You can't log into a competitor's Google Analytics. There's no dashboard anywhere that shows you their real, verified visitor count, and anyone who tells you otherwise is guessing. What you can do is run a proper competitor traffic analysis: build a solid, honest estimate of their website traffic, compare it to your own numbers on equal terms, and read the trend and channel mix well enough to make a real decision. If you've been asking yourself how much traffic does a competitor get compared to you, the answer is never one clean number from one tool. It's a small, defensible process. This guide walks through the seven steps that get you there, using a total-traffic estimator alongside your own Google Analytics 4 and Search Console data, so the comparison holds up when someone asks how you got the number.
A lot of teams skip straight to typing a domain into a free checker and taking whatever number comes back at face value. That's how you end up in a meeting quoting a figure nobody can defend, and it falls apart the moment someone asks what period it covers or whether it includes mobile. The steps below take a little longer, but they leave you with a number, and a story behind that number, that actually holds up.
Why the terms matter before you compare website traffic
Before you pull a single number, it helps to know what you're actually looking at, because different tools answer different questions and use the same words to mean different things. Public data on organic traffic benchmarks by industry also helps you sanity-check whether an estimate looks plausible at all before you build a whole conclusion on it.
| Term | What it means here | The trap |
|---|---|---|
| Visits or sessions | A repeatable site visit. One person can generate several in a day. Similarweb reports estimated visits, GA4 reports sessions. | Don't treat a visits figure as a count of unique people, and don't assume two vendors define a visit the same way. |
| Monthly visits | Similarweb's estimate of a site's average monthly traffic across desktop and mobile for a chosen period. Its Total Visits metric sums visits across the whole selected window. | A three-month total isn't the same unit as a monthly average. Read the label the tool actually shows before you copy the number. |
| Unique visitors or users | An estimate of distinct people or devices, defined by whichever provider produced it. | Never place it next to a sessions figure as though the two are interchangeable. |
| Search clicks | Search Console's count of clicks from Google's results into your own property. | It's neither all-channel traffic nor guaranteed to match GA4's organic-search sessions. |
| Estimated organic traffic | Ahrefs' modeled search traffic only, not total traffic across direct, email, referral, social, and paid. | Don't set an Ahrefs organic number next to a Similarweb total-visits number as if they answer the same question. |
| Channel share | One channel's estimated visits divided by all estimated visits, for the same domain, period, and filters. | A channel's share can go up even while its own visits fall, if total traffic falls faster. |
Keep this table nearby while you work through the steps. Most disagreements between two "traffic numbers" for the same competitor trace back to one row in it.
Step 1: Define the sites and the comparison you actually need
Before you open any tool, write down your domain and the domain of each competitor you're checking. Decide whether you're comparing entire websites or just a piece of them, like a blog, since those are two different questions. Pick the same month (or a matching multi-month window), the same country or worldwide market, and the same device view (desktop, mobile, or combined) for every site in the comparison. If your product and site structure are roughly similar to a competitor's, start with a whole-domain comparison. It's simpler and less likely to hide a mismatch.
You should be able to state the comparison in one line: which domains, which dates, which geography, which devices, which unit. If you can't say that sentence, you're not ready to pull a number yet.
Common mistake: comparing a competitor's worldwide, entire-domain traffic against your own US-only blog numbers, or a desktop-only chart against an all-device figure. If you genuinely can't isolate equivalent scopes, say so in your notes instead of treating two mismatched numbers as comparable.
Step 2: Establish your own measured baseline first
Before you estimate anything about a competitor, get a solid read on your own site, because that's the one number you can actually trust. In Google Analytics 4, open Traffic acquisition for a session-scoped view of where your visits come from. Pull the sessions total for your chosen period, along with the breakdown by channel or source your property reports. Save the period and filters next to the numbers you export, because a figure with no context attached is nearly useless a month later. It also helps to know what counts as organic traffic in GA4 before you treat that channel row as complete.
Use User acquisition only if your question is really about where new users first arrived. It's scoped to first-user attribution, which answers a different question than session-source dimensions do, and mixing the two up leads to numbers that don't add up the way you expect. If search performance matters to your comparison, also open your verified Search Console property and pull Google Search clicks for the same period from the Search results Performance report.
You'll end this step with an internally measured sessions total and a channel breakdown, plus a separately labeled Google-search click total if you need it.
Where people go wrong: Search Console clicks aren't the same thing as total website sessions. Its anonymized-query handling means the visible query rows can actually sum to less than the reported total. If you want the fuller picture of why so many Google Search Console queries are hidden, that's worth a separate read before you lean too hard on the click totals. GA4's first-user and session-level attribution can also split traffic differently depending on which report you're reading. Call this number your measured baseline, not a universal count of every human who visited your site, because your own tracking setup shapes what it captures too.
