Prompts • Skills • Connectors • Agents • School workflows • Tools
AI Playbook
for K-12 School Administrators
For heads of school, business managers, and admissions and advancement staff at private, charter, and faith-based schools. A practical guide to running the school with AI: admissions and enrollment, tuition and financial aid, budgeting and reporting, fundraising and alumni relations, board reporting, and compliance.
Who this playbook is for
This playbook is for the people who run the school, not for classroom teaching. It covers the business, enrollment, advancement, and governance side of private, charter, and faith-based schools. It doesn't cover lesson planning, grading, or student AI use, and it isn't written for public school districts or for parents.
- Heads of school and their offices
- CFOs, business managers, and controllers
- Admissions, enrollment, and financial aid staff
- Advancement, alumni relations, and development staff
- Charter school operations and compliance leads
- Teachers, lesson planning, grading, or curriculum
- Student use of AI (see your academic AI policy)
- Public school districts
- Parents
Start With Your Role
Choose the role closest to your day-to-day work. Start with one useful prompt, make it repeatable, then add the information it needs.
- Start with: head of school report or strategic plan progress
- Save as a skill: board report format and dashboard metrics
- Connect: board folder, dashboard data
- First agent: board packet assembler
- Measure: packet prep time, board meeting focus
- Start with: budget vs. actual narrative or tuition setting analysis
- Save as a skill: reporting format, ratios, assumptions
- Connect: accounting exports, enrollment forecast
- First agent: monthly finance narrative drafter
- Measure: close-to-report days, forecast accuracy
- Start with: funnel analysis or visit follow-up
- Save as a skill: facts sheet, follow-up standards, committee criteria
- Connect: admissions system reports
- First agent: admissions follow-up coordinator
- Measure: conversion, yield, follow-up time
- Start with: aid budget model
- Save as a skill: aid policy, bands, letter templates
- Connect: aggregate aid reports only
- First agent: financial aid budget monitor
- Measure: budget variance, consistency exceptions
- Start with: annual fund plan or donor segments
- Save as a skill: constituency goals, appeal formats, acknowledgment rules
- Connect: Raiser's Edge NXT connector or CRM exports
- First agent: annual fund progress watcher
- Measure: participation, retention, dollars
- Start with: compliance calendar or authorizer report draft
- Save as a skill: obligations list and evidence matrix
- Connect: policy library, evidence folders
- First agent: compliance calendar monitor
- Measure: missed deadlines, renewal readiness
100 AI Prompts for K-12 School Administrators
Ready-to-use prompts for admissions, re-enrollment, tuition, financial aid, budgeting, the annual fund, alumni and major gifts, board work, compliance, and operations. Copy one, add your school's context, and review the result.
Use IDs and aggregates instead of student or family details, verify every figure, and review the result before it reaches families, donors, or the board.
Prompts for inquiries, tours, applications, and yield. Use applicant IDs, not student names. Charter schools: follow your lottery and open-enrollment rules.
Shows the exact stage and grade where families drop out, so effort goes where it matters.
Risks: Small grades swing a lot. Control: Director of admissions reviews with context before acting.
Fast, consistent answers matter when families compare several schools at once.
Risks: Facts change each year. Control: Update the facts block every admission season.
Personal details from the visit set your follow-up apart.
Risks: AI may embellish. Control: Officer checks against notes before sending.
Most applications that stall are one or two items short. This focuses follow-up.
Risks: Status data may lag. Control: Confirm in your enrollment system before contacting families.
Consistent summaries help the committee compare files fairly.
Risks: Summaries can carry bias. Control: Committee reads full files; decisions stay with the committee.
Yield is where enrollment targets are won or lost. A planned sequence beats ad hoc outreach.
Risks: Over-contacting can feel like pressure. Control: Limit touchpoints and respect 'not this year'.
Clear lottery communication protects fairness and trust in open enrollment.
Risks: Charter admissions are regulated. Control: Compliance lead and authorizer rules govern; counsel reviews changes.
Well-run events convert inquiries to applications.
Risks: Student volunteers need supervision and consent. Control: Follow your student participation policy.
Shows where to invest relationships and marketing.
Risks: Demographic data may be dated. Control: Verify sources.
Families judge schools by the website before they visit.
