By Quang Do, PE — Founder, Pyra Engineering
Between 2021 and 2026, my firm sealed mechanical, electrical, and plumbing drawings for 374 commercial projects in Texas. Restaurants, dental offices, small medical, multifamily, hospitality, retail tenant improvements, and ground-up industrial. We’re licensed in 14 states — Texas, Florida, Arizona, Arkansas, California, Colorado, Georgia, Louisiana, Michigan, Minnesota, New Jersey, Ohio, Utah, and Wisconsin — but the dataset behind this piece is our Texas portfolio, sealed on a standard-scope drawing-set turnaround of one business day.
That production cadence isn’t the point of this piece. The point is what we saw doing it.
When you’re the MEP engineer of record on 374 projects, you sit at the bottom of the design stack. The architect handed you the floor plan. The owner already signed the lease. The GC already bid the job (or is about to). Whatever decisions everyone above you made about scope, code, jurisdiction, and timeline land on your desk as constraints. Then you draw and stamp.
From that seat, certain patterns repeat. Not because we’re seeing one developer’s quirks or one city’s idiosyncrasies — because we’re seeing the same things across hundreds of projects in dozens of cities. Here are the three patterns I think are most useful to anyone planning, designing, or financing commercial construction in the U.S. right now.
1. Permit timelines vary 4× between cities — and the variance has nothing to do with the project
If you’ve only built in one city, you assume permit review takes about as long as it takes. It doesn’t.
Across our 374-project Texas dataset, the permit review timeline for a comparable scope-and-occupancy commercial TI varied roughly four-fold between the fastest cities and the slowest — from under two weeks at one end to well over a month at the other. Same drawing set quality, same code basis, same level of contractor competence. The variable that drove most of the variance was the city’s plan review department.
The fast-end cities had three things in common:
- Predictable intake. You can log into the portal and see your project’s status without calling. Email the reviewer and get a response in 24 hours.
- Single-cycle review. First-cycle comments cover everything the reviewer cares about. Resubmittal is a single revision, not a six-month back-and-forth.
- Plan reviewers who answer technical questions. When you call to clarify something the comment is ambiguous about, you can actually reach a human with code authority.
The slow-end cities tended to have one or more of: portal that goes silent for two weeks at a time, comment letters that release new objections on every cycle, no way to reach the reviewer except through a clerk.
This matters for owners way more than they realize. If you’re financing a restaurant build with a six-month construction loan, the difference between a 10-day permit cycle and a 34-day permit cycle is the difference between profitable and bleeding.
What we learned to do: pick the jurisdiction before we pick the building. When clients give us two options on adjacent lots in different cities, we tell them — frankly — which jurisdiction will get them to occupancy faster. About a quarter of the time, that changes the deal.
2. The single MEP spec that drives 30-40% of cost variance on dental and small-medical TI
We’ve done a meaningful number of dental and small-medical-office TIs. There’s a pattern that holds across almost all of them.
The single design decision that drives the largest swing in MEP cost — across DD through occupancy — is the medical-gas / vacuum / dental-air infrastructure decision. Specifically:
- Bottle-supply medical gas (cheapest, simplest, restrictive on layout and operations)
- House-line manifold supply (mid-cost, mid-flexibility)
- Pre-piped tenant infrastructure shared with adjacent dental suites (cheapest per suite but only available in purpose-built dental condos)
- Central system with reserve manifold (highest cost, highest operating flexibility)
The decision is usually made at lease-negotiation stage — meaning before MEP is engaged. By the time the project hits our desk, the gas-source decision is locked, and the MEP scope and cost flow downstream from it.
What we tell dental clients now: before you sign the lease, get an MEP engineer to spend 30 minutes on a phone call with you and your broker. Walk through the four options against the specific suite. Pyra does this for free, because the cost of not doing it is so much higher than the cost of doing it.
The same lease-stage MEP review applies to a quieter version on restaurant TI: make-up air. If the building’s existing make-up air infrastructure can’t carry the Type I hood you want, you’re paying for roof-top equipment, duct routing, and possibly structural mods that will swamp your TI budget. Make-up air should be evaluated at LOI stage, not at MEP DD.
3. The ASHRAE 62.1 calculated-to-measured gap is wider than the literature reports
This one is technical. Skip it if you’re a layperson.
ASHRAE Standard 62.1 governs commercial outdoor-air rates. The calculation method is well-documented and gives you a required cfm of outside air based on occupancy, floor area, and use category. Engineers calculate it, building owners approve it, mechanical contractors install equipment sized for it, and TAB contractors certify it.
Then it goes into use, and the real measured outdoor-air rate is meaningfully different from the design value on a notable share of buildings we’ve followed up on (where we had post-occupancy measurement available). The deltas are bidirectional — sometimes the building is over-ventilated (waste), sometimes under-ventilated (IAQ risk). The most common drivers we’ve identified:
- Restaurant occupancy categories that don’t match real operation. ASHRAE 62.1’s “restaurant dining room” rate assumes a use profile that fast-casual concepts don’t follow. The miss can be 25-50%.
- Make-up air interaction with general ventilation. When a kitchen exhaust hood is operating, the make-up air it’s pulling in counts toward general ventilation for occupied zones if it ducts through them. Most calculations don’t credit this correctly.
- Operator behavior: dampers manually overridden, economizers disabled, schedules misaligned with actual hours.
Our internal response was to build a revised default-assumption table for the occupancy categories where we see the gap most often (restaurant, dental, small medical) and to add a post-occupancy verification recommendation to every drawing set. A minority of clients act on it. Of those, the corrections paid for themselves in operating cost within roughly the first two years on every project we’ve followed up on.
We’re publishing the revised default-assumption table in the 2026 Texas Field Report.
What this all adds up to
You can summarize these three patterns one way. The information that determines whether a commercial project succeeds is mostly invisible to the people making the decisions.
Jurisdiction-picking happens at site selection, before plans are drawn. Medical-gas and make-up-air decisions happen at lease signing, before MEP is hired. Ventilation gaps show up after occupancy, when nobody’s looking. By the time a problem is visible, it’s too late and too expensive to fix cleanly.
The fix isn’t more rigor in MEP design. The fix is bringing MEP into the conversation earlier — at LOI, at lease negotiation, at site selection.
I started Pyra Engineering in 2021 because I wanted to compress the MEP design cycle from weeks to a day for standard scope. The unexpected lesson from 374 projects is that the cycle compression matters less than the timing: a day of MEP advice at the right moment is worth weeks of MEP design at the wrong moment.
If you’re an owner, GC, architect, broker, or operator who’s about to sign a lease on a commercial space — call your MEP engineer first. If you don’t have one, call us. The first call is free, and the cost of skipping it usually isn’t.
Quang Do, PE is founder and principal engineer of Pyra Engineering, an MBE/HUB-certified MEP design firm in Houston. Pyra has sealed 374+ commercial drawings since 2021 and is licensed in 14 states. The 2026 Texas Field Report — anonymized data and findings from those projects — is free. Pyra also publishes a free suite of MEP calculators.
