Construction data gets sold four different ways, and the model matters more than the sticker price — the same records can look cheap or expensive depending purely on how they are packaged. Cost per record is the one figure that lets you compare across models. Ours runs from about 20 cents down to about 8, published on a page rather than quoted on a call.
Why this question is hard to answer
Search for what construction data costs and you will find a great many pages that discuss the value of construction data, the return on construction data, and the importance of choosing the right construction data partner. You will find very few numbers.
That is not an accident and it is not incompetence. Quoting on a call lets a seller set the price against the buyer rather than against the product. Two companies with similar needs can be charged materially different figures for the same file, because the number is anchored to whatever budget the salesperson believes exists. It is a legitimate commercial strategy — most enterprise software works this way — and it is also why comparing options in this category is unusually hard.
The practical consequence for a buyer is that you cannot shop. You can only enter a series of sales processes, each of which will take several weeks and produce a number you have no way of calibrating. By the time you have three quotes you have spent a month, and the quotes will be structured so differently that comparing them is its own project.
This article is an attempt to make that comparison possible: how the models work, what genuinely drives the price, and what we charge. If you would rather skip the reasoning, our numbers are on a page — as is the coverage those numbers buy and the field specification, since price without either is meaningless.
The four ways it gets sold
Nearly every offer in this market is one of four shapes, whatever the service around it is called. Recognizing which one you are looking at is the first step, because each has a different failure mode.
| Model | How it works | Suits | Watch for |
|---|---|---|---|
| Per record | You pay for each record you take. Often sold as credits bought in advance. | Low or irregular volume; testing a market before committing | Cost scales with no ceiling. A good month becomes an expensive one, and budgeting is guesswork. |
| Per seat | A monthly fee per user, usually with a soft or hard cap on records. | Small teams where everyone works the same territory | Cost tracks headcount rather than data value. Adding a rep who needs occasional access costs the same as adding a heavy user. |
| Territory and volume | Priced by the geography you need and the number of records it produces. Seats are usually unlimited. | Most suppliers and builders with defined markets | What counts as one market. A "metro" and a "state" can differ tenfold in record volume. |
| Enterprise license | An annual negotiated contract, typically for national coverage or the right to build products on the data. | National programs; anyone redistributing or embedding the data | Multi-year terms, automatic uplifts, and no published baseline to negotiate against. |
Territory and volume is highlighted because it is the model that most closely matches how suppliers and builders actually consume this data — by market, not by user.
The reason the model matters more than the sticker price is that it decides what happens when your usage changes. A per-record offer that looks cheap at 200 records a month is not cheap at 2,000. A per-seat subscription that suits four reps stops suiting you at fourteen. Nobody is misleading you; the shape simply stops fitting. Ours is territory and volume, for the reason in the note above.
The same records can be priced four different ways and look like four different products. Convert everything to cost per record before you compare anything else.
That conversion is the single most useful thing a buyer can do. Take the total annual cost, divide by the records you will actually receive in a year, and you have one number that survives across all four models. It will not tell you whether the data is any good — for that, what a vendor means by verified matters far more than what they charge — but it makes the offers comparable, which nothing else does.
What actually moves construction data pricing
Five things drive price in this category. Only two of them are about the data.
How many issuing authorities, and which. This is the largest genuine cost driver, because every authority is a separate integration that has to be built and then maintained forever. A provider covering one state and a provider covering forty are not selling the same thing at different scales; they are selling different products. Why that is true is worth understanding before you compare two coverage claims.
Daily collection costs more to run than monthly collection, and it is worth more, because timing is most of the value in this data. A cheap feed that arrives six weeks late is not a discount, it is a different and less useful product.
Almost every provider discounts as volume rises, because the cost of serving the ten-thousandth record is near zero once the pipeline exists. Ours falls about 60% per record from the smallest published tier to the largest. If an offer does not get cheaper per record as you grow, ask why.
Named account management, service level commitments and quarterly reviews are real costs and appear in higher tiers everywhere. They are also the easiest thing to be sold that you will not use. Be honest about whether you want a quarterly business review or just the file.
The uncomfortable one, and the reason published pricing is unusual. Where the number is set on a call, your industry, your size and your evident urgency are inputs to it. Nothing about the file changes. Which buyer you appear to be can be worth more than any of the four factors above.
What we charge, and why it is on a page
Three published tiers, billed monthly, no contract — against the coverage set they draw from.
| Tier | Per month | Records | Per record | Territory |
|---|---|---|---|---|
| Starter | $99 | 500 | 19.8¢ | One metro area or county |
| Professional | $299 | 2,500 | 12.0¢ | One state, or five metro areas |
| Enterprise | $799 | 10,000 | 8.0¢ | Multi-state |
| Custom | Quoted | — | — | Nationwide programs and data licensing |
Cost per record is derived by dividing the monthly price by the included record volume. Published prices as of August 2026; the pricing page is authoritative if the two ever differ.
