Applied AI means the admissions office answers every inquiry the same day, the business office stops re-keying one family into four systems, and the head of school gets one honest answer about a student instead of five partial ones. We train your people on their own tasks, put a trained lead in each office, and build the rest into the tools you already pay for. A person approves anything that reaches a family.
Works with FACTS, Schoology, HMH, GoHighLevel, Stripe, and Cloudflare.
Here's what we find when we walk into a school. A teacher writes the newsletter in a free assistant on a personal account. The admissions coordinator pastes inquiries into a chatbot to "clean up the reply." Every vendor has bolted an AI assistant onto its portal that knows nothing about this school. There's an AI policy for students and none for staff. And the work AI is best at, the re-keying, the inquiry backlog, the month-end exceptions, the accreditation spreadsheets, is still done by hand by the people who have the least time.
Answers the questions the website already answers and sends everything else to a form. Families notice.Generic
Reads the inquiry, matches the grade to open seats, drafts the reply in the school's voice with two tour times, and waits for the director to press send.Applied
Useful once. Then it forgets, and so does everyone else.Generic
Vendor, term, fee basis, renewal date, notice period, and the clause that matters, with each renewal raised ninety days out for a keep, renegotiate, or retire decision.Applied
Perfectly reasonable and about nobody's child.Generic
Generated from the school's own curriculum and assessment data, reviewed by the teacher, and only then seen by a parent.Applied
Most schools start with one and add the others as the first one becomes routine. None of them require new software; all of them require a person who approves.
A working morning with one office at a time, built on that office's real material: last week's inquiries, the contract folder, the board packet. Each task gets done the old way and the new way, the team decides what to keep, and the office writes its own approval rule before anyone leaves. Not a webinar about what AI is.
One person in admissions, the business office, the front office, and communications, trained in more depth than the rest, with a direct line to us and a short weekly job: notice the next task worth trying, test it on one case, and show the team. It's the reason the training still holds a year later.
When a task deserves more than a prompt, we build it into the SIS, the CRM, the learning system, and email: inquiry triage, reteaching kits from your own standards and results, at-risk signals for the head of school, drafted parent communication, accreditation report prep. Grounded in your data, reviewed by a person, owned by the school.
Every inquiry read the moment it arrives, matched to open seats, and answered with a draft that offers two tour times. Post-tour follow-ups on day one, day seven, and two weeks before the deadline. A Monday list of families who've gone quiet. Tour notes dictated on the walk back and filed against the family record without typing.Read the admissions guide
One family record behind billing, lunch, the store, giving, and athletics, so data is entered once. Month-end becomes an exception list with each exception explained. A vendor contract table that raises renewals. Tuition questions drafted from the live balance. The board packet's tables assembled on schedule and its narrative drafted from the actual movement.Read the business office guide
The newsletter drafted from the week's calendar and the school's own voice, the allergy entered once and mirrored everywhere it's needed, the parent email that asks something the handbook answers drafted with the handbook's answer, and a person reading every one before it goes.
Signals instead of reports: which students' attendance, assessment results, and engagement moved in the same direction in the same month, surfaced weekly with the underlying numbers a click away. One question about a student, one answer, drawn from every system instead of three logins.
Reteaching material generated from the school's own standards and a student's actual results, in the school's phrasing, reviewed by the teacher before a parent sees it, so what a family sees at home matches what's taught in class.
Portals that replace form scripts and spreadsheets for candidacy, annual reports, and member directories, with drafted narrative sections and a staff review queue. Work we've already done for national associations.
Every system the school pays for, what data it holds, who enters it, and where it's re-entered, on one page, with a plain recommendation for what to keep, what to connect, and what to retire. Scope and price are agreed before we start.
The use cases, ranked by the hours they return and the risk they carry, with the approval rule and the approver named for each. Two to start, usually one in admissions and one in the business office, chosen because they save real time without touching a family or a dollar on their own.
The first use cases go live inside the tools the office already opens, and the office is trained on them with its own material in the room. Integrations where the systems need to share a record; custom tools where nothing fits; prompts and templates where that's enough.
A monthly look at what ran, what the approval step changed or rejected, the hours returned, and the next use case the office lead has found. The policy gets rewritten after the first semester from what actually happened.
The people on the call are the people who build and train. Nothing is handed to a junior team after the sale.
Code in the school's own repository, data in the school's own systems, tools under the school's own logins. If we disappeared tomorrow, everything keeps running and you can hand it to anyone.
Anything that reaches a family, touches money, or changes a student record waits for a named person's click. The approval log shows who accepted what and when.
The work shows up in the inbox, the CRM, and the SIS, not in a new destination people have to remember to visit. Adoption fails when AI is a separate place.
Your curriculum, your records, and your families' information are used to do your work and nothing else. Every vendor involved answers that question in writing before it sees a record.
We also won't rip out a system that's doing its job or run AI on generic data and call it personalized. If the honest answer after the audit is that your stack is fine, that's the answer you'll get.
Who want one answer about a student or a family instead of five partial ones, staff who answer families the same day, and a technology bill the board can understand.
Who are carrying a dozen renewals, re-keying every new family into four systems, and closing the month from a ledger instead of an exception list.
Who run accreditation, membership, and dues on forms and spreadsheets and know the office deserves better than that.
Five guides on applied AI for schools, written for the people who'll run it.
Generic against applied, where the hours go, and the three ingredients every use case needs.
Read the guide →Same-day replies, the follow-up sequence, the gone-quiet list, and what stays with a person.
Read the guide →One record, exceptions instead of ledgers, the contract table, tuition replies, and board packets.
Read the guide →Staff before students, the five rules, six vendor questions, and a one-page outline.
Read the guide →Sessions on real work, a lead in every office, the playbook the school keeps, and what to measure.
Read the guide →Start with the audit. Two weeks, a fixed fee, and every system your school pays for on one page. Whatever you decide to do next starts from an honest map.
Usually not. Most of the useful work connects the systems you already pay for and adds AI where the data meets. The audit comes before any purchase, and the honest answer is sometimes that the stack is fine and the connections aren't.
It stays in your systems and your accounts. Nothing about your students or families is used to train a model, and every vendor involved answers in writing where it runs and how long it keeps anything before it sees a record.
It drafts, and a person sends. Every use case keeps a named approver between the AI and the family, and the approval log shows who accepted what and when.
A working morning with one office, built on that office's real tasks in its own tools and accounts, ending with a written approval rule and a playbook the school keeps. One person per office gets more depth and owns it afterward.
The audit takes about two weeks. The first use cases, typically inquiry triage and a contract table, run within weeks of the plan because they need little integration. The connected record behind the business office takes longer and is where the re-keying disappears.
No. The audit, the training, and the build all run remotely, and the systems involved are the same ones schools use everywhere. On-site sessions are available where it helps.