What does web scraping actually cost?
Compare scraping prices without mixing subscriptions, credits, actor start events and per-result rates. Learn the formula and the hidden cost lines.
The short answer
A scraping bill usually combines a plan, a billable event and usage charges. The useful comparison is the cost of the records you need after the input, retries and field add-ons are included. A price per 1,000 results is a starting unit, not the final invoice.
Catalogue and method checked 2026-11-12. See how tools are scored.Scraping prices are hard to compare because vendors sell different units. One charges per 1,000 results, another per actor start, another by compute time, and a marketplace adds platform usage on top. Convert every option to the same job before comparing the number.
The basic formula
For a published per-result rate, the estimate is rows divided by 1,000, multiplied by the price per 1,000. A 10,000-row job at $0.99 per 1,000 is $9.90 before start events, retries, storage or plan minimums.
A subscription is not automatically more expensive. If the plan includes usage credit that covers the job, the effective cost can be lower than a pay-as-you-go rate. The comparison changes with volume, so compare a realistic monthly run rather than a single row.
Cost lines a per-1,000 rate can hide
Actor start events are often charged once per run, so many small runs cost more than one larger run. A field enrichment may be a separate priced event. Proxy traffic, storage and compute can be bundled or billed separately depending on the platform.
- Plan minimum and included usage credit.
- Actor start or run event.
- Per-result or per-dataset-item rate.
- Enrichment add-ons such as email or contact lookup.
- Proxy, storage and compute usage where not bundled.
- Retries and failed runs that still consume paid usage.
How to compare a one-off with a monthly workflow
For a one-off list, compare the total bill for the final row count, including a test run. For a recurring workflow, compare the monthly plan plus overage at the expected volume. Do not compare a monthly subscription with a single-run estimate without saying which horizon you are using.
The catalogue cost estimator normalises published per-result rates for a target and a row count. It deliberately excludes plan differences and start events, and it says so. Use it to narrow the field, then confirm the billing model on the Store page.
Questions people ask
Is the cheapest per-1,000 tool always the cheapest job?
No. Start events, add-ons, retries, plan minimums and missing fields can change the total. A cheap tool that requires a second enrichment step may cost more than a tool with the fields included.
How do I estimate a scraping run?
Multiply rows by the published per-1,000 price, then add start events and any separate enrichment. Run a small test first and compare the record count and field completeness, not only the advertised rate.
Are free credits enough for a real list?
They are useful for a test, not usually for an ongoing list. Check what the credit covers and whether add-ons consume it faster. A free test can still reveal a schema mismatch before you pay for volume.
Why does a price change between the listing and the invoice?
The published primary rate may exclude platform usage, plan differences, failed runs or add-ons. The invoice follows the actual billable events. Confirm the pricing model before scaling a run.
Sources
These sources support the general method on this page. A specific tool's price, schema or access route is on its target page with the check date.
Pick the website you need data from
Each website has a route map: every catalogue Actor for it, the fields each one returns, the price per 1,000 results and the official API beside them. Or search all 394 Actors by the field you need.