"Sold permit" sounds like it means a permit for a project that's about to start. It is the opposite of that.
The confusion costs Houston-area contractors real time. The City's sold-permits search is one of the most-clicked surfaces on the entire permitting site, and a meaningful share of the contractors hitting it every week are working from a backwards interpretation of what they're looking at. They're treating the list as a pipeline of upcoming projects. It's actually a list of projects whose paperwork phase is complete — and where the GC, if there is one, has already been chosen.
What "sold" actually means
In Houston Permitting Center vernacular, a "sold permit" is one that has been fully issued and paid for. The fee has been collected, the document has been generated, the file is closed on the City's end. The work is now allowed to begin.
It does not mean the work has started. It does not mean the GC has been selected. It does not even mean the project is going to happen — about 8% of issued Houston permits expire without a single inspection ever being called. "Sold" is a transactional state at the City. It is not a project status.
The reason this matters: by the time a permit is sold, most of the decisions that affect a GC's pursuit are already made. The architect was hired months ago. The GC was usually shortlisted weeks ago. The owner's calendar for shopping vendors closed about the time the plan-review was submitted. You are not early. You are, in fact, late.
Why the name confuses people
The word "sold" carries an inventory connotation — the way a car gets sold, or a lot gets sold. In permit terminology it just means "transacted." The City uses it because permits used to be physical documents that were literally sold over a counter. The phrase stuck even though the workflow is online.
Other municipalities use cleaner language. Austin says "issued." Dallas says "approved and paid." Harris County's unincorporated permitting workflow says "finalized." Houston kept "sold," and the result is a small but persistent misunderstanding that costs contractors hours a week.
Where the actual buying signal lives
The right signal in the permit lifecycle is two or three steps earlier than "sold." Working backwards:
- 01 ·Plan review submitted. The applicant has filed the full drawings and the City has accepted them for review. This is typically 30–120 days before the permit is issued. The GC is being shortlisted right now.
- 02 ·Plan review in routing. Internal review cycle — structural, MEP, planning, fire. About 60% of projects make it through routing without major corrections. The other 40% bounce back, which means another 30–60 days before issuance. Both groups are still in play.
- 03 ·Plat preliminary approval. Even earlier. The parcel has been subdivided or replatted. Six to twelve months before any permit. The earliest visible signal that a real project is going to happen.
If your team is doing project-driven bid pursuit, those are the records to watch. The sold-permit search is the wrong endpoint of the lifecycle.
When sold-permit data is actually useful
I'm not arguing the sold-permits search is worthless. There are two specific use cases where it earns its keep.
First: subcontractor pursuit. If you're a sub — concrete, framing, MEP, glazing — the sold-permit list is exactly the right surface. The GC has just been awarded, the GC has just been forced to actually list contact info on the document, and the GC is now about to start procuring sub trades. Subs who time their outreach to the week after a permit is sold get a meaningful lift on bid invitations.
Second: market intelligence. The sold-permits search is the cleanest weekly proxy for how busy your market is. Permits sold per week, broken down by valuation bracket and zip code, is a very accurate leading indicator of subcontractor labor pricing 60–90 days out. We track this internally and it's one of the most predictive feeds we have.
How to read a sold-permit search result
If you're going to use the sold-permits search at all, read each result carefully. The contractor-of-record field is the one to anchor on — if it's populated and matches an entity you compete with, the project is closed for you. If it's empty or names an owner's LLC instead of a GC, the door may still be open even after sold.
Also worth checking: the gap between application date and sold date. A short gap (30–45 days) typically means a straightforward residential or small-commercial project. A long gap (120+ days) means routing pain — usually a contested commercial project where the design needed to be revised. The long-gap projects are more interesting from a sub-trade perspective; the routing pain often correlates with budget pressure, and budget pressure correlates with last-minute trade selection.
If you want the full field-by-field walk-through of what each line on a Houston permit actually says, I wrote a separate piece on that. The sold-permit search is just one view onto the same underlying record.
The takeaway
"Sold" is the City's word for "done with paperwork." It is not the start of a project. It is the moment when most of the decisions you care about as a GC have already been made.
The right place to spend pursuit time in Houston is one or two records upstream — the plan review, the plat. Those records are less convenient, more sparsely indexed, and require more work to monitor. They are also where the actual buying signal lives. The convenience of the sold-permits search is what makes it the wrong tool for the job.