Bright Data Pricing: Cost per 1,000 Pages
Bright Data sells six scraping products on four different meters, and the same $499 plan buys 383,000 unlocked requests on one of them and roughly 36,000 browser page loads on another. Every rate read firsthand, converted to cost per 1,000 pages.
By the ClawEngine team
September 2026 · 9 min read
Hit Extract to turn this page into clean, LLM-ready data.
robots.txt respected · public data only
Short answer: Bright Data prices every scraping product on its own meter. The Web Scraper API, Web Unlocker API and SERP API each cost $1.50 per 1,000 successful records or requests on pay-as-you-go, or $499 a month for roughly 380,000 on a Scale plan with extras at $1.30 per 1,000, and each includes 5,000 free a month. The Crawl API starts at the same $1.50 and falls to $1.00 per 1,000 on its $1,999 plan. The Browser API and residential proxies are billed by bandwidth at $8 per GB pay-as-you-go. That split decides the bill: a 2MB page costs $1.50 per 1,000 through Web Unlocker and about $15.60 per 1,000 through the Browser API.
Bright Data is not expensive in the way its reputation suggests. The request-priced products sit at a very ordinary $1.30 to $1.50 per 1,000, and the company bills only what it successfully delivers. What makes the invoice hard to forecast is that Bright Data sells several overlapping ways to fetch the same page, and they do not share a unit. Records, requests and gigabytes all sit on the same pricing hub, and the product you pick moves the cost of one page by an order of magnitude.
Every figure below was read from Bright Data's own product pricing pages and billing documentation on 10 September 2026. Where a number is our arithmetic rather than a published rate, the working is shown next to it.
Bright Data pricing by product in 2026
Six products, four meters. The free tiers and the $499 entry plan look nearly identical across the request-priced products, which is exactly why teams end up on the wrong one.
| Product | What is metered | Free each month | Pay-as-you-go | $499 plan includes | Lowest published rate |
|---|---|---|---|---|---|
| Web Scraper API | Records delivered | 5,000 records | $1.50 per 1,000 | 384,000 records | $1.30 per 1,000, Enterprise by quote |
| Web Unlocker API | Successful requests | 5,000 requests | $1.50 per 1,000 | 383,000 requests | $1.30 per 1,000, Enterprise by quote |
| SERP API | Successful requests | 5,000 requests | $1.50 per 1,000 | 380,000 requests | $1.30 per 1,000, Enterprise by quote |
| Crawl API | Requests | None listed | $1.50 per 1,000 | 380,000 requests | $1.00 per 1,000 on $1,999 for 2M |
| Browser API | Gigabytes of traffic | 1 GB | $8 per GB | 71 GB | About $7.03 per GB on Scale |
| Residential proxies | Gigabytes of traffic | None listed | $8 per GB | About 71 GB at the $7 list rate | $5 per GB list on the $1,999 plan |
Two details in that table are easy to misread. The residential proxy page is running a promotion right now, a coupon for 50 percent off for three months plus a matched first deposit up to $500, and the included gigabytes it prints track the promotional rate: 141 GB on the $499 plan is about double the 71 GB that $499 buys at the $7 list price. Budget on the list figure, because the promotion ends and the plan does not. The other detail is that the Crawl API is the only request-priced product with a published ladder above $499: $999 for 900,000 requests at $1.10 per 1,000, and $1,999 for 2 million at $1.00.
How much does Bright Data cost per 1,000 pages?
For the request-priced products the conversion is simple, because one successful request is one page: $1.50 per 1,000 on pay-as-you-go, $1.30 on a Scale plan, and $1.10 or $1.00 on the two larger Crawl API plans. On those products Scale pays for itself at about 333,000 requests a month ($499 divided by $1.50 per 1,000), so below that volume pay-as-you-go is the cheaper way to buy the same thing.
The bandwidth products work differently. A real browser downloads the page and everything it pulls in: scripts, fonts, images and third-party tags. Your cost per page is the weight of all that multiplied by the per-GB rate, which means the same list of URLs costs very different amounts depending on what those sites ship.
| Page weight in a browser | Pages per GB | Browser API at $8 per GB | Browser API on Scale | Web Unlocker, pay-as-you-go |
|---|---|---|---|---|
| 500KB, a lean text page | 2,097 | $3.81 per 1,000 | $3.35 per 1,000 | $1.50 per 1,000 |
| 2MB, a typical modern page | 512 | $15.62 per 1,000 | $13.73 per 1,000 | $1.50 per 1,000 |
| 5MB, an image-heavy listing | 205 | $39.06 per 1,000 | $34.32 per 1,000 | $1.50 per 1,000 |
Working: 1,000 pages at 2MB is 1.95 GB, which is $15.62 at $8 per GB and $13.73 at the Scale rate of $499 for 71 GB. So at an ordinary page weight the Browser API costs about ten times what Web Unlocker charges for the same URL, and about 26 times on image-heavy pages. Put another way, the same $499 buys 383,000 unlocked requests or roughly 36,000 browser loads of a 2MB page.
That premium is not a rip-off. It buys a real browser you drive with Puppeteer, Playwright or Selenium, so you can click, scroll and wait for content that only appears after interaction. It is wasted money when all you needed was the page content. If you do need the browser, blocking image and font requests is usually a bigger saving than negotiating a lower per-GB rate, which we worked through in our residential proxy cost comparison.
