Similarweb API pricing: data credits and how to estimate
Similarweb API pricing from Similarweb's docs: how data credits are counted, why no price is public, how to estimate a query, and per-call traffic data.
The looot team · · 7 min read
On this page
Similarweb API pricing has no public price list. The API is sold in custom packages, or on its own, through Similarweb's sales team, and what you buy is an allowance of data credits. What Similarweb does publish is how credits are counted, and that is enough to estimate what a query will spend before you call it.
This post reads Similarweb's pricing page and its data credits guide as they stood on 2026-11-03, shows how one request turns into credits, works through the two examples Similarweb gives, and covers the per-call route for website traffic estimates an agent can use today. One honest note first: Similarweb API endpoints are listed in looot's catalog from Similarweb's own docs, but they do not run through looot yet: each one needs a Similarweb API key, and looot has none.
Similarweb API pricing: what the pricing page says
Similarweb's pricing page shows no price, seat count or credit amount for any plan. It says the API is part of customized packages for businesses, and that you can also buy the API as a standalone product without a platform subscription.
| Question | What the page says |
|---|---|
| Is the API in a plan? | API is part of customized packages for Businesses |
| Can you buy only the API? | Yes, as a standalone product without a platform subscription |
| Data products listed | API, Data Feeds, Data Hub, Data Licensing, none priced |
| Price shown | None, packages go through sales |
Read on from similarweb.com/corp/pricing. No price, seat count or credit amount is shown for any plan or for the API.
In practice that means a sales call, a quote for a credit allowance, and a contract term. Unused credits expire at the end of each billing cycle, so size the allowance to what you will actually query.
How does the Similarweb API count data credits?
Most Similarweb endpoints charge one credit per data point returned. A data point is one result for one domain, one country and one period, so the same metric costs more for more countries, more months or finer granularity. Keyword endpoints are cheaper: 100 data points cost one credit.
| Rule | What the guide says |
|---|---|
| Most endpoints | 1 credit per data point returned |
| Keyword endpoints | 100 data points = 1 credit |
| Granularity | Monthly is 1 result a month, weekly about 4, daily about 30 |
| Countries | Credits multiply by the number of countries in the request |
| Example: visits, amazon.com, US, 6 months, monthly | 6 credits |
| Example: traffic sources, walmart.com, US and CA, 3 months, weekly | 576 credits |
| Errors and requests with no data | Not charged |
| Unused credits | Expire at the end of each billing cycle, no rollover |
| Over the allowance | Requests are throttled or blocked with a 429 error |
Read on from Similarweb's data credits guide. The guide gives no dollar price for a credit and sends you to your account manager. Total credits = domains x endpoint base cost x granularity x countries x time range x results per domain.
The formula in the guide multiplies domains, the endpoint's base cost, granularity, countries, time range and results per domain. Granularity is the one that surprises people: the same six months cost about four times as much by week as by month, and about thirty times as much by day.
How do you estimate a Similarweb query before you run it?
Write the request out as the formula's factors and multiply. Visits for one domain, one country, six months, month granularity is 1 x 1 x 1 x 6 data points, which is the guide's own six-credit example. Change one factor at a time to see what it costs: a second country doubles it, week granularity multiplies it by about four.
Similarweb's support and FAQ page suggests checking a large query with its Request Validate endpoint, which previews the cost before you run it, and lists a limit of 10 requests per second per account. Errors and requests that return no data are not charged, but a query that returns data you did not need is.
Can you get website traffic data per call instead?
Yes, as an estimate from another provider. Similarweb's own endpoints do not run through looot yet: each one needs a Similarweb API key, and looot has none. What runs per call today is a domain traffic overview from DataForSEO and a company traffic lookup from Datagma, priced per call from one prepaid balance. The table below reads their live price when the page loads.
These are estimates from other data sources, and they will not match Similarweb's numbers. Use them to compare sites with each other, not to reconcile with a Similarweb report. To check one site by hand first, the free website traffic checker runs the same lookup. The feature by feature view, including what has no equivalent, is on the Similarweb alternative page, and can I replace Similarweb lists which jobs an agent can take over. For the SEO side, see Semrush vs Similarweb.
Similarweb API or a per-call estimate?
Use the Similarweb API when you need Similarweb's own numbers, its audience and app data, or day-level history, and you can sign a contract for a credit allowance. Use a per-call estimate when your agent compares a list of competitor sites once in a while and you would rather pay per lookup than per month.
Run it with an agent
Paste this into Claude Code, Cursor or any agent that reads a skill file.
Prompt for your agent
Estimate traffic for five competitor sites
What it cannot do
- Similarweb's own data. Similarweb API endpoints do not run through looot yet: each one needs a Similarweb API key, and looot has none. looot does not resell Similarweb.
- Audience demographics and app analytics. No provider in the live table returns these.
- Tracking a site over time. Re-checking traffic on a timer needs scheduled runs, not available yet.
- Free credit. looot has no free trial credit; you top up a prepaid balance first. See pricing.
Last checked 2026-11-03. Similarweb facts were read that day from the pages linked in the tables and above. looot prices are read from the live catalog when the page loads.
Questions
Keep reading
Same job, different price: what data costs per provider
The same Google search is listed at 28 times the price depending on the provider. Five data jobs priced side by side, read from the looot catalog on 2026-10-11.
Walid Boulanouar · · 11 min read
One API key for many data APIs, without separate sign-ups
One API key for many data APIs: AnyAPI, FetchAPI, upAPI and looot compared on billing, free start and fallback, read 2026-10-23.
The looot team · · 8 min read
AI SEO agent: the SEO jobs it can run with live data
What an AI SEO agent can do with live data today: keyword research, SERP checks, trend checks and backlink audits. Costs, review steps and limits.
Walid Boulanouar · · 8 min read
Try it with the agent you already use
Start your workspace, top up, and paste one prompt into your agent.

