A subcontractor's growth problem is a list problem: more general contractors need to know who you are before the invitation to bid goes out. Building records name the contractor on most filings, so that list can be built rather than guessed at. The harder question is timing, and the answer is not a number of weeks — it is where your trade sits in the sequence.
Bid lists are set before the portal posts
Most subcontractors find work the same way: referrals, associations, and the major bid portals. All three work. All three share a limitation, and it is worth naming plainly rather than pretending the tools are bad.
By the time an invitation to bid reaches a portal, the general contractor has already decided who is on the list. You are pricing against everyone else who saw the same notice, on a project whose team was assembled without you in the room. Portals are an excellent way to price work you were invited to. They are a poor way to become someone who gets invited.
The firms with the best backlog tend to fill trade packages from relationships that predate the project. Which means the useful question is not how to win more of the bids you see — it is how to be known to the right contractors before the bid exists. That is a marketing problem, and unusually for construction it is one with a public data source behind it, the same one general contractors use in the other direction and that the subcontractor view of the same records reads differently.
The GC list you can build yourself
When a general contractor pulls a building permit, their name is on it. That is the whole mechanism, and it is more useful than it sounds.
Take one city authority. Its commercial filings named a contractor on 89% of records — 1,134 distinct firms. That is a market's worth of general contractors, identified without a single phone call — and visible in real rows rather than inferred.
But 1,134 is a directory, not a target list. Group by how often each firm files and the picture sharpens considerably.
| Commercial filings | Firms | Share | What it means for you |
|---|---|---|---|
| One only | 560 | 49% | Half the list built once. Owner-builders, one-off projects, firms passing through |
| Two to four | 396 | 35% | Occasional. Worth knowing, not worth a campaign |
| Five to nine | 102 | 9% | 178 firms that build repeatedly. This is the list worth working — small enough to approach properly, large enough to matter |
| Ten to twenty-four | 51 | 4% | |
| Twenty-five or more | 25 | 2% |
One city authority, commercial building records only. Percentages are of the 1,134 firms named, and total 99% after rounding. The ten busiest firms alone account for 31% of commercial filings — higher than the 15% and 18% ten-firm shares reported in the market-level read, because that analysis covered all record types across county authorities while this one is commercial work in a single city.
Half the firms in that market filed exactly once. Chasing all 1,134 would mean spending half your effort on companies that will not build again this year. Chasing the 178 that filed five times or more is a different exercise entirely — a list one person can genuinely work, made of firms that have proven they generate repeat trade packages.
A market with 1,134 general contractors in it does not have 1,134 worth knowing. It has about 178, and the records tell you which.
The same grouping does other useful work. Declared value tells you which firms build at your scale. Property and record type tell you whether they build what you are set up for. And for a subcontractor working several states, this is the difference between a coherent territory plan and a stack of business cards, which is why which authorities you cover matters more than how many records you receive.
When to reach out depends on your trade
The old advice was to contact a general contractor a fixed number of weeks after permit approval. That advice is wrong for most trades, because trades are bought out at wildly different points relative to when they show up on site.
Commercial projects follow a broadly fixed order — sitework, foundation, structure, envelope, MEP rough-in, then finishes — with each trade depending on the one before it. But the buyout does not follow the same order at the same pace. Long-lead packages in particular are committed early: structural steel now runs 14 to 20 weeks in most US markets after a 38% rise since 2024, and electrical switchgear can run 30 weeks or more for complex assemblies. Those decisions get made near the start of a job that will not need the equipment for the better part of a year — which is why general contractors are advised to lock long-lead procurement during the permit-review window rather than after the permit issues.
| Your trade | On site | Which records to work |
|---|---|---|
| Sitework, excavation, concrete | First | The earliest records available. Applied and in review — by the time a permit is issued the earthwork sub is frequently already engaged |
| Structural steel, long-lead equipment | Middle | The earliest records too, despite being on site later. Lead times force the decision forward. This is the trade group most often approached too late |
| Mechanical, electrical, plumbing | Middle | Approved and issued records. Rough-in follows framing, but the package is usually bought out well before that |
| Drywall, flooring, paint, millwork | Last | Issued and active records. The one group for whom a later record is genuinely still early |
Sequencing is indicative and varies by project type, delivery method and region. The point is the ordering, not a number of weeks — which is why what each status actually means is a better clock than the calendar.
The highlighted row is where most subcontractors lose work without knowing it. A steel fabricator or an equipment supplier who waits for a project to look real has waited too long, because the procurement decision on their package was made while the record still said applied — which is an argument about how quickly the file reaches you as much as what is in it.