Risks: Copy must match policy. Control: Admissions director approves changes.
Learning from lost admits improves next year's yield.
Risks: Families soften reasons. Control: Pair with personal conversations.
12 Claude-Ready School Administration Skills
These are downloadable Claude Skill packages, not just longer prompts. Each includes a SKILL.md your team can tailor with your policies. A copy-and-paste version works in ChatGPT and other AI tools.
Connect Your Work
AI can only help with the information it can see. Start with one system, see what it can do, then add more when you are ready.
Easy start
Start with a limited, low-risk connection your team can test quickly.
- Access: Blackbaud's native connector for Claude, available to Raiser's Edge NXT customers on Claude Team and Enterprise plans
- Useful for: donor questions, appeal segments, and drafting with records your role can already see
- Setup & limit: access follows Blackbaud role permissions; start with read and draft tasks and keep gift entry with advancement services
- Access: Gmail, Drive, and Calendar connections in Claude and ChatGPT
- Useful for: policy search, board folders, and drafting from templates
- Setup & limit: start with administrative folders; keep student records out of scope
- Access: Outlook, SharePoint, OneDrive, and Teams connections, or Microsoft Copilot in your tenant
- Useful for: board packets, policies, and finance documents
- Setup & limit: your Microsoft admin controls access; start with one SharePoint site
- Access: Canva's connection for AI assistants
- Useful for: admissions viewbooks, appeal graphics, and event materials
- Setup & limit: use real student photos only with consent; don't generate images of real students
Needs an administrator
Useful, but someone needs to set access and permissions first.
- Access: Finalsite (Ask AI family chatbot and others), Veracross, Blackbaud, and FACTS are adding AI features inside their products
- Useful for: answering family questions from your website, and working inside the system you already use
- Setup & limit: review what data each feature sees, training settings, and admin controls before enabling
- Access: QuickBooks, Sage Intacct, and Blackbaud Financial Edge NXT; AI access varies by product and plan
- Useful for: budget vs. actual questions and variance narratives
- Setup & limit: read-only reporting role; keep postings and payments with the business office
- Access: official Slack and Notion connections for Claude and ChatGPT
- Useful for: internal project notes and the administrative knowledge base
- Setup & limit: keep HR, student, and family channels out of scope
Integration project
Plan the use case, source data, permissions, and owner before connecting.
- Access: no official direct Claude or ChatGPT connection is confirmed here for Veracross, Blackbaud SIS, FACTS SIS, or Ravenna
- Useful for: scheduled, de-identified exports (funnel counts, enrollment by grade) into a shared folder
- Setup & limit: strip student names and records; name an owner and review every export field
- Access: FACTS, Blackbaud Tuition Management, TADS, SSS, and Clarity hold family financial data
- Useful for: aggregate reports (aid by band, billing totals) for modeling
- Setup & limit: family financial data never goes into general AI tools
- Access: a curated, view-only dataset for enrollment, finance, and advancement metrics
- Useful for: board dashboards and multi-year models
- Setup & limit: an IT project; define fields, access, and logging first
Good to know
Limits and honest gaps to keep in mind.
- Access: transcripts, discipline, health, learning support, and counseling records
- Useful for: nothing in this playbook; these stay in your SIS and are governed by your privacy policy and applicable law (FERPA for charter schools and other federally funded schools)
- Setup & limit: administrative AI work uses aggregates and IDs only
- Access: tools for teachers and students are covered by your academic AI policy
- Useful for: not covered here; this playbook is for administrators
- Setup & limit: keep administrative and classroom AI policies separate but consistent
Tools
Choose the AI tools and school software that fit your team. The goal is to make your existing systems easier to use, not replace everything at once.
School Information Systems
5Admissions & Enrollment
5Tuition & Financial Aid
6Advancement & Alumni
7Finance & Planning
5Board & Governance
4Operations
512 AI Agents for K-12 School Administrators
An agent is an assistant for one recurring job. It gathers the right information, follows your process, and prepares the work for someone to review and decide.
A good first agent handles one task your office already does, reads only approved reports, never contacts families or donors, and hands its work to a named person.