Step 3: Estimate whole-site traffic in one comparable tool
Now look up your domain and the competitor's domain in the same total-traffic estimation product, using the same settings for both. In Similarweb's website-performance view, use +Compare to add up to four more sites to the one you started with, then set the time period, geography, and All Traffic / Desktop / Mobile filters the same way for each. Read the metric the interface actually labels, whether that's Monthly Visits or Total Visits, and write down which one it is. You can do the same comparison in Semrush's Traffic & Market tool, using Traffic Analytics to inspect both domains side by side. Ahrefs has also written its own comparison of ways to estimate a competitor's traffic, if you want another perspective before you commit to one tool.
A free public checker is a fine first look, but don't assume its free view shows the same geography, history depth, or detail a paid plan would. Build a small comparison sheet: site, provider, metric, period, geography, device, domain scope, and the estimate itself, for both your site and the competitor's, from the same provider under the same settings. This sheet is the actual deliverable of a competitor traffic analysis; the single number people ask for is really just its last column.
Pricing varies by provider, and it's worth knowing roughly what you're signing up for before you pick one. Similarweb's Starter plan starts around $199 a month (or $1,500 billed yearly), with one user and three months of history; Professional starts near $399 a month and extends that to six months. Semrush's Traffic & Market toolkit lists a Pro tier at $289 a month. Both offer a limited free public checker if you just need one quick look rather than an ongoing comparison.
Where people go wrong: treating a modeled number as if it were access to a rival's real analytics. Similarweb and Semrush build their estimates differently from how GA4 measures your own site, and the two can disagree meaningfully, a difference Semrush documents in its own comparison of traffic analytics data against Google Analytics. Similarweb's country-level filtering also depends on your plan tier, so an entry-level account may only show you a worldwide view.
Step 4: Calculate the relative gap, without adding false precision
Divide the competitor's estimated visits by your own site's estimated visits, using numbers from the same tool and the same configuration. That gives you a within-provider traffic ratio, and it's the number worth reporting. For example, if a tool estimates 45,000 monthly visits for a competitor and 30,000 for your site, that's a 1.5x ratio, a roughly 50% estimated lead for the competitor. Those specific figures are illustrative, not a benchmark to expect. Some teams track this ratio over time as a form of share of voice against named competitors, a discipline that extends naturally into AI search results too.
It's also worth comparing the tool's estimate for your own site against your real GA4 sessions, just to see how closely that provider tracks reality for you. If GA4 shows 24,000 sessions while the tool estimates 30,000 visits for your own domain, note the gap. Don't apply that same correction factor to the competitor's number, though, since that assumes the tool is off by the same amount for a site you don't actually measure, which you have no way to confirm.
Common mistake: averaging an Ahrefs organic-search estimate with a Similarweb total-visits figure. They measure different slices of traffic, and neither the average nor the midpoint of two providers is a verified count. Where tools disagree, describe a range of estimates or point out what they agree on directionally. Don't call that a statistical confidence interval, because it isn't one.
Step 5: Read the trend before you try to explain it
Pull the same provider's monthly estimates for each site across the same window and look at the shape, not just one snapshot. Is the competitor's lead steady, growing, shrinking, or was it really just one unusual month? Calculate the period-over-period change as (later visits minus earlier visits) divided by earlier visits, times 100. Where you have enough history, compare the same calendar month a year apart, since that separates a seasonal pattern from a genuine shift. It's worth knowing how seasonality shapes organic traffic benchmarks in general, since a single month rarely tells the real story. Look at a multi-month window before treating any single estimated spike as a strategic gain.
Report the time window you used, the direction each site moved in, and whether the estimated gap widened or narrowed. A simple two-line chart or a short table of the underlying monthly values can make this easier to follow than prose alone. If your own trend line is the one moving rather than the competitor's, that calls for a different diagnosis: work through how to diagnose an organic traffic drop before you draw any competitive conclusion.
Where people go wrong: pinning a modeled spike on a specific campaign with no independent evidence for it. Also watch your history limits: Similarweb's Starter plan shows three months of history and Professional shows six, so a year-over-year comparison simply isn't available on those tiers, and it's better to say that plainly than to fake it with a shorter window.
