❯ git show --details cases/waste-sync.md
Payments

Digitising the back office for operators who pay for every mistake in cash

Waste Sync is a cleantech SaaS venture for U.S. waste haulers running multiple sites. The product digitised compliance and payments. My work was the unglamorous middle: the payment flows, the reconciliation, and what happens when something does not match.

role
Product lead, venture studio
timeline
2024 – 2025
users
Multi-site U.S. haulers and their back-office staff
programs
Founder Institute, Cleantech Open
outcome
Manual payment and exception work replaced with defined paths

The problem

A regional hauler runs several sites, each with its own trucks, routes, permits and customers. Money moves in two directions: customers pay the hauler, and the hauler pays landfills, transfer stations and contractors. In the operators we worked with, all of it ran through spreadsheets, paper tickets and a back office that spent its week chasing what did not match.

The failure mode in this business is not a bad user experience. It is cash. A missed receivable, a duplicated vendor payment or an unreconciled tipping fee comes straight out of the operator's margin, which is already thin.

What I defined

01
Payment pathsEach type of money movement got an explicit path: who initiates, what evidence is attached, what gets posted and when. Customer receipts, vendor payments and inter-site transfers no longer shared a single bucket.
02
ReconciliationDaily matching between tickets, invoices and bank activity, with exact matches cleared automatically and everything else landing in a queue with the reason it failed.
03
Exception handlingA defined escalation path for every mismatch type, with an owner and a deadline. The queue was built for the person clearing it, not for the person who designed it.
04
Compliance recordsPermit and manifest data captured once at the point of work, so the compliance filing was a report rather than a project.

The principle was the same one I later applied to JournalRelay: clear the easy matches without a human, make the hard ones fast for the human, and never let an unresolved exception disappear.

Results

Manual payment and reconciliation work was replaced with defined, auditable paths. The venture graduated from Founder Institute and Cleantech Open and was part of the portfolio that reached $500K ARR across the studio. An NSF pitch was drafted on the compliance side of the platform.

This project is also the one that taught me what two-sided payment flows with an accountable operator look like from the inside. Reconciliation, exception handling and a human decision at the boundary turned out to be the same problem in waste, in accounting and in fund administration.

What I would do differently

  • Start with the exception queue. We built the happy path first. The operators judged the product entirely by how it handled the mismatches.
  • Sit in the back office for a week. Discovery calls told us what people thought they did. Watching told us what they actually did at 4pm on a Friday.
06.

connect()

Building something where a wrong number costs money?
Fintech, accounting, operations, applied AI. I am always open to talking with people who ship.

$ echo "Let's build something that survives an audit"