Written and maintained by CASRAI Editorial Board
Last updated
Tip: try code CASRAI at checkout for 15% off, if the offer is currently active for this program — codes vary by vendor and aren’t guaranteed.
If you’re reading this, you’ve probably already opened a Bright Data pricing tab and a sign-up form, and you’re trying to answer one question before you type in a card number (or don’t): does the free trial actually cover what I need for this one project, or am I about to sign up for enterprise infrastructure sized for something much bigger than a literature-harvesting scrape or a one-off dataset pull? Below is what Bright Data’s own pricing pages say as of August 2026, exactly what its free trial does and doesn’t include, and — for researchers running a single, defined data-collection project rather than an ongoing commercial scraping operation — where Thordata is the cheaper, simpler fit instead.
What Bright Data actually costs (as of August 2026)
Bright Data’s residential proxy network is priced pay-as-you-go by the gigabyte, with volume tiers on top:
- Pay-as-you-go: $4.00/GB, no minimum commitment (an active promotional rate off an $8.00/GB list price, per Bright Data’s own residential-proxy pricing page)
- Tier 1: 141 GB for $499/month (~$3.50/GB)
- Tier 2: 332 GB for $999/month (~$3.00/GB)
- Tier 3: 798 GB for $1,999/month (~$2.50/GB)
- Enterprise: custom pricing above roughly 1 TB/month, sales-negotiated
Bright Data also sells purpose-built scraping products (Web Scraper API, Web Unlocker, SERP API, Scraper Studio) priced separately, generally starting around $0.75 per 1,000 records with a limited ongoing free tier layered on top of the proxy network itself. If your project is a single scrape or a modest literature/citation pull rather than continuous production scraping, the proxy network’s per-GB pricing — not the API products — is almost always the number that matters, since that’s what a small research budget actually consumes against.
Bright Data free trial limits: what you actually get
Bright Data’s sign-up for residential proxies is genuinely no-card-required, and traffic used during the trial period is tracked on your dashboard but isn’t billed while no payment method is attached. What Bright Data does not publish, on its own pricing page as of August 2026, is a fixed “you get X free GB” number for the residential proxy trial itself — unlike a lot of competitors, there’s no stated GB or day cap you can plan a project budget against in advance. That’s a meaningfully different shape of “free trial” than researchers often expect: it’s closer to “sign up, use it, see what you’re charged if you add a card” than “here’s your fixed free allowance, spend it however you like.”
Where Bright Data does publish a concrete free allotment is on its separately-priced scraper products — the Web Scraper API, Web Unlocker, and SERP API each advertise a limited free tier (on the order of a few thousand free records per month) before paid usage kicks in at the per-1,000-record rate above. If your project is API-shaped (structured extraction from a known set of sites) rather than raw-proxy-shaped (routing your own scraper’s requests through residential IPs), that free tier is the more relevant number — but it’s still capped in volume, not a substitute for a full project’s worth of collection.
Bright Data proxy free trial vs. a real research-project budget
The practical question for a researcher isn’t “is there a free trial” — there is — it’s “will the free trial’s unclear cap get me through an actual collection run, or will I be adding a card and paying full per-GB rates partway through.” Because Bright Data doesn’t publish a fixed trial ceiling for the residential proxy network, the honest answer is: plan your budget around the paid per-GB rates above, and treat the trial as a way to test the tooling and IP quality, not as free capacity you can count on for the whole project.
Is Bright Data worth it for a small research project?
For some research use cases, yes — and worth saying plainly rather than glossing over. Bright Data’s residential network is large, audited, and built with compliance tooling (KYC on customers, network-level abuse controls, documented ethical-sourcing practices for its residential IP pool) that matters when the data being collected touches human subjects, is IRB-reviewed, or otherwise needs to hold up to an ethics or compliance review after the fact. If that’s your project, the answer to “is Bright Data worth it” is genuinely yes, and the pricing above is the honest cost of that assurance — see the tradeoff section below.
For a smaller, lower-stakes project — harvesting citation/reference data, pulling publicly available web pages for a literature review or a systematic scoping exercise, building a one-off dataset from openly accessible sources with no human-subjects dimension — Bright Data’s enterprise-grade network and pricing is usually more infrastructure than the project needs, and the per-GB cost adds up fast against a typical research budget or a single grant line item.