Enterprise is $9,588 a year for multi-state coverage at 10,000 records a month. We publish that figure because a buyer should be able to know it without booking a call, and because the alternative — making you sit through a discovery process to learn whether we are in your range — wastes your time to preserve our negotiating position.
Two honest caveats. The tiers are volume caps, not guarantees of what you will find useful — 10,000 records in a market where only a fraction match what you sell is worth less than 2,500 in a market where most do, which is why reading a market before buying it matters more than the tier you pick. And Custom genuinely is quoted, because nationwide programs and licensing arrangements vary too much to publish a number that would mean anything. We would rather admit that than invent a headline figure with an asterisk.
Not sure which tier fits? Tell us the markets you sell into and we will tell you roughly how many records a month that produces — before you pick a plan, and without a discovery call.
Ask for a volume estimate or call 888-888-1214What to ask before you sign anything
Six questions. They work on us as readily as on anyone else, and the answers are more informative than the price — including when you ask us.
What is the cost per record at my expected volume? If the seller cannot answer immediately, they are not thinking about it in those terms, which means you should.
What happens when I exceed the cap? Overage rates are frequently several times the effective in-plan rate, and they are frequently not on the pricing page.
What exactly counts as one market? A metro, a county and an issuing authority are three different things, and a "county" may exclude its largest city. Get the definition in writing before you agree a territory-based price.
What am I committing to, and what does renewal look like? Annual terms with automatic uplift are standard in enterprise licensing. Monthly billing with no contract is not, and it is worth asking for.
Can I see a sample from my own jurisdictions first? Price is meaningless until you know what the file contains. Ask for real rows, not a screenshot of a dashboard.
Is this your published rate? A simple question with an informative answer. If there is no published rate, you have no way of knowing whether you are being quoted the same figure as the company down the road — and that is worth knowing before you negotiate rather than after.
None of this makes price the deciding factor. Coverage, freshness and whether the file actually contains the markets you sell into all matter more, and a cheap feed of the wrong jurisdictions is not a bargain. But you should be able to find out what something costs without a month of meetings, and in this category you usually cannot. That seemed worth fixing on at least one website.
Why Alliance Data Solutions
Price is the last question, not the first. Here is the case on the four that come before it, each stated against what is normal in this category rather than in the abstract.
The number is on a page. The category standard is a discovery call, a needs assessment and a proposal — three weeks to learn a figure. Ours is published, which means you can rule us in or out this afternoon. It also means the company down the road pays what you pay, which is not true of any quote-on-call arrangement and is worth more than a discount.
Monthly billing, no contract. Annual terms with automatic renewal and a built-in uplift are the norm at the enterprise end of this market. We bill monthly and you can stop. That is a genuinely worse commercial position for us — no locked revenue, no switching friction — and we would rather earn the renewal than trap it.
Coverage counted by issuing authority. Most coverage claims are made in states, because states are a bigger number. A state is not a market and a county label is not always a whole county, which is why we publish the authority list instead of a headline. 300+ jurisdictions across 39+ states, counted the harder way.
The specification is published, not described. Twenty data points, named on a page you can hold us to, rather than walked through on a call. Records arrive deduplicated, scored on how many fields carry a value, and filtered to the record types and thresholds you set — reconciled into one specification regardless of which of three hundred authorities each row came from. That reconciliation is the actual product; the underlying filings are public and anyone can scrape a county.
Next day, not next month. A record filed at its source is generally available in a delivered file the next day, and within a few days for authorities that publish on slower cycles. Since most records in any permit feed are applications rather than issued permits, that timing is worth more here than it would be for a feed of completed work.
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. If a competitor covers your specific markets better than we do, that is a real reason to buy from them, and you should be able to work that out from published information rather than from two sales processes.
Common questions
How much does construction data cost?
It depends far more on the pricing model than on the data itself. The same records can be sold per record, as a per-seat subscription, as a territory and volume tier, or as an annual enterprise license. Alliance Data Solutions publishes tiers at $99, $299 and $799 a month for 500, 2,500 and 10,000 records respectively, which works out to roughly 20, 12 and 8 cents per record. Nationwide programs and data licensing are quoted separately.
Why do most construction data providers not publish prices?
Because quoting on a call allows the price to be set against the buyer rather than against the product. A provider that does not publish can charge two comparable companies different figures for the same file, and can anchor to a budget rather than to a cost. That is a legitimate commercial strategy and it is also why comparing options in this category is unusually hard.
What is a fair price per record for construction data?
Cost per record is the only figure that lets you compare offers built on different models, so calculate it before comparing anything else: divide the total annual cost by the number of records you will actually receive in a year. At Alliance Data Solutions that ranges from about 20 cents at the smallest published tier down to about 8 cents at the largest. Whether a figure is fair depends on coverage, freshness and how much of the file matches what you sell.
Should construction data be bought per record or by subscription?
Per record suits low, irregular volume and testing, because nothing is spent when nothing is needed. Subscription suits steady volume in defined markets, because the unit cost falls as volume rises and the budget is predictable. The trap in per-record pricing is that cost scales without a ceiling; the trap in subscriptions is paying for volume or territory you do not use.