# Admissions Follow-Up Coordinator ## Job to be done Make sure every inquiry and visit gets timely, personal follow-up. ## When it runs Daily during admission season; when a visit is logged. ## Approved context Inquiry and visit records (applicant IDs), facts sheet, templates. ## Operating loop Find families due for follow-up; draft personal notes; flag stalled files and missing items; summarize the day. ## What it delivers A daily follow-up worklist with drafts. ## Human owner Admissions officers personalize and send. ## What it must never do Never send messages, change application status, or comment on admission likelihood. ## How to measure it Follow-up within 24 hours, inquiry-to-application rate, completed files. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Application Completeness Monitor ## Job to be done Get every file complete before the committee deadline. ## When it runs Weekly from application open; daily in the final 2 weeks. ## Approved context Application checklist status (IDs), recommendation tracking, deadlines. ## Operating loop List incomplete files by missing item; draft family and recommender reminders; prepare committee-ready status. ## What it delivers A completeness report and reminder drafts. ## Human owner Admissions operations coordinator. ## What it must never do Never contact families or recommenders directly, or summarize file contents for decisions. ## How to measure it Files complete by deadline, days to completion. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Enrollment & Re-Enrollment Tracker ## Job to be done Give leadership a daily view of enrollment against budget in peak season. ## When it runs Daily from contract release through the deadline; weekly otherwise. ## Approved context Contract status, enrollment by grade, budget, aid status, balances. ## Operating loop Update enrollment vs. budget by grade; list families yet to return contracts; flag families to check in with. ## What it delivers A dated enrollment snapshot and check-in list. ## Human owner Director of enrollment and head of school. ## What it must never do Never contact families or assume reasons for not re-enrolling. ## How to measure it Re-enrollment rate by deadline, budget variance, attrition. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Third-Party Payment Reconciler ## Job to be done Make sure ESA, voucher, and scholarship-organization payments land on the right accounts. ## When it runs When program payments arrive; before monthly statements. ## Approved context Award records, payments received, billing, program rules. ## Operating loop Match expected vs. received by student ID; flag unapplied or missing funds and possibly non-allowable charges; update family balance view. ## What it delivers A reconciliation report for the business office. ## Human owner Business office specialist. ## What it must never do Never post payments, change charges, or contact program administrators. ## How to measure it Unapplied funds, statement corrections, days to apply payments. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Past-Due Account Watch ## Job to be done Catch balances early and route them to the right conversation. ## When it runs Weekly; when an account crosses a policy threshold. ## Approved context AR aging, payment plans, aid status, collections policy. ## Operating loop Identify next policy step per account; flag aid families for personal outreach; draft messages. ## What it delivers A collections worklist. ## Human owner Business manager. ## What it must never do Never send messages, apply fees, or hold enrollment or records. ## How to measure it Balances over 60 days, collection rate, write-offs. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Financial Aid Budget Monitor ## Job to be done Keep aid commitments within budget and consistent with policy during award season. ## When it runs After each committee meeting; weekly during award season. ## Approved context De-identified award data, aid budget, policy. ## Operating loop Track committed vs. budget by division; check consistency flags; project remaining capacity. ## What it delivers An aid budget status and consistency flags for the committee. ## Human owner Financial aid director and committee. ## What it must never do Never make, change, or communicate awards. ## How to measure it Budget variance at season end, consistency exceptions resolved. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Monthly Finance Narrative Drafter ## Job to be done Cut the time to produce monthly and board financial narratives. ## When it runs Monthly after close. ## Approved context Approved financials, budget, prior year, enrollment, cash. ## Operating loop Compare to budget; identify variances; draft narrative with possible reasons; update full-year projection; list questions. ## What it delivers A draft narrative and question list. ## Human owner CFO or business manager. ## What it must never do Never distribute reports or change budgets. ## How to measure it Prep time, edits needed, on-time board packets. