Shaping a permitting tool
from the inside out

A problem no one in the
Wrench Group had solved yet

Morris-Jenkins handles hundreds of HVAC, plumbing, and electrical installations a year. Every one of them requires permits, inspections, and compliance documentation across dozens of municipalities. Before Wrench Connect, none of that had a real home.

Most of the 24+ companies under the Wrench Group didn't even have a formal permitting process for HVAC and plumbing changeouts. Morris-Jenkins was the exception, and that meant we were essentially building the playbook from scratch, for ourselves and potentially for the entire group.

The "before" was genuinely chaotic. Coordinators were simultaneously managing work across four systems: ServMan (our primary field service platform), Wrench Connect (the new permitting tool), ServiceTitan (incoming replacement), and Excel spreadsheets to bridge the gaps between all three. When an inspection date changed, you had to remember to update it in multiple places. When a permit stalled at a jurisdiction, you found out when the customer called, not before.

Wrench Connect was supposed to fix this, a centralized permitting management tool covering the full lifecycle from application through inspection scheduling, status tracking, customer contact history, and compliance documentation. My job as SME was to make sure what got built actually worked for the people who would live in it daily. That role was formally designated by leadership. I was selected as one of the HVAC coordinator trainers for the ServiceTitan Conference Room Pilot, responsible for identifying functionality issues, building training materials, and representing the permitting team's needs across the transition.

Two dev teams, one deadline,
and a platform we had to abandon

The story of Wrench Connect isn't a clean product success arc. It's more honest than that, and more instructive.

We worked with the first development team for over a year. They were based overseas, with a team lead who would meet us on-site in Charlotte while the developers joined by video call. Regular meetings, detailed feedback sessions, documented wish lists. But the pattern was frustrating and consistent: fix one thing, break two others. The UI stayed skeletal and beta-looking. The ServiceTitan integration was unreliable. Search was nearly unusable at volume. Jobs and permit data didn't transfer cleanly from ServMan, which meant coordinators were toggling between three platforms just to schedule a single inspection.

In February 2024, we connected with that second team, local developers from NC and GA who came to our office, observed how we actually worked, and came back with wireframes that finally captured what we needed. But we were already running up against our March 7, 2024 ServiceTitan launch deadline. Stakeholders decided not to extend the timeline. We moved everything to ServiceTitan, and Wrench Connect was shut down.

Of the 1,067 permitting records in Wrench Connect, I personally imported 898 into ServiceTitan's environment. The other 169 either had data integrity issues or couldn't be migrated cleanly. That migration, unglamorous, meticulous, and high-stakes, was also my work.

Four categories of work,
one goal: make it real

My contributions fell into four areas. Each one directly improved how the tool functioned for coordinators who lived in it daily. This wasn't feedback for its own sake. Every issue I raised, every change I advocated for, was grounded in what I knew the workflow actually required.

What the tool was vs.
what it became

These aren't hypothetical improvements. They're the direct result of systematic UAT documentation, feedback sessions, and persistent advocacy through the product refinement cycle over more than a year.

The problems Wrench Connect couldn't solve
became the blueprint for ClearTrack

When we moved to ServiceTitan, the operational complexity didn't go away, it just moved. Coordinators still had to toggle between the salesman's ticket, the installer's ticket, and the inspection ticket to gather the information needed to permit a single job. In busy season, that meant chasing information across 2-3 browser tabs for each of up to 25 jobs a day.

Permit data still lived in too many places. Inspector contact information still had to be looked up manually from municipal websites. Regions couldn't be toggled cleanly. And when Morris-Jenkins opened a Greenville location, duplicate status options started appearing for everything, CLT and GSP versions of every dropdown, because ServiceTitan wasn't built for multi-region permitting operations.

What this work taught me
about design that matters

The billing address vs. service address bug is the one I think about most. On the surface it sounds like a minor data issue, two address fields, wrong one selected. In practice, it meant every permit for a rental property was being routed to the wrong jurisdiction. That's not a UX problem. That's a compliance failure waiting to happen.

Catching it required thinking beyond "does this look right" to "does this work right for the specific operational context it lives in." That's the difference between testing a tool and understanding it. My background as a permit coordinator wasn't incidental to this work, it was the reason I could see what a developer couldn't.

The second dev team taught me something equally important: good design process isn't just about talent. It's about proximity. Those developers flew to Charlotte, sat in our office, watched us work, and asked real questions. They got it immediately. The first team never visited. They never would have found the billing address bug. They didn't know what a rental property inspection meant operationally.

And the migration, importing 898 records by hand, flagging 169 that couldn't be trusted, taught me that design work doesn't end when a feature ships or a platform launches. Sometimes it ends with you cleaning up someone else's mess so the team can move forward.