Want to know how much traffic a competitor takes from search each month? You don’t need to get into their backend. Public data is enough to piece together a usable range — as long as you don’t trust any single tool’s number blindly.
Estimate competitor traffic with the three-source cross-check method: third-party tools give you a total range, your own ranking data reverse-engineers traffic on shared keywords, and keyword difficulty weighting works out where the share gap sits. With three lines calibrating each other, the error shrinks to a level you can actually make decisions on — not just report upward.
Use third-party tools to bracket the total range
Similarweb gives site-wide monthly visits, including direct, social, and paid; Ahrefs and Semrush Site Explorer give organic search estimates. These numbers usually differ by 20–40%, because each vendor’s click-curve model and keyword coverage differ.
The right use isn’t picking one to believe — it’s treating them as the bounds of a range: take the lowest as the floor, the highest as the ceiling, and the middle value only as reference. When reporting externally, give a range; it holds up to questioning far better than a fake number precise to the last digit.
| Data source | What it provides | Main source of bias | How to use it |
|---|---|---|---|
| Similarweb | Site-wide monthly visits and channel share | Includes non-search traffic; small sites have thin samples | First derive the search portion from channel share |
| Ahrefs | Organic search estimate and keyword count | Incomplete long-tail coverage, conservative | Use as the floor of the range |
| Semrush | Traffic estimate and keyword distribution | Clear geographic differences in its database | Cross-check against Ahrefs |
| Competitor’s site itself | Number of sections and update frequency | Can’t be converted directly into traffic | Judge content volume and investment intensity |
For small competitors with under 10,000 monthly visits, third-party estimates are basically unusable — the sample is too thin and the error often exceeds double. For that kind of rival, skip straight to the reverse-engineering method in the next section.
Reverse-engineer the competitor with your own data
You and the competitor are certainly ranking for a batch of shared keywords. You hold real click data for that batch — that’s the most reliable yardstick. Use your own known “ranking → clicks” relationship to infer what the competitor earns on the same set of keywords.
- Export keywords where you both rank, taking a 20–50 sample with a reasonable volume spread
- Record each keyword’s position for both sides, plus the clicks you actually get from it
- Fit a click-curve from your own data — categories differ a lot, don’t just apply a generic value
- Plug the competitor’s positions into your curve to estimate their clicks on this batch
- Use the share of shared keywords in the competitor’s total keyword set to extrapolate site-wide
Step 3 is the key. Generic click curves come from cross-industry averages and often deviate by more than double in a specific category. Fitting from your own Search Console data noticeably improves accuracy. While doing this you’ll also see your own ranking structure clearly, which pairs well with an analysis of the distribution issues hidden by average positions in rank tracking.
Break down the competitor’s traffic structure
Total volume is only the first layer. What really shapes your play is the distribution: is the competitor’s volume concentrated in a few posts, or spread evenly across hundreds of pages?
- Highly concentrated, with the top 10 pages taking the bulk: a few hits are carrying the site. Build deeper, fresher pages on the same topics and you can carve off a meaningful slice
- Evenly distributed, with the top 10 pages’ share very low: they win on section breadth. Single-point breakthroughs don’t work — you have to fill the surface, building pages in batches around topic clusters
- High brand-term share: the search advantage actually comes from brand awareness; fighting this head-on is poor value, so go after non-brand keywords instead
Copy the titles, structure, word counts, and update dates of the competitor’s 20 highest-traffic pages into a sheet. Read through them a few times and you’ll grasp their content playbook — long lists, deep tutorials, or tool-type pages. The playbook is worth studying more than any single piece of content.
Find where the share gap sits
The gap is the batch of keywords where the competitor ranks and you don’t — or where you sit beyond position 20. Sort these by “monthly volume × business value ÷ difficulty” and you get your fill-in list.
Don’t take difficulty scores at face value; the KD tools give is mostly an approximation on the backlink dimension. For how to read it, see how to correctly read keyword difficulty scores: first check whether the top 10 includes any site close to your size — if so, the term is beatable; if not, set it aside for now.
- Export both keyword sets and take the difference to get the gap terms
- Remove the competitor’s brand terms and clearly irrelevant ones — these two categories often make up a big share
- Prioritize the batch with medium difficulty and solid volume: fastest to show results and easiest to verify the method works
- Put high-difficulty head terms on the list and wait until your backlink resources and content base catch up
Build a monthly tracker and watch changes, not absolute values
The absolute precision of an estimate is always limited, but the month-over-month change produced by the same method is trustworthy. Take a snapshot at a fixed time each month and focus on the direction of trends and the share gap — don’t obsess over decimals.
| Month | Competitor estimate range (10k) | Your actual (10k) | Share-gap trend |
|---|---|---|---|
| Month 1 | 110–135 | 80 | Baseline |
| Month 2 | 118–142 | 88 | Slightly narrowing |
| Month 3 | 115–140 | 97 | Consistently narrowing |
If the competitor suddenly jumps one month, don’t panic — go see what they published. Most of the time it’s a new section going live, or one piece picking up a batch of backlinks. Tearing down that section’s structure is far more useful than staring at the total number.
When it actually comes to execution, it’s really five things: pick one main competitor and pull an organic estimate from three tools, noting the floor and ceiling; export 20–50 keywords where you both rank and fit a click curve from your own click data; organize the competitor’s 20 highest-traffic pages into a sheet and judge whether they’re concentrated or breadth-driven; take the difference to find gap terms, sort by volume and difficulty, and pick the 30 to fill this quarter; then build a monthly snapshot table, updating on the same day each month, watching trends rather than absolute numbers. After one round you’ll have a real feel for the competitor’s “volume” and “playbook” — whether you’re chasing share or avoiding a head-on fight, you’ll know where to put your effort.