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Annual Fund Progress Watcher ## Job to be done Keep the annual fund on pace and highlight segments to re-engage. ## When it runs Weekly; daily in the final month of the fiscal year. ## Approved context Gift data by donor ID and constituency, goals, appeal calendar. ## Operating loop Track progress vs. goal and last year; identify LYBUNT donors and lagging constituencies; draft appeal adjustments. ## What it delivers A progress report and segment lists. ## Human owner Director of advancement. ## What it must never do Never contact donors or change gift records. ## How to measure it Participation rate, dollars vs. goal, donor retention. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Major Gift Pipeline Reviewer ## Job to be done Keep major gift relationships moving with clear next steps. ## When it runs Biweekly before the advancement team meeting. ## Approved context Pipeline records (prospect IDs), contact reports, proposals outstanding. ## Operating loop Find prospects without next steps or with stalled stages; balance portfolios; prepare meeting agenda. ## What it delivers A pipeline review packet. ## Human owner Director of advancement and gift officers. ## What it must never do Never contact prospects, update stages, or research beyond approved sources. ## How to measure it Prospects with next steps, proposals outstanding, gifts closed. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Board Packet Assembler ## Job to be done Deliver a complete, consistent board packet draft on time. ## When it runs Two weeks before each board meeting. ## Approved context Committee reports, dashboard data, head report notes, prior minutes, agenda template. ## Operating loop Collect inputs; draft agenda and dashboard narrative; list missing reports; draft head report outline. ## What it delivers A board packet draft and missing-items list. ## Human owner Head of school and board chair. ## What it must never do Never distribute materials or include executive session content. ## How to measure it Packet delivered on time, prep hours, missing items at deadline. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Compliance Calendar Monitor ## Job to be done Make sure no filing, notice, renewal, or report deadline is missed. ## When it runs Weekly; 60 and 30 days before each deadline. ## Approved context Compliance calendar, evidence inventory, policy review dates. ## Operating loop Find upcoming deadlines; check evidence status; draft reminders to owners; flag overdue items. ## What it delivers A compliance status report. ## Human owner Head of school or compliance lead. ## What it must never do Never file, publish notices, or certify compliance. ## How to measure it Missed deadlines, overdue policy reviews, audit findings. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
# Vendor Renewal & Contract Watch ## Job to be done Avoid auto-renewal surprises and prepare renewal decisions. ## When it runs Monthly; 90 days before each renewal or notice deadline. ## Approved context Contract register, spend data, satisfaction notes. ## Operating loop List upcoming renewals and notice dates; summarize cost trends; draft negotiation questions. ## What it delivers A renewal watch list. ## Human owner Business manager. ## What it must never do Never renew, cancel, or negotiate. ## How to measure it Unplanned renewals, savings identified, notices sent on time. ## Operating rules 1. Use only approved sources and follow the linked school skill. 2. Cite source records and separate facts from assumptions. 3. Route policy exceptions, material variances, and sensitive matters to the human owner. 4. Keep a record of trigger, inputs, output, reviewer, decision, and override. 5. Stop and ask for human review when policy, law, or approval authority is unclear. 6. Never handle student records, family financial data, or personnel details beyond what the owner approved.
AI for Admissions & Enrollment
Use AI to analyze the funnel, follow up faster, and prepare committee materials. Admission decisions stay with people.
- Prompts: funnel analysis, visit follow-up, file completeness
- Skills: admissions funnel & follow-up; admission committee preparation; enrollment forecasting
- Connect: admissions system reports (Ravenna, Finalsite Enrollment, Blackbaud, Veracross), calendar
- Agent: admissions follow-up coordinator; application completeness monitor
- What AI does: Conversion by stage, grade, and source
- Helps: Fix the right step
- Control: Watch small numbers
- What AI does: Drafts notes from visit details
- Helps: Stand out among schools
- Control: Officers personalize and send
- What AI does: Tracks missing items and drafts reminders
- Helps: More complete files by deadline
- Control: Confirm in system first
- What AI does: Consistent summaries against criteria, no ranking
- Helps: Fairer, faster committee work
- Control: Committee reads full files
- What AI does: Plans touchpoints by family segment
- Helps: Higher yield
- Control: Respect families' pace
- What AI does: Projects enrollment by grade with scenarios
- Helps: Staffing and budget decisions
- Control: Leadership approves
Admissions Implementation Checklist
WorkflowBefore You Start
- Export funnel data by grade and source (IDs)
- Write committee criteria
- Build a facts sheet for templates
- Confirm charter lottery rules if applicable
- Baseline: follow-up time, conversion
After You Start
- Review drafts before sending
- Audit committee summaries for consistency and bias
- Track conversion monthly
- Update templates each season
- Survey admitted and lost families
- Decisions: AI never ranks, scores, or recommends admission. The committee decides.