What Bright Data bills, and what it does not
The Web Scraper API, Web Unlocker and SERP API pages all say the same thing: pay only for success. The Web Scraper API page is the most explicit, promising no charges for failed deliveries. Pay-as-you-go accounts can also set a monthly spend limit, and the request-priced products advertise unlimited concurrency, so you are not paying for a thread ceiling the way credit-based vendors charge you.
That success rule is more generous than it sounds, because it is not the norm. Firecrawl charges a credit for a page that comes back as a 403 or a 404, as our Firecrawl pricing breakdown shows, and Oxylabs counts any 2xx or 4xx response as a billable result, covered in Oxylabs pricing per 1,000 results. If your URL list is stale or your targets block often, Bright Data's request products are the ones where that waste stays off the invoice. The bandwidth products are the exception to the pattern: the meter counts bytes, so heavy pages and retries use gigabytes whatever the outcome.
Commitments, billing dates and rollover
The billing documentation adds three rules you will not find on the product pages. Each plan commits you to a minimum payment at the start of every month while the account is active, and the billing cycle starts on the 1st, with a first month that begins mid-month prorated by active days. Usage is calculated daily and drawn down from your account balance. And unused plan volume does not roll over: prepaid plans renew monthly by default, so a quiet month on a $499 plan is $499 spent.
Those mechanics are worth flagging to whoever closes the books. A vendor that bills a prorated commitment on the 1st and then draws usage down daily from a prepaid balance produces charges that rarely match the plan price on any single statement, which is the kind of line that slips through month-end unless someone is reconciling it against the bank feed. The pricing pages also display estimates in local currency, while charges are made in US dollars.
Which Bright Data product should you buy?
- Known URLs on defended sites, and you only need the content: Web Unlocker. Stay on pay-as-you-go below about 333,000 requests a month and move to Scale above it.
- Structured records from sites Bright Data already supports: the Web Scraper API, which bills per record rather than per request.
- Search engine results: the SERP API, on the same $1.50 and $1.30 rates.
- Whole sites rather than URL lists: the Crawl API, the only one with a published path down to $1.00 per 1,000.
- Clicking, scrolling or anything interactive: the Browser API, and budget per gigabyte, not per page.
- You already run your own scraper and only need clean IPs: residential proxies, sized on the list rate rather than the promotion.
When you do not need Bright Data at all
Bright Data earns its price on marketplaces, travel and ticketing sites and anything behind serious bot protection. Most scraping jobs are not that. Documentation, company sites, blogs, news and public directories serve their content to any well-behaved client, and for those a proxy platform is a large bill for a problem you do not have. It is worth reading whether you need a proxy network at all before you size a plan.
ClawEngine is built for that second group, and it is priced the other way around. One crawled, rendered page returned as clean markdown or typed JSON counts as one page of plan allowance: $39, $99 and $399 a month for roughly 50,000, 250,000 and 1,500,000 pages, which is $0.78, $0.40 and $0.27 per 1,000. That pricing is published but not yet open for purchase. Against Web Unlocker at $1.30 to $1.50 we are cheaper from the $99 tier, and the extraction into typed fields is included rather than being code you maintain. We also defeat no anti-bot system, sell no proxies and do not scrape search results, so every job in the first half of that product list is one we cannot take. If you are choosing between the enterprise platforms themselves, compare Bright Data vs Oxylabs, and for the tools that fit ordinary public pages better, see Bright Data alternatives.
Frequently asked questions about Bright Data pricing
How much does Bright Data cost per month?
There is no single monthly price. The request-priced products run $499 a month on their Scale plans, covering about 380,000 records or requests, or you pay $1.50 per 1,000 with no commitment. Residential proxy plans run $499, $999 and $1,999, and Enterprise is quoted. Most small teams start on pay-as-you-go and move to a plan above about 333,000 requests.
Does Bright Data have a free tier?
Yes, per product. The Web Scraper API includes 5,000 records a month, and Web Unlocker and the SERP API include 5,000 requests a month, all without a credit card. The Browser API includes 1 GB a month. The free allowance is enough to measure what your own targets cost before you commit to anything, which is the best use of it.
Is Bright Data pay as you go?
Yes. Every scraping product has a pay-as-you-go option with no monthly commitment: $1.50 per 1,000 records or requests on the request-priced products and $8 per GB for the Browser API and residential proxies. You can set a monthly spend limit. The trade is a higher unit rate than the $499 plans, which drop extras to $1.30 per 1,000.
How much does the Bright Data MCP server cost?
The Web MCP server is billed like the request products: 5,000 free requests a month, then $1.50 per 1,000 on pay-as-you-go, with a $499 monthly plan covering about 383,000. Its Rapid mode exposes two tools and its Pro mode exposes 69, and the browser tools draw on bandwidth-priced capacity, so an agent that opens browsers will cost more per task.
Does unused Bright Data volume roll over?
No. The billing documentation states that unused plan volume does not roll over and prepaid plans renew monthly by default. Plans also carry a minimum monthly commitment billed from the 1st. If your volume swings from month to month, pay-as-you-go with a spend limit is usually cheaper than a plan you only fill half the time.
See ClawEngine turn pages into clean data
Point ClawEngine at any public or permitted site and get back clean markdown, JSON, or typed structured fields in one call. Crawl at scale, render JavaScript, and feed your RAG pipelines and AI agents, robots.txt and Terms of Service respected.