Step 6: Decompose the gap by channel, then check the search slice separately
Open the channel breakdown in your total-traffic provider for both domains, under the same market, period, and device settings. Compare both the share of traffic by channel and, where the interface supports it, the estimated visits by channel. The provider's own labels will usually separate direct, organic search, paid search, referral, social, and email traffic. A competitor can have more total visits while getting a smaller share of theirs from organic search, and that distinction matters for what you do next. It's also worth reading your own branded vs non-branded organic traffic split separately, since that tells you how much of your organic strength is demand you already own.
For a second, narrower view of the search picture, run Ahrefs' traffic checker or Site Explorer on both domains. It reports estimated organic and paid-search traffic specifically, which you can compare against your own Search Console clicks as a plausibility check, while keeping the two metrics labeled separately since they aren't measuring quite the same thing. Once you know which channel is doing the work, deciding whether it's worth chasing further is really a paid and social channel economics question.
Where people go wrong: calling a "direct" traffic share proof of brand demand, assuming a paid-search campaign from a chart alone, or treating an Ahrefs organic estimate as if it described the competitor's whole channel mix. Search Console validates your own Google Search performance. It has no visibility into a competitor's private Search Console account.
Ahrefs' paid plans run from around $129 a month on Lite up to $449 on Advanced, with a free version of the traffic checker for a quick look at a single domain. It's worth remembering that Ahrefs has published its own comparison of Similarweb's estimates against verified figures from sites it had access to, and found meaningful overestimation for unique visitors and pageviews in that particular sample. That's one dated study on one sample, not a general accuracy rating for every domain you'll ever look up, so use it as a reason to stay cautious rather than a number to quote as fact.
Pro tip: keep your comparison sheet from step 3 open while you do this step, and add a channel column to it. A gap that looks huge on the total-visits line often narrows or widens once you see which channel is actually driving it, and that's usually the detail someone in the room will ask about first.
By this point in the process you've touched all three of these numbers separately: your measured GA4 sessions, the modeled total visits from step 3, and the narrower organic-search estimate from this step. Keeping them visually distinct is what stops someone from quietly averaging a search-only figure with a whole-site one later.
Step 7: State your conclusion, and be honest about what it doesn't prove
Pull it together into one decision-ready statement: which sites and market you compared, the dates and devices, the provider and exact metric name, the estimated traffic ratio, the direction of the trend, the leading channel differences, how that compares against your own measured baseline, and any real disagreement between tools or gaps in the data. Say plainly what the evidence doesn't establish, especially an exact competitor session count or a firm explanation for why their traffic moved.
Someone else should be able to take your stated inputs and reproduce the comparison, and tell at a glance which numbers are your own measured data and which are modeled estimates for a site you don't control. That's what it actually means to compare website traffic responsibly: the process is reproducible even when the underlying numbers are estimates.
Where people go wrong: reporting extra decimal places that imply precision you don't have, quietly picking whichever tool gave the biggest number, or presenting a directional estimate as a settled fact.
This is really the whole answer to how much traffic does a competitor get: not a single verified figure, but a stated estimate, from a named provider, under named conditions, next to your own measured baseline. That's a more useful sentence in a strategy meeting than a number with no context behind it, because it survives the first follow-up question.
Where DeepSmith fits, and where it doesn't
DeepSmith's AI Visibility tracks a different thing entirely: how often your brand gets mentioned or cited when someone asks ChatGPT, Perplexity, Gemini, or another AI engine a question in your space, and how that compares to your competitors' presence in those same answers. That's a real and increasingly important signal, but it isn't a substitute for total website traffic, GA4 sessions, or a competitor's channel mix, and it shouldn't be treated as one. If you want to see how AI referral traffic specifically shows up in your own numbers, there's a separate walkthrough on how to measure AI search traffic in GA4. If your traffic comparison in this guide turns up a competitor pulling ahead, AI Visibility is worth a separate look afterward, specifically for whether they're also winning citations in AI answers for the topics your buyers are asking about. Pro tracks ChatGPT, Grow adds Perplexity, and Scale adds Gemini, so the platform grows with you as AI search becomes a bigger part of the picture.
What to do next
Build the comparison sheet from step 3, run the ratio and trend math from steps 4 and 5, and write the one-paragraph conclusion from step 7. That's a repeatable exercise you can rerun quarterly with the same providers and settings, which is what makes the comparison mean something over time instead of being a one-off number you can't defend later. Save the sheet somewhere your team can find it next quarter, since the real value of a competitor traffic analysis shows up on the second and third time you run it, when you can finally see a trend instead of a snapshot.
If you're also curious whether your content is showing up in AI answers the way it shows up in search, DeepSmith's free trial gives you real data on that within a week, no long-term contract required.