- Fairness: Never include or infer protected characteristics.
- Charter schools: Admission follows the lottery and open-enrollment rules in your charter and state law.
- Privacy: Student records, testing, and recommendations stay in your admissions system.
- Accuracy: Tuition, aid, and deadlines must match the current facts sheet.
AI for Tuition & Financial Aid
Use AI to model tuition and aid, check billing, and reconcile third-party payments. Award decisions and family financial data stay with people.
- Prompts: tuition setting analysis, billing exceptions, aid budget model
- Skills: tuition billing & third-party payments; collections; financial aid budget & consistency
- Connect: tuition management reports (FACTS, Blackbaud Tuition Management, Veracross, TADS), aid summaries
- Agent: third-party payment reconciler; financial aid budget monitor
- What AI does: Compares cost per student, tuition, aid, and net revenue
- Helps: Clearer tuition decisions
- Control: Board decides
- What AI does: Finds discount, aid, and fee exceptions
- Helps: Fewer statement errors
- Control: Business office corrects
- What AI does: Reconciles ESA, voucher, and scholarship payments
- Helps: Accurate family balances
- Control: Verify program rules
- What AI does: Models demand, renewals, and discount rate
- Helps: Stay within budget
- Control: Committee decides
- What AI does: Checks de-identified awards against policy
- Helps: Fair, defensible awards
- Control: Committee reviews flags
- What AI does: Drafts award letters and FAQs
- Helps: Fewer confused calls
- Control: Amounts verified
Tuition & Aid Control Checklist
ControlsBefore You Start
- Document tuition, discount, and fee rules
- Collect third-party program rules
- Write aid policy and appeal process
- Define de-identified bands for analysis
- Baseline: statement errors, aid budget variance
After You Start
- Verify every exception in the system
- Reconcile third-party payments monthly
- Review consistency flags with committee
- Track discount rate
- Survey families on clarity
- Do NOT: Paste tax returns, income, or need analysis reports into general AI tools.
- Do NOT: Make, change, or communicate aid awards.
- Do NOT: Decide refunds or hold records without the business manager.
- Do NOT: Apply program funds without verifying current rules.
- Do: Keep need analysis in your provider (SSS, Clarity, FACTS Grant & Aid, TADS) and use de-identified bands for modeling.
AI for Budgeting & Financial Reporting
Use AI to draft narratives, build models, and prepare for audits. Finance leaders own the numbers.
- Prompts: budget vs. actual narrative, multi-year model, audit prep
- Skills: school budget & financial reporting
- Connect: accounting system exports, budget model, enrollment forecast
- Agent: monthly finance narrative drafter
- What AI does: Drafts monthly explanations from approved figures
- Helps: Faster reporting
- Control: CFO verifies reasons
- What AI does: Builds revenue and expense frameworks with ratios
- Helps: Focus on decisions
- Control: Finance committee reviews
- What AI does: Projects 5 years with sensitivities
- Helps: See structural gaps early
- Control: Stress-test assumptions
- What AI does: Maps tuition timing against payroll
- Helps: Avoid cash surprises
- Control: Compare to actuals
- What AI does: Builds PBC lists and schedules
- Helps: Shorter fieldwork
- Control: Confirm with auditors
- What AI does: Tracks restrictions, draws, and availability
- Helps: Honor donor intent
- Control: Counsel on restrictions
Finance Implementation Checklist
ControlsBefore You Start
- Close books on a consistent calendar
- Export budget and actuals by category
- Document endowment spending policy
- List board-required metrics
- Baseline: reporting hours
After You Start
- Reconcile every AI number to the GL
- Mark AI drafts clearly until reviewed
- Track edits needed
- Review models quarterly
- Brief the auditor on AI use
- Accuracy: Every figure ties to the general ledger.
- Explanations: Variance reasons are hypotheses until confirmed.
- Restrictions: Donor restrictions and spending policy govern endowment use.
- Charter schools: Public reporting and procurement rules apply.
- Approvals: Budgets and tuition are board decisions.
AI for Fundraising & Alumni Relations
Use AI to segment donors, draft appeals, prepare briefs, and steward gifts. Relationships stay human.
