Semrush vs Similarweb Numbers Don't Match? How to Compare Competitor Traffic in 2027

You may have noticed that the traffic numbers for three competitors differ between Semrush and Similarweb. One site may have 410,000 visits a month in one tool and 260,000 in the other. It can be tempting to rely on the tool you trust more or average the two figures. However, I recommend not doing either. The numbers from these two vendors rarely match, and that alone provides little insight.

What truly matters in competitor research is whether the two tools agree on the market structure: which competitor is larger, by roughly how much, and who is growing. You can usually determine this even when the totals are vastly different, as long as you follow a careful comparison process. Here's a step-by-step guide:

1. Ensure you're comparing the same metric:

Sometimes, Semrush and Similarweb disagree not because of a real discrepancy but due to comparing different metrics. Semrush provides two traffic figures for a domain using two different methods. Its Domain Analytics estimates traffic from keyword positions, considering search volume multiplied by an average click-through rate. Traffic Analytics, on the other hand, is built from clickstream data, covering all channels like direct, referral, search, social, email, and display.

Similarweb's headline figure, Total Visits, also covers the entire site, not just search. So, if you compare the organic traffic number from Semrush with Similarweb's Total Visits, you'll likely notice a significant gap. The definition of a visit also differs between the two tools. Similarweb counts a visit when someone opens one or more pages and stays inactive for over 30 minutes; a new visit starts at midnight.

Semrush defines a visit as a session, a set of actions a user takes on a website during a limited timeframe. These small differences in rules add up over millions of sessions, causing variations in reported numbers.

2. Adjust the settings to line up the numbers:

Both tools allow you to change settings, and their defaults may not match. Before examining any figures, set the following four filters to be identical in both tools:

- Device: All devices (desktop or mobile)

- Geography: Worldwide

- Subdomains: Excluded (consider only the root domain)

- Period: Monthly data

The subdomain row often catches people off guard. For instance, if a SaaS competitor's logged-in product lives on app.example.com, including it can dominate the total, while excluding it may make the marketing site appear modest. The period also matters. Semrush has moved to a daily processing model, which can cause some shifts in metrics. Comparing a month from spring with one from today may attribute some changes to the model rather than the site itself.

3. Compare competitors within each tool first:

Instead of directly comparing Semrush numbers with Similarweb numbers, compare competitors within each tool. Then compare the two results. For example, consider three competitors for one month, worldwide, all devices, and subdomains excluded:

- Site A: 410,000 visits (55% of the set in Semrush, 50% in Similarweb)

- Site B: 220,000 visits (29% in Semrush, 33% in Similarweb)

- Site C: 120,000 visits (16% in Semrush, 17% in Similarweb)

In this case, both vendors put Site A first, with roughly half the traffic, followed by Site B and C. The structure remains largely consistent across tools. For most competitor decisions, such as benchmarking or assessing market share targets, agreeing on the rank order is sufficient. Averaging the totals is unnecessary, as it produces a number that neither vendor would stand behind.

4. Read the disagreement, not just the numbers:

When the two vendors disagree on the structure, the type of disagreement can provide valuable insights. If the sites maintain the same order but differ in scale, report the order and the range for the gap. For example, Site C might have 120,000 visits in Semrush and 15,000 in Similarweb, while Sites A and B show little difference.

In this scenario, something specific to Site C is likely going on, such as subdomain settings not matching, most of its audience arriving through a channel one model sees poorly, or the site being small, so both models are working from limited data. In cases like these, Semrush publishes a Data Accuracy Report, which can help identify the root cause of the discrepancies.