Want the general contractor list for your markets? Tell us the jurisdictions and categories you work in and we will show you who is filing, how often, and at what values.
Request a GC list or call 888-888-1214What to say when you get there
A list only helps if the message earns a reply. Estimators receive a great many generic subcontractor introductions, and the ones that get answered share three properties — all three built on knowing what the record says.
Reference the actual work. Not "we do commercial electrical" but the category, the market and ideally the project. The entire advantage of working from records is that you know what they are building; a message that does not show it has thrown that advantage away. The same principle decides whether any list is worth working.
Bonding capacity, certifications, license numbers and relevant completed projects belong in the first message. None of that comes from building records — it is yours to supply, and it is what turns a name into a bidder. The records get you the right door; your credentials get you through it.
Offer a number on their next project rather than asking for a meeting. Estimators are measured on coverage, and a subcontractor who makes their package easier to fill is solving their problem rather than adding to it. Being easy to work with is a competitive position, not a soft skill.
What the records will not tell you
Four limits, and the last one matters more to a subcontractor than to anyone else who reads this data.
Revenue, employee count and company size are not in a building record. Those come from an entirely different kind of vendor, and the published field list makes clear which is which. What a record gives you instead is arguably better for this purpose: how much a firm actually builds, which is a more useful measure of a general contractor than headcount.
License status, bonding capacity and insurance are not published on the record. Verify them the way you always have — we do not claim to.
The record carries a permitting status, not a commercial one. It will not tell you whether trade packages have been let, or to whom. Any provider offering to filter on that is describing something other than public records.
Nothing in a building record tells you about payment history, retainage practice or how a firm behaves when a job goes badly — and for a subcontractor that is the question that matters most. Other subcontractors are the only real source. A firm filing twenty-five permits a year is worth knowing about; whether they are worth working for is a different inquiry, and worth making before the first bid rather than after.
Why Alliance Data Solutions
Any complete records set supports the approach above. The case for this one, against what is normal in the category.
Multi-market coverage without multi-market effort. A subcontractor working one metro can look up filings on a city portal for free, and should. The calculation changes across state lines, because every authority publishes differently and reconciling them never finishes. 300+ jurisdictions across 39+ states, counted by issuing authority rather than by state.
Grouping needs the whole market. The frequency analysis in section two is a share calculation, so a partial file gives a confidently wrong answer rather than an incomplete one — you would rank the wrong firms. Records arrive deduplicated, scored on how many of their up to twenty fields carry a value, and filtered to the categories and thresholds you set.
Timing that suits the early trades. A record filed at its source is generally available in a delivered file the next day. For the long-lead group in section three, a monthly file arrives after the decision has been made.
Published pricing, monthly billing, no contract. Plans run from a single metro up to multi-state, with the numbers on a page rather than a call — which for a subcontractor deciding whether this is worth trying is the difference between an afternoon and a procurement exercise.
And the limit, stated the same way it is stated everywhere else on this site: this is public building activity, assembled, deduplicated, scored and filtered at national scale. It is not a proprietary intelligence network and we do not claim one. Every other subcontractor in your market can read the same filings. Most of them will not.
Common questions
How can a subcontractor find general contractors to work with?
Building records name the contractor on most filings, so a subcontractor can build a list of the general contractors actually filing work in their markets rather than guessing. In one city authority, commercial filings named 1,134 distinct firms — but only 178 of those filed five times or more. That smaller group is the list worth working, because it is the firms that build repeatedly rather than once.
When should a subcontractor contact a general contractor about a project?
It depends on your trade, because trades are bought out at very different points relative to when they appear on site. Sitework and concrete are needed almost immediately. Structural steel now runs 14 to 20 weeks in most US markets and electrical switchgear can run 30 weeks or more for complex assemblies, so those packages are committed early even though installation is months away. Finishing trades are bought out latest. Read the record's stage against your own position in the sequence rather than against the calendar.
Are bid portals enough to grow a subcontracting business?
They are useful and they are reactive. By the time an invitation to bid appears on a portal, the general contractor has already decided who is on the list, and every other subcontractor who saw the same notice is bidding it. Portals are a good way to price work you were invited to. They are a poor way to become someone who gets invited.
What do building records not tell you about a general contractor?
They do not carry revenue, employee count, bonding capacity, license status, credit standing or payment history. A record shows that a firm filed work, where, in what category and at what declared value. Whether they pay on time is something only other subcontractors can tell you, and that question is worth asking before the first bid rather than after.