Submit up to 1,000 domains and get the published overview for each - global, category and country rank with their movement, bounce rate, pages per visit and average visit duration, plus the company behind the site. Third-party estimates, and the page is explicit about that.
This distinction decides whether the numbers you get are useful or misleading, so it goes first rather than in a footnote.
Nobody outside a company can measure its website traffic. Figures like these are modelled - inferred from panels, sampling and other signals, then published as an estimate. That's a legitimate and widely used methodology, and it is not the same thing as the numbers in someone's own analytics.
The practical rule: these figures are strong for comparison and weak as absolutes. Ranking three competitors against each other, or watching whether a rival's rank is climbing over successive months, is exactly what this data supports. Quoting a bounce rate as though it were measured fact - in a board pack, a pitch or a valuation model - is where teams get into trouble.
For your own sites, your own analytics are authoritative and this isn't a substitute for them. Its value is that it covers domains you have no access to.
A domain is the only input. Nothing else to configure.
A bare domain or a full URL both work - apple.com or https://www.google.com/.
Each domain returns the query you submitted plus an overview object. These are the documented fields inside it.
US.00:02:27.15 documented fields per domain - the overview block is the one the reference specifies.
Each of these has caused a wrong chart somewhere. They're cheap to handle if you know first.
The documented value is a decimal like 0.524, not 52.4. Multiply before you display it, or your dashboard will report a rate of half a percent.
The field name ends in _formatted and the documented value is 00:02:27. You'll need to parse it before you can average or sort on it numerically.
Both end in _min, and the documented example shows band boundaries rather than real values. Read them as "at least this many" and never as a precise headcount or revenue.
The documented block covers engagement and rank, not a visit count - which is the number most people expect. If a monthly-visits figure is what your analysis needs, check the output on the free tier before you plan around it.
Four of the fields here describe the company rather than the website, which invites a comparison with another service.
Company Insights is the firmographics product: twenty flat columns, an exact founding year, a normalised industry token, full split address, phone and social accounts. This service returns four company fields as context alongside the traffic picture, and two of them are band floors rather than figures.
So if you want to know who a company is, use Company Insights. If you want to know how its website performs relative to others, that's this page - and the company fields here are best treated as a label on the traffic data rather than as firmographics you'd build a segmentation on.
All three are comparative, which is what this data is good for.
A single rank tells you little; the same rank measured repeatedly tells you a lot. Fix a set of competitor domains, re-run on a cadence, and the movement fields plus your own diffs show who is gaining ground - the direction being far more reliable than the absolute figure.
When someone pitches a partnership or an ad placement on the strength of their traffic, an independent estimate is a useful counterweight. You're not looking for a precise number, only for whether the order of magnitude matches the story.
Run every domain in a sector and sort by category rank to see the shape of the market - who leads, how steep the drop-off is, and where the mid-tier sits. Engagement fields add texture: two sites at similar ranks can behave very differently.
No subscription, no minimum, no per-seat licence. Your first 500 rows are on us - after that, pay-as-you-go.
Every new account, one-time. No credit card required. Enough to pull a competitor set and see which fields populate for the domains you actually care about before committing.
Roughly $2 per 1,000 domains, one row per domain - easy to budget exactly. Considerably cheaper than a seat on an analytics platform when all you need is a periodic comparison.
Volume pricing, SLAs, dedicated workers and bespoke onboarding for teams tracking a large domain set on a schedule. Tell us your numbers and we'll quote.
LIVESCRAPER10 at checkout.Competitive research usually needs more than one view of the same company.
The firmographics counterpart from the same input: twenty flat columns with an exact founding year, industry token, address and socials. Use it when the question is who the company is rather than how the site performs.
Rank data of a different kind - where a domain actually appears for the queries you care about, with a documented field set. Search visibility and site-wide rank answer different questions.
Most services here read public listings. This one reads an analytics vendor's published pages, which is a different situation and worth saying so.
The overview pages are publicly viewable, and collecting publicly available information for research is long-established practice. But the figures are a commercial product rather than incidental public data - a vendor compiled and modelled them, and sells access to them. That matters in two ways.
