of resolution quality or net FTE change.","architectureImplications":"Interpretation: multi-phase onboarding with website crawl readiness, department content optimization, and integration with campus platforms matter as much as the model. Measure channel diversion by unit, not only aggregate chat volume.","governanceImplications":"Interpretation: expansion without published accuracy audits or escalation SLAs still needs content ownership and human backup for unresolved chats. Treat resolution rate as operator-defined, not verified outcome.","securityPrivacyImplications":"Personalized residency, FAFSA, and balance queries imply sensitive student data pathways; the release does not detail data retention, vendor controls, or FERPA handling.","caveats":"Self-reported institutional metrics; August call drop may reflect other factors; resolution definition unpublished; no workforce displacement or accessibility evaluation.","streamIds":["campus-operations"],"roles":{"sales":"Campus service leaders need after-hours coverage and call diversion without claiming headcount cuts. UH reports Shasta growing from 11 to 23 departments, an 83% resolution rate, nearly 40% of conversations outside office hours, and a bursar August call drop from 3,417 to 2,807. Use those local metrics as a bounded discovery story for multi-unit chatbot programs. Ask which units own content quality, how resolution is defined, and how unresolved chats escalate. Do not sell guaranteed FTE savings or FERPA-safe personalization from this press release alone.","engineering":"Fit is a multi-tenant campus chatbot with crawlable departmental web content and SIS integrations for personalized residency, aid, and balance answers. Prerequisites include department content-review gates, identity for personalized queries, and a measurable human escalation path. Constraints are operator-defined resolution metrics, a one-month bursar call comparison, and thin vendor data-flow detail in the release. Personalized financial-aid data raises FERPA, retention, and processor-control requirements that must be designed locally rather than inferred from UH marketing language.","delivery":"Work is phased department onboarding, website content remediation so crawls stay accurate, transcript review, and per-unit diversion dashboards against call and email baselines. Ownership sits with UIT AI solutions plus a named content owner in each department. Train authors to write crawl-ready FAQs and agents to handle the unresolved share. Acceptance should include after-hours share, sampled accuracy of resolved chats, and published escalation SLAs—not aggregate conversation counts alone. Risks include content drift, over-claiming resolution quality, and sensitive-data exposure."},"roleDetails":{"sales":{"customerProblem":"Students need around-the-clock answers on bursar, aid, and campus services while staff capacity is constrained during peaks.","stakeholders":["CIO/UIT","Bursar/Student Business Services","Auxiliary and card services","Department content owners"],"discoveryQuestions":["Which peak call types are candidates for deflection?","Who owns website content quality for each department?","How is resolution defined and sampled?","What is the human escalation path after hours?"],"valueHypothesis":"A multi-department chatbot with content-review onboarding can divert routine inquiries and improve after-hours access, as UH reports locally, without proving net FTE reduction.","potentialEngagement":"Bounded pilot: onboard 2 to 3 high-volume service units with baseline call counts, content remediation, and 90-day diversion sampling.","unsupportedClaims":["Guaranteed campus-wide FTE savings","Verified accuracy equal to the stated resolution rate","FERPA-safe personalization without further design"]},"engineering":{"fit":"Campus service chatbot spanning public FAQ and authenticated personalized queries across many departments.","architectureIntegration":"Crawlable departmental web content plus integrations to campus systems; phased onboarding with content review gates.","prerequisites":["Department content owners","Baseline channel metrics","Escalation workflow","Identity for personalized answers"],"constraints":["Resolution rate is self-defined","August call comparison is one unit and one month","Vendor and data-flow details thin in the press release"],"security":"Personalized financial and residency data require documented FERPA, retention, and vendor processing controls not evidenced here.","proofOfValue":["Replicate August-style call baselines per unit","Sample unresolved chats for accuracy","Track after-hours share without treating it as outcome quality"]},"delivery":{"work":"Phase department launches, remediate web content for crawl quality, train staff on escalation, and instrument diversion dashboards.","dependencies":["UIT AI platform/vendor","Department website owners","Identity/SIS access for personalization"],"ownership":"UIT AI solutions manager with named departmental service owners.","skillsAdoption":"Content authors must write crawl-ready FAQs; agents must handle escalations the bot cannot resolve.","governanceCheckpoints":["Pre-launch content review","Periodic transcript accuracy review","Expansion gate before next phase"],"acceptanceCriteria":["Documented baseline vs post call volume for pilot units","Published escalation SLA","Sampled accuracy review of resolved chats"],"risks":["Content drift after launch","Over-claiming resolution quality","Sensitive data exposure via personalization"]}},"retrievedAt":"2026-09-15T18:48:20Z","enrichedAt":"2026-09-15T18:57:40Z","enrichmentBasis":"retrieved source","accessibilityWorkforceImplications":"After-hours access may help students outside business hours; the release does not assess disability accessibility of the chatbot UI or impacts on frontline staffing beyond complementarity claims.","procurementImplications":"Buyers should require measurable diversion KPIs and data-processing terms rather than department-count growth alone.","operatingModelImplications":"Phased multi-department rollout with central UIT enablement and local content ownership is the operating pattern; expansion depends on content readiness, not only licenses.","sourceVerification":{"openedUrl":"https://www.uh.edu/af/news/press-releases/releases-articles/2026/aug-26/shasta-chatbot.php","referenceExcerpt":"Since its launch, the chatbot has handled 72,433 conversations, generating 112,104 responses and achieving an 83% resolution rate.","promptVersion":"sled-research-v4.0","model":null,"basis":"agent-reported inspection"}}}]}