Where the number comes from, and what that costs you
Traffic estimates for a site you do not own are modelled, not measured. A panel provider observes a sample of browsing behaviour, extrapolates it, and produces a figure. That is a real method with a real error bar, and it is the same class of number as an ad spend estimate.
Which means it has to be labelled every single time it appears. We mark these as estimates in the data, on the page and in every row they touch, and they are never allowed into a score. A modelled number standing bare next to measured ones is the most common way a report becomes untrustworthy, because the reader has no way to tell which is which.
Read direction rather than the decimal. If a competitor sits at roughly two million visits a month, that is a fact about their scale. Whether it is 1.9 or 2.3 is not a fact about anything.
Read scale first, then the mix
Scale sets the strategy. A rival four times your size is not a company you outspend, and knowing that early saves a year of trying.
The mix decides what you actually do. Three quarters direct is a brand you cannot outbid, only outlast: people are typing them in, so the fight is on retention and reputation, not on ad auctions. Heavy on paid search is a company renting its demand, and rented demand stops the day the card does. Heavy on organic is a compounding asset that took years and will take you years, which is worth knowing before you commit to catching up.
Bounce and pages per visit are the quality check on all of it. Large numbers with a bad bounce rate is traffic that was bought badly, and it is one of the few places where a bigger competitor is quietly weaker than you.
The number that did not exist two years ago
Some share of their visits now arrives after somebody asked an assistant instead of a search engine. That share is measurable from the outside, and it is worth separating from everything else, because it is the one channel where being invisible has no warning sign.
It also splits a question most tools blur. Being named inside an AI answer and getting a visit from it are two different things, and the gap between them is one of the few genuinely new things to measure right now. The ranking half is here; this is the traffic half.
What it cannot tell you, stated plainly
Not their analytics. No sessions, no conversion rate, no revenue. Anyone selling you those for a site they do not own has invented them.
Small sites model worst. Below roughly ten thousand visits a month the estimate gets noisy enough that we would rather you ignored it than argued with it.
Small monthly moves are noise. A month over month change of a few percent sits inside the model's own error bar. If a report presents that as a trend, the report is decorating.
It says nothing about why. Traffic is a lagging indicator: it moved because of things that happened weeks ago. Pair it with the inputs that caused it or it is just weather.
How to do it here
Point us at a domain. Traffic runs weekly alongside everything else, lands on the Benchmark next to your own figures, and carries the estimate label in every row it appears in. The month by month history builds from the day you start, which is the only argument for starting now rather than later.
The longer version of what this page measures, with the caveats attached, is on the traffic page. The first run against one competitor is free.
Pick one competitor. See what we find.
Your first benchmark is free. One domain, measured against you, no card and no call.