Elevon in ForbesStudent Discounts Verified and Sold in Minutes
The helpdesk agent reads every school confirmation in SupportBox, checks Edu eligibility against C-Engineering's rules, and sends the purchase link on its own. Qualifying students buy in minutes, and the team steps in only for the real edge cases.
Client
C-Engineering
Industry
IT & software
Solution
Custom AI automation (Elevon suite)
Deployment
Production
Confirmation in. Eligibility checked. Purchase link out. Nobody had to touch the queue.
The Challenge
The Edu version of the software is available only to eligible schools. Applicants send a school-enrolment confirmation to SupportBox. Someone then has to open each request, read the confirmation, decide whether the institution is an eligible higher or technical school, and only then send the purchase link. The task is predictable, repetitive, and entirely manual.
The judgment calls are where it gets inconsistent: secondary schools, non-accredited institutions, foreign schools, or an ambiguous document each need a decision, and the decision depends on who is on the helpdesk that day. When the queue is busy, response times stretch.
The cost is not one mishandled request. It is a steady drain of support time on a rules-based task, and applicants waiting longer than they should for a link that a clear ruleset could send in seconds.
Why Elevon
The work was never about difficult decisions. It was about applying a clear ruleset consistently, at any time of day.
Eligibility rules owned by C-Engineering
The agent checks each confirmation against a list of eligible institution types maintained by C-Engineering, not a general model guess. Anything outside the rules (secondary or non-accredited schools, unclear documents) is routed for human review, never approved by default.
End-to-end inside SupportBox
The agent reads the request and its attachment directly in SupportBox and, on a positive result, composes the reply with the Edu purchase-form link. No copy-pasting between tools, no separate queue to watch.
From confirmation to order
Once the applicant completes the form, a follow-up flow creates the order automatically in the billing system and confirms it back to the customer. The predictable part runs on its own; people handle only the exceptions.
“Eligibility checks used to sit in a support queue. Now a qualifying student gets the purchase link in minutes, and the team only looks at the genuine edge cases.”
The Rollout
An applicant sends a school-enrolment confirmation to SupportBox. The agent reads the request and the attached document, classifies the institution against C-Engineering's eligibility ruleset, and decides: eligible → compose a reply with the Edu purchase-form link; unclear → flag for a support agent with the reason it was held.
When the applicant completes the form, a follow-up flow creates the order in the billing system and confirms it back to the customer. The helpdesk stays the system of record; the agent handles the repeatable steps.
What this looks like in practice
A student emails a school confirmation. Within minutes they receive a link to the Edu purchase form. A confirmation from a secondary school, or an unreadable document, is routed to a support agent with the reason it was held. The team reviews exceptions instead of processing every request.
The suite
Inputs
AI agents
Order flow
Output
Illustrative reconstruction of the production suite.
Eligibility reply
Real output format, recreated with blind sample data.
The Results
The rules-based work moved to the agent; people kept the judgment calls.
Every enrolment confirmation triaged automatically; eligible applicants receive the purchase link without manual review
Edge cases (secondary / non-accredited / unclear) routed to a support agent with a reason, never silently approved
Reply time for a qualifying request reduced from hours in the queue to minutes
Order created automatically once the applicant completes the form
Eligibility ruleset maintained by C-Engineering; scope changes take effect the next cycle
0 manual
eligibility checks left in the queue
~150 h
manual eligibility checks removed / yr
minutes
reply time, was hours in queue
How we estimate: the agent takes a rules-based task out of the support queue entirely. The visible change is response time, from hours waiting in a queue to minutes, and support time moves to the cases that genuinely need a person.
“The rules were always clear. The work was just checking them by hand, one request at a time. That part is now automatic.”
Why It Worked
Eligibility checking is a rules-based task wearing the clothes of a judgment task. The information needed to decide, the institution and its type, is right there in the confirmation. What made it manual was habit and the absence of a consistent ruleset applied to every request.
Two design decisions kept it safe. The agent only sends the link when the institution clearly matches the eligibility list; anything unclear is routed to a person with the reason, never approved by default. And the ruleset stays with C-Engineering, so the team can widen or narrow eligibility without waiting on Elevon.
Want a similar transformation in your organization?
Let's talk about how Elevon can help your team too.
Book consultation