First, terms: automated access is restricted by the site's own terms of service, so this remains a terms question rather than a settled right. Second, redistribution: publishing someone else's compiled dataset as your own is a separate risk from collecting it, and in some jurisdictions compiled databases attract protection in their own right. Using these figures for internal research and comparison is the ordinary case; republishing them as a dataset is not something we'd advise without your own legal view.
There's no personal data involved here - these are facts about websites and companies - so the privacy questions that dominate our contact services don't apply. On our side: nothing behind a login, no third-party trackers on the data layer, and exports auto-delete after 30 days.
The questions we hear most. Anything else? Talk to us - humans, not bots, write the answers.
overview block covers rank, rank movement, bounce rate, pages per visit and average visit duration - engagement and position rather than a visit count. Since a monthly-visits number is what many people assume they're buying, check the actual output on the free tier before you build a report around it.bounce_rate is a decimal fraction, not a percentage - the documented example is 0.524, meaning about 52%. Multiply by 100 when you display it. It's the single most common mistake with this field._min and they're the lower bound of a band. Treat them as "at least this many" rather than a figure. If you need firmographics you can segment on properly, Company Insights is the service built for that.Paste your competitor set and get rank, movement and engagement for each - the comparative view that matters, with the estimate caveats stated plainly. Your first 500 rows are free.
Competitive research runs into the same wall every time: you can measure your own website precisely and you can measure nobody else's at all. Third-party rank and engagement estimates exist to fill that gap, and the SimilarWeb Scraper collects them in bulk. Submit up to 1,000 domains - a bare domain or a full URL both work - and each returns an overview block carrying global, category and country rank along with the movement on each, the two-letter code of the country the country rank refers to, bounce rate, pages per visit, average visit duration, and four fields describing the company behind the site. Fifteen documented fields per domain, one row each, exportable as CSV, JSON or Excel.
The most important thing on this page is not a feature. Figures like these are modelled rather than measured: nobody outside a company can observe its traffic, so estimates are inferred from panels and sampling and published as approximations. That methodology is legitimate and widely relied on, and it has a specific consequence for how you should use the output. These numbers are strong for comparison and weak as absolutes. Ranking three competitors against one another, or watching whether a rival's position is climbing across successive runs, is exactly what the data supports. Quoting a bounce rate as measured fact in a board pack or a valuation model is where the same data becomes a liability. For your own properties your own analytics remain authoritative; the value here is coverage of domains you have no access to.
Four details will save whoever writes the importer an afternoon. Bounce rate arrives as a decimal fraction rather than a percentage - the documented example is 0.524, meaning roughly 52% - and displaying it unscaled produces a rate of half a percent, which is the most common error with this field. Average visit duration is a pre-formatted string such as 00:02:27 rather than a number of seconds, so it needs parsing before it can be sorted or averaged. The two company size fields both end in _min and represent the lower bound of a band rather than a figure, so they should be read as "at least this many" and never as a headcount. And the documented block contains no total-visits field at all, despite monthly visits being the metric most people expect, which is worth confirming against real output on the free tier before any report depends on it.
Two of the fifteen fields overlap conceptually with Company Insights, which is the dedicated firmographics service: twenty flat columns with an exact founding year, a normalised industry token, full address, phone and social accounts. The company fields here are context for the traffic data rather than a segmentation basis, so the split is straightforward - Company Insights answers who a company is, this answers how its website performs relative to others. On the legal side, this service differs from most of the catalogue in one respect worth flagging. The pages read are publicly viewable, but the figures are a vendor's compiled commercial product rather than incidental public information, which makes automated access a terms question and republication a separate risk from collection, since compiled databases attract protection in some jurisdictions. Internal research and competitive comparison is the ordinary use; publishing the figures as your own dataset warrants your own legal view. No personal data is involved, nothing behind a login is touched, and exports auto-delete after 30 days. Start free: your first 500 rows cost nothing and need no credit card.