TOOL · RUNS IN YOUR BROWSER
Web scraping cost calculator
How much will a month of scraping cost? Compare the proxy bandwidth bill for running your own scraper with the per-page bill of a managed scraping API. Nothing you type leaves this page.
DIY BANDWIDTH VS MANAGED API
Calculator
Prices as of October 2026. Pick a preset or type your own numbers.
Assumptions and what is excluded
- Bandwidth = pages ÷ success rate × page size. Every failed attempt is counted as a full page, which overstates the cost of failures (blocked responses are usually small): treat it as an upper bound. At 74% success that adds up to about 35%.
- Proxy price is billed per GB as each provider measures it; the presets use entry or pay-as-you-go rates.
- API price is per 1,000 successful pages, assuming you use a plan's credits fully. All presets bill successes only, with fine print (ScraperAPI also bills 404s; Firecrawl bills 403/404 pages). Subscription plans have a monthly minimum ($19–$49 at the entry tier) whether or not you use them.
- Not included: servers or serverless compute for your own scraper; CAPTCHA solving (2Captcha lists reCAPTCHA v2 at $1–$2.99 per 1,000 and Turnstile at $1.45 per 1,000); engineering and maintenance time (the Web Scraping Club estimates about 30 minutes a month per simple scraper in 2026); data storage; taxes.
- Not comparable on capability: raw proxy bandwidth and a premium unblocker do not get you the same pages on a heavily protected site. Bandwidth is only cheap if the requests succeed; see the success rates in the scraping API guide and proxy guide.
Example results for 100,000 pages a month
Computed with the same formula, at 100% success. Multiply the DIY figures by about 1.35 for a 74% success rate.
| Page size | Evomi Core ($0.99/GB) | PacketStream ($1.00/GB) | Decodo ($4.00/GB) | Bright Data ($8.00/GB) |
|---|---|---|---|---|
| 100 KB (HTML-only fetch) | $9.90 | $10.00 | $40.00 | $80.00 |
| 500 KB (headless render, images/fonts/CSS blocked) | $49.50 | $50.00 | $200 | $400 |
| 2 MB (full headless render) | $198 | $200 | $800 | $1,600 |
| API configuration | Per 1,000 | Per month |
|---|---|---|
| ScrapingBee Freelance, no JS | $0.20 | $20.00 |
| ScrapingBee Freelance, JS | $0.98 | $98.00 |
| Bright Data Web Unlocker | $1.50 | $150 |
| PacketStream Fetch, no JS | $2.50 | $250 |
| ScrapingBee Freelance, premium + JS | $4.90 | $490 |
| ScraperAPI Hobby, premium + render | $12.25 | $1,225 |
Reading the result
Page weight moves the DIY number more than anything else. At 100 KB per page, even $8/GB bandwidth undercuts every managed API above about $0.80 per 1,000 pages; at 2 MB per full render, $4/GB bandwidth costs $8 per 1,000 pages, more than Bright Data's $1.50 unblocker or ScrapingBee's $0.98 JavaScript rendering. That is why blocking images, fonts and CSS (see the open-source tools guide) is the first thing to do before you pay for more bandwidth.
The second lever is the target. On lightly protected pages, DIY with a cheap residential proxy is hard to beat. On sites behind Cloudflare, DataDome, Akamai or Kasada, your success rate drops and every retry, CAPTCHA and hour of debugging lands on your side of the ledger, while an API that bills successes only absorbs it. The Web Scraping Club (independent, sponsored by Decodo and Oxylabs; article) puts the break-even for DIY on hard anti-bot sites above roughly 100,000 requests. The build vs buy guide walks through the full decision.