How to evaluate a competitor’s traffic without access to their analytics

Choose competitors for a specific customer job

Make three lists: businesses selling an alternative, publications competing for the same searches, and substitutes customers use instead of buying. A local marketing consultant may compete with another consultant, a software tutorial, and doing nothing. Choose three to five relevant sites. A famous brand with an unrelated audience can teach you very little about your next customer.

Write the market, language, device scope, and time period you will compare. Analyse domains, subdomains, and individual pages consistently. Comparing one company’s entire international domain with your local service page creates an impressive but unhelpful gap.

Know what the tools estimate

Similarweb blends multiple data sources and models to estimate website activity. Ahrefs estimates organic search clicks from its keyword and ranking data; that figure is not total website traffic. Different products and reports answer different questions. Read the metric description before copying a number into a spreadsheet.

Neither estimate gives you unrestricted access to a competitor’s GA4, conversion rate, customer list, or profit. No result or a very small estimate is not proof of zero visitors. Use relative patterns and repeated observations, especially when coverage is limited. Avoid averaging two tools and calling the result “actual traffic.”

Run a focused one-hour review

1. List the three customer questions most closely related to your offer. Search them and note which page types appear: guides, product pages, directories, videos, or local results.
2. Open five relevant pages on each competitor. Record the question answered, distinctive evidence, format, destination offer, and next action.
3. If you have a traffic tool, inspect estimated top pages, organic topics, trend direction, countries, and referral sources with matching settings. Export the date and metric definition.
4. Look at visible distribution: newsletter, original research, partnerships, videos, directories, and public ad libraries. An ad’s existence does not prove it is profitable.
5. Choose one gap you can fill with stronger usefulness or a different audience focus.

A free version or manual research can be enough to identify a first test. Do not buy a subscription just to admire a domain total. Pay for a tool only when a repeated decision needs the additional data.

Build an evidence ledger

Add a confidence column and the original URL. “Observed on the page,” “estimated by tool,” and “our hypothesis” should never be interchangeable labels.

Turn a gap into an original page

Imagine competing consultants publish broad articles about marketing budgets, but none provides a clear template for a solo owner choosing a first $300 experiment. Interview a few suitable owners, document the decision they struggle with, and build an original planning example. Link it to a relevant consultation or worksheet. Do not copy the competitor’s text, images, layout, or unsupported claims.

Your hypothesis might be: a small-budget planning guide will bring more suitable consultation inquiries than broad marketing tips. Measure its qualified inquiries, resource use, and customer questions—not whether your traffic estimate catches up with a large publisher.

Review what happened, then decide whether to repeat

Keep a dated baseline, a distribution log, and a review window appropriate to the channel. Direct outreach can produce feedback quickly; search discovery often takes longer. Revisit the competitor set periodically without letting it replace customer research.

Your deliverable today is one page with three observed patterns, two uncertainties, and one test. The best competitor research ends with work that is useful to your customers, even if the rival’s estimated traffic was wrong.