Bright Data vs Thordata for researchers
The table below compares the two directly on the dimensions that actually matter for a research data-collection decision, not just headline price.
On raw per-GB cost, Thordata is substantially cheaper across every tier: its published residential proxy pricing runs from roughly $2.00/GB at small volumes down to about $0.65/GB at 5,000 GB+, versus Bright Data’s $4.00/GB pay-as-you-go rate and $2.50–$3.50/GB tiered plans. For a one-off project measured in tens or low hundreds of gigabytes rather than a production scraping operation measured in terabytes, that difference is the gap between a proxy line item that fits inside a small grant budget and one that doesn’t.
| Dimension | Bright Data | Thordata |
|---|---|---|
| Best for | IRB-reviewed / human-subjects data collection requiring audited, ethically-sourced residential IPs | One-off, non-human-subjects research scraping — citation/literature harvesting, public web data |
| Residential proxy pay-as-you-go price | $4.00/GB (promotional rate off $8.00/GB list, no commitment, as of Aug 2026) | ~$2.00/GB at low volume, down to ~$0.65/GB at 5,000 GB+ (as of Aug 2026) |
| Cheapest tiered rate | ~$2.50/GB (798 GB / $1,999 per month) | ~$0.65/GB (5,000+ GB) |
| Free trial | No card required; trial usage tracked but not billed; no published fixed GB cap for the proxy network | Free trial available for residential proxies; no published fixed GB cap on the public pricing page either |
| Compliance / ethical sourcing documentation | Audited residential IP sourcing, KYC, documented abuse controls | Not positioned or documented as an audited/compliance-grade network |
| Minimum commitment | None on pay-as-you-go; tiers are monthly plans | None — pay-as-you-go by GB |
| Good fit for a single small research project | Only if human-subjects/IRB stakes require it — otherwise more capacity/cost than needed | Yes, for non-human-subjects public-data or literature-harvesting scrapes |
The honest tradeoff: when you should stay with Bright Data
This is the part worth stating clearly rather than burying under a comparison table: if your project involves IRB-reviewed human-subjects data collection, stay with Bright Data. Its residential IP network comes with the audited sourcing, KYC, and compliance documentation that an IRB or a research-ethics review is likely to ask about, and that’s exactly the kind of assurance you do not want to be the first to test with a cheaper, less-audited alternative. Thordata is a genuinely good fit for lower-stakes, non-human-subjects scraping — citation and literature harvesting, public web data collection, building a dataset from openly available sources — but it is not the product to reach for when the data being collected is about people and a compliance reviewer will be asking questions about your infrastructure’s sourcing later. Match the tool to the actual stakes of the project, not just the price.
When Thordata is the better fit
For everything short of that IRB/human-subjects threshold — a single literature-review scrape, a citation-mapping project, a one-off public-web dataset build for a paper or a course — Thordata’s lower per-GB pricing and simpler sign-up flow make it the more sensible default. It’s the same conclusion CASRAI’s own Thordata review and guide to proxies for web scraping in academic research reach in more depth: for a defined, non-human-subjects collection project, there’s rarely a reason to pay Bright Data’s enterprise rate for capacity you won’t use.
FAQ
What are Bright Data’s free trial limits?
Bright Data’s residential-proxy sign-up doesn’t require a credit card and trial usage isn’t billed while no card is attached, but as of August 2026 the company doesn’t publish a fixed free-GB allowance for that trial on its pricing page — plan around the paid per-GB rates rather than counting on a specific free allocation. Its separate scraper API products do publish a concrete limited free tier (on the order of a few thousand free records/month).
Does Bright Data have a proxy free trial specifically, not just its scraper APIs?
Yes — the no-card-required sign-up applies to the residential proxy network itself, not only the API products. The difference is that the proxy network’s trial has no stated GB cap, while the API products’ free tier does have a published limit.
Bright Data vs Thordata for researchers — which should I pick?
If your data collection is IRB-reviewed and touches human subjects, pick Bright Data for its audited, ethically-sourced residential network. If it’s citation/literature harvesting or public web data with no human-subjects dimension, Thordata’s lower per-GB pricing is the more cost-effective choice for a one-off project.
Is Bright Data worth it for a small research project?
Only if the project’s compliance stakes justify the premium — IRB oversight, human-subjects data, or a need for documented ethical sourcing. For a smaller, lower-stakes literature or public-data scrape, Bright Data’s enterprise pricing is usually more than the project needs.