- Prompts: annual fund plan, donor segments, prospect brief
- Skills: annual fund & campaign support; alumni & major gift preparation
- Connect: Raiser's Edge NXT via Blackbaud's Claude connector, or your CRM exports
- Agent: annual fund progress watcher; major gift pipeline reviewer
- What AI does: Groups donors by giving pattern and constituency
- Helps: Better-targeted appeals
- Control: Clean CRM data
- What AI does: Writes appeals by constituency
- Helps: Faster, more relevant asks
- Control: Advancement approves
- What AI does: Drafts timely, specific thanks
- Helps: Higher retention
- Control: Tax language verified
- What AI does: Summarizes sourced, public information
- Helps: Better-prepared visits
- Control: Ethical research only
- What AI does: Drafts tailored major gift proposals
- Helps: Reflect donor interests
- Control: Officer owns the ask
- What AI does: Plans reunions, stories, and class agent toolkits
- Helps: Lifelong connection
- Control: Consent for stories
Advancement Implementation Checklist
WorkflowBefore You Start
- Clean CRM data and constituency codes
- Set goals by constituency
- Write prospect research ethics guidelines
- Confirm acknowledgment language
- Baseline: participation, retention
After You Start
- Review all donor-facing drafts
- Track participation and retention
- Audit briefs for sources
- Steward every major gift
- Report to the board quarterly
- Ethics: Prospect research uses public and school-held information only.
- Privacy: Donor data stays in your CRM; share minimum necessary.
- Honesty: Never invent impact or overstate how gifts will be used.
- Restrictions: Honor every gift restriction.
- Relationships: Gift officers and the head own every ask.
AI for Board Reporting & Governance
Use AI to assemble packets, draft reports and minutes, and track strategic goals. Boards and heads of school make the decisions.
- Prompts: board agenda, head of school report, dashboard narrative
- Skills: board packet preparation
- Connect: shared board drive, dashboard data, strategic plan
- Agent: board packet assembler
- What AI does: Builds timed agendas with Inform/Discuss/Decide
- Helps: More strategic meetings
- Control: Chair approves
- What AI does: Drafts consistent reports to the board
- Helps: Saves the head hours
- Control: Head edits candidly
- What AI does: Narrates key indicators and trends
- Helps: Focus on what matters
- Control: CFO verifies
- What AI does: Drafts minutes from notes
- Helps: Timely, consistent records
- Control: Secretary verifies
- What AI does: Summarizes progress by goal
- Helps: Keep the plan alive
- Control: Leadership agrees on status
- What AI does: Orientation guides and self-assessments
- Helps: Stronger trustees
- Control: Governance committee owns
Board Reporting Checklist
WorkflowBefore You Start
- Standardize dashboard metrics
- Collect bylaws and committee charters
- Set packet deadlines
- Define executive session handling
- Baseline: packet prep hours
After You Start
- Verify all figures
- Keep personnel and student matters out of AI
- Board approves minutes
- Review dashboard metrics yearly
- Survey trustees on packet usefulness
- Confidentiality: Executive session, personnel, and student matters stay out of AI tools.
- Records: Minutes are legal records; the secretary verifies.
- Charter schools: Open meetings and public records laws apply to board materials.
- Accuracy: Dashboards tie to approved financials.
- Roles: AI supports the head and board; it doesn't advise on governance disputes.
AI for Compliance & Risk
Use AI to build calendars, map evidence, and draft policy updates. Counsel, accreditors, and authorizers are the authorities.
- Prompts: compliance calendar, policy review, accreditation organizer
- Skills: compliance & policy management
- Connect: policy library, compliance calendar, evidence folders
- Agent: compliance calendar monitor
- What AI does: Organizes filings, notices, and renewals
- Helps: No missed deadlines
- Control: Counsel confirms the list
- What AI does: Finds gaps and drafts updates
- Helps: Current policies
- Control: Counsel approves
- What AI does: Maps standards to evidence
- Helps: Organized self-study
- Control: Accreditation lead reviews
- What AI does: Maps systems holding student and family data
- Helps: Know where data lives
- Control: Review vendor agreements
- What AI does: Structures risks, controls, and owners
- Helps: Better board risk discussion
- Control: Leadership rates risks
- What AI does: Organizes authorizer requirements and evidence
- Helps: Reduce renewal risk
- Control: Board reviews
Compliance Implementation Checklist
ControlsBefore You Start
- List obligations by school type and state
- Collect policies and review dates
- Inventory systems with student data
- Gather accreditation or authorizer requirements
- Name a compliance owner
After You Start
- Review calendar monthly
- Track policy review dates
- Confirm vendor privacy agreements
- Update risk register each semester
- Report to the board annually
- Authority: Counsel, accreditors, and authorizers decide what compliance means.
