-Sep-11-2026-04-21-22-8168-PM.png?width=45&height=45&name=Untitled%20design%20(10)-Sep-11-2026-04-21-22-8168-PM.png){kind=small}
The Background
When a child cannot read by the end of third grade, they are four times less likely to graduate high school. That number is why Ignite Reading exists. The company was built around the conviction that every student, regardless of zip code, income, or learning profile, deserves a trained, dedicated tutor who shows up every single school day to help them crack the code of reading.
Each student receives 15 minutes of daily 1-to-1 instruction from a tutor trained in evidence-backed foundational literacy. Those results are consistent across every demographic: students with IEPs, multilingual learners, students of color, and students receiving free or reduced-price lunch all achieve the same outstanding outcomes, averaging better than five months of additional reading growth per student.
Delivering that promise at scale, across 25 states and growing, requires every new district partnership to launch cleanly and quickly. The faster a district is onboarded, the sooner students are in sessions with their tutors. Managing that operational work was Matthew Ruxton’s responsibility — Ignite Reading’s Senior Manager of Revenue Operations — and one piece of it was taking far longer than it should.
The Challenge
For the students and schools depending on Ignite Reading, a delayed launch is real and immediate. It means a week or more of missed sessions — a child who needed a tutor on Monday still waiting while onboarding gets sorted out. Getting setup right determines when students start reading.
Every district that partners with Ignite Reading runs on a different school calendar. Different first days, different last days, different holidays, different breaks. Before tutoring can start, those non-program days have to be mapped and loaded into the scheduling tool, so tutors and students are never matched on a school holiday or break day. Multiple team members owned that process, and it was entirely manual.
A Process That Did Not Scale
For every new deal, a rep would search for the district, navigate to its website, find the right school-year calendar, open the PDF or website page, read it line by line, and hand-key each date into HubSpot. That took 15 to 20 minutes per district. And because every district formats its calendar differently, a missed holiday or misread date could mean a tutor session firing on a day no one was at school.
"Every district has a different calendar format. We were Googling each one, hunting through PDFs, and keying dates by hand. As we grew into more states, that process wasn’t built to scale," Matthew said.
As deal volume grew and Ignite Reading expanded into new states, the problem multiplied. Matthew could see exactly where the time was going, and the time was going somewhere that did not move deals or serve schools. Onboarding associates were reformatting data that arrived inconsistently from whoever had done the lookup. The work was slow, duplicative, and growing every season.
The Solution
For every district that signs on with Ignite Reading, getting to the first tutoring session is the moment that matters. The districts who have entrusted their struggling readers to the program are waiting. Matthew’s answer was to build the calendar lookup directly into the deal, so the data was there before anyone had to ask for it.
Matthew built a custom agent inside HubSpot himself, iterating on the prompt with an LLM and running tests until the logic held. Along the way, he found something that changed the scope of what the agent could do: a recently released beta feature that let the agent pull live properties straight from the deal record. That meant the school year property on the deal could drive the lookup automatically, making the agent reusable year over year with minimal edits.
Built on the Data Already There
The agent uses the associated company’s domain to find the district’s calendar, opens and parses the official page or PDF, extracts the first day, last day, and every break, and structures the output into the exact format their scheduling product expects. Matthew added a custom UI extension that exports the dates as an upload-ready CSV. The agent fires as part of the deal’s natural progression, so data lands precisely when the team starts scheduling.
"Our data was already in HubSpot, so building there was an easy call. I did not need outside developers or a separate tool. I described what I needed, and the agent built itself from the deal record," Matthew said.
Matthew chose to build on HubSpot because Ignite Reading’s data already lived there, an important factor for an organization with SOC 2 certification. Keeping the work in-house meant no customer data traveling between systems, no additional integrations to manage, and no added compliance risk. The decision to build inside the platform he already used removed every barrier between the idea and a working agent.
The Transformation
The students in Ignite Reading’s partner districts never see the agent. They have no idea their tutor showed up on the right day because Matthew’s agent parsed a calendar in seconds rather than someone researching it over 20 minutes. What they experience is a program that is ready when they are, a tutor who appears consistently, a schedule that holds.
Matthew made that possible. A lookup that took 15 to 20 minutes per district now completes in seconds. Based on last year’s deal volume, that is roughly an hour saved per district and an estimated 350 hours saved per year from this single automation.
From One Fix to a Company-Wide Efficiency Goal
The calendar agent counts toward a company-wide goal of saving over 2,000 hours this quarter through automation, and Matthew’s agent already accounts for a meaningful share of that target. Sales reps no longer touch web research at all. Onboarding associates move faster through fall and spring peaks because data arrives clean and structured rather than needing reformatting. Every district now gets the same structured treatment — clean, consistent, and handled the same way every time.
The agent opened up Matthew’s thinking about what is possible. As Ignite Reading scales beyond 25 states, the custom agents he builds are becoming what is actually making the team’s efficiency goals reachable, turning recurring manual tasks into automated steps inside the deal workflow. A fix for one process has turned into a blueprint for how he approaches automation.
"This one agent already counts toward our 2,000-hour savings goal for the quarter. And I am already thinking about what I build next," Matthew said.
Matthew is already identifying the next automation — driven by what he has seen work, not a top-down directive — because he now knows what it looks like when a good agent is built on the right foundation: customer data already in HubSpot, a clear input, a structured output, and an outcome that ultimately speeds up when students get into sessions.
-Sep-11-2026-04-21-22-8168-PM.png?width=48&height=48&name=Untitled%20design%20(10)-Sep-11-2026-04-21-22-8168-PM.png){kind=small}
-May-15-2026-08-50-57-5826-PM.png?width=215&height=50&name=image%20(4)-May-15-2026-08-50-57-5826-PM.png)
-Jul-31-2026-02-22-13-7631-AM.png?width=215&height=50&name=image%20(7)-Jul-31-2026-02-22-13-7631-AM.png)
-Sep-15-2026-04-19-07-7995-PM.png?width=215&height=50&name=Untitled%20design%20(12)-Sep-15-2026-04-19-07-7995-PM.png)
.webp?width=215&height=50&name=filters_quality(90).webp)