- Private schools: Annual nondiscriminatory policy notice requirements apply; confirm with counsel.
- Charter schools: FERPA, open meetings, public records, and authorizer reporting apply.
- Faith-based schools: Religious exemption questions go to counsel.
- Honesty: Never create or alter evidence after the fact.
Set Up & Get Running
If you already use prompts, here is how to turn one into a repeatable skill and then a useful first agent. Start with a task that already works.
From prompt to first agent
- Choose a task you do every month or season
- Use real, de-identified data
- Good first choices: budget variance narrative, funnel analysis, or board agenda
- Write down inputs, output, and reviewer
- Try it on three real cycles
- Improve it when results are unclear
- Start with reports your reviewer trusts
- Use vendor connectors (e.g., Raiser's Edge NXT) with role permissions
- Keep student and aid data out
- Start with a monitoring task
- Set when it runs; a person checks every output
- Track time saved and edits
It's useful, low-risk, and easy to check. It compares actuals to budget after close and drafts the narrative and question list. It doesn't change budgets, distribute reports, or touch student data.
Copy these builders
Use the first after a prompt works. Use the second only after the skill has been tested on real work.
Days 1-30: Foundation
- Head of school names an AI lead for administration
- Write the 'never put into AI' list
- Set up business accounts with training off
- Review AI features in SIS, CRM, and tuition systems
- Pick 1 pilot (finance narrative or funnel analysis)
- Record baseline hours
Days 31-60: Expand
- Save the pilot as a skill; test 3 cycles
- Add a 2nd workflow (board agenda or annual fund segments)
- Turn on one vendor connector with role permissions
- Draft the administrative AI policy
- Measure hours saved and edits
- Share a prompt library across offices
Days 61-90: Standardize
- Add a 3rd workflow (compliance calendar or re-enrollment tracker)
- Head of school or board adopts policy
- Pilot one review-only agent
- Write SOPs
- Report results to leadership
- Plan the next wave
Implementation Success Measures
Measurement30-Day Targets
- Staff follow 'never put into AI'
- Pilot running with review
- Baseline recorded
- No privacy incidents
60-Day Targets
- Pilot time down noticeably
- Second workflow live
- One connector in use safely
- Draft policy reviewed
90-Day Targets
- 3 workflows with SOPs
- Policy adopted
- First agent piloted
- Time redirected to families, donors, and staff
- Decision on next wave
How the Pieces Fit Together
Start with a prompt. Save the ones that work. Add the systems and files they need. Then use an agent for recurring work, with someone checking the result.
- Ask AI to do one clear task
- Use it for a quick draft, analysis, or explanation
- Check the result before you use it
- Save the step-by-step instructions for a task your team repeats
- Add your policies, examples, and review steps
- Update it when your process or regulations change
- Bring the right systems, files, and reports into the task
- Stop copying and pasting information by hand
- Start read-only, with only the sources the task needs
- Run a skill on a schedule or when something happens
- Gather the information, prepare the work, and flag issues
- Leave decisions about students, families, money, and donors with people
Where AI Can Help School Administrators
Use AI to get routine school administration into a clear first draft. Keep decisions about students, families, money, and people with your team.
- Funnel analysis and follow-up drafts
- Enrollment and re-enrollment forecasts
- Billing and third-party payment checks
- Aid budget modeling
- Variance narratives and multi-year models
- Board packets and head reports
- Audit preparation
- Strategic plan tracking
- Appeals, segments, and acknowledgments
- Prospect briefs and proposals
- Compliance calendars and evidence
- Policy drafts
- Admission and aid decisions
- Tuition, budget, and policy approval
- Donor relationships
- Personnel and student matters
AI Capabilities Explained
No jargon. What each kind of AI does, and what it looks like in school administration.
ChatGPT, Claude, Gemini, and Copilot write, summarize, and follow instructions.
At school: appeal letters, tuition announcements, board report drafts, policy drafts.
AI that can read approved systems through a connector.
At school: asking Claude about donor segments through the Raiser's Edge NXT connector.
Projects future numbers from patterns and assumptions.
At school: enrollment by grade, aid demand, 5-year financial models.
Compares records to policies and lists exceptions.
At school: billing vs. tuition rules; de-identified aid awards vs. policy.
Answers questions from documents you supply.
At school: 'What does our accreditor's standard 4 require?'
Answers family questions from approved website content.
At school: admissions and business office FAQs after hours.
Guardrails: Student Privacy, Fairness & Stewardship
Simple rules for administrative AI use at a school. Families trust you with their children and their finances; donors trust you with their gifts.
- Student names with grades, discipline, health, or learning support details
- Family income, tax returns, or aid files
- Employee personnel files or SSNs
- Executive session or confidential board matters
- AI never ranks or recommends admission or aid
- No use of protected characteristics
- Committees read full files and decide
- Document exceptions and who approved them
- Every figure ties to the general ledger
- Donor restrictions are honored
- Tuition and budgets are board decisions
- Prospect research is ethical and sourced
- Business or enterprise accounts, training off
- Review AI features in SIS, CRM, and tuition systems
- Signed privacy agreements with vendors
- Remove access when staff leave
- Charter: FERPA, lotteries, open meetings, public records
- Private: nondiscriminatory policy notice requirements
- Faith-based: religious exemption questions to counsel
- All: state law and accreditor standards
- Admission, aid, and discipline decisions
- Tuition, budget, and policy approval
- Donor relationships and asks
- Communication with regulators and authorizers
School AI Readiness Checklist
ControlsLeadership & Policy
- Head of school sponsors administrative AI use
- Administrative AI policy adopted (separate from academic policy)
- Approved tools listed with training off
- Vendor AI features reviewed
- Staff trained on 'never put into AI'
Practice & Review
- Human review on all family, donor, and board materials
- Exports de-identified
- Committee decisions documented without AI ranking
- Quarterly review of AI use
- Access removed when staff leave
Scope: This policy covers AI use by administrative staff (business office, admissions, financial aid, advancement, operations, and the head of school's office). Classroom and student AI use is governed by our academic AI policy.
Approved tools: [e.g., Claude Team with the Blackbaud connector; Microsoft Copilot in our tenant; reviewed AI features in our SIS and website platform]
Approved uses: drafting communications, analyzing de-identified or aggregate data, preparing budgets and board materials, drafting policies, and organizing compliance evidence.
Never enter into AI tools: student records (grades, discipline, health, learning support, counseling), family financial and aid data, employee personnel information, SSNs, or confidential board matters.
Decisions: AI does not rank or recommend admission, aid, hiring, or discipline decisions.
Human review: A named staff member reviews everything sent to families, donors, the board, or regulators.
Training & review: Annual training; policy reviewed each year.
Approval: [Head of School / Board], [date]
AI Maturity Model for School Administration
See where your administrative teams are today. Choose a realistic next step.
Curious + Ad Hoc
- Individuals use AI on their own
- No administrative AI policy
- Vendor AI features unreviewed
- No shared prompts
Shared Prompts & Rules
- 'Never put into AI' list in place
- Business accounts
- Prompt library by office
- Human review on all outputs
Embedded Workflows
- Saved skills for finance, admissions, advancement, and board work
- Adopted policy
- Vendor connectors with role permissions
- Quarterly review
Connected & Measured
- Review-only agents with owners
- Governed data for dashboards
- Time measurably redirected to people
- Annual policy review
Maturity Self-Assessment
AssessmentLeadership
- AI lead named
- Policy adopted
- Staff trained
- Vendor AI reviewed
Practice
- Prompt library
- Saved skills
- SOPs
- De-identified exports
Protection
- No student or aid data in AI
- No AI ranking of decisions
- Human review
- Access removal
Measurement
- Baseline
- Time saved
- Errors caught
- Quarterly review