Can you tell me exactly how much traffic my competitor gets?

No, and no honest tool can. Real clicks and impressions come from Search Console, and Search Console only ever shows you a property you own. Every number anyone quotes about a rival's traffic is a third-party index modelling what it thinks happens. We use those estimates for direction and relative size, never as fact, and we label them that way in the report so you do not repeat a modelled number in a board meeting as if it were measured.

Which competitors should I give you?

Give us the businesses you lose deals to, because that tells us what a buyer compares you against. Then expect the list to change. In search your competitor is whoever holds the result page for the queries that matter, and that is often a directory, a marketplace listing, a national publisher's roundup or a Reddit thread rather than a company you have ever pitched against. We reconcile both lists and tell you which of your named rivals are as invisible as you are.

Will this tell me how to overtake them?

It tells you where the gaps are, which of them are reachable and in what order to attack them. What it will not do is promise a position. Rankings sit with Google's algorithm, not with us. We commit to the analysis, the evidence and the build order, and we write those as scope. Overtaking a named competitor is a goal, and anyone selling it as a deliverable is selling something they cannot control.

We are a new site with almost no history. Is this still worth doing?

Yes, but it changes shape. With no Search Console history there is no first-party baseline to compare against, so the whole picture comes from external estimates and from reading result pages by hand. That makes the output less about closing measured gaps and more about deciding what to build first, which is usually exactly what a new site needs anyway. We say up front which half of the normal method is unavailable to you.

Do I need to buy Ahrefs or Semrush for this?

Not necessarily. Ahrefs Webmaster Tools is free on a site you have verified, Bing Webmaster Tools gives away both keyword and backlink data that goes unclaimed because the traffic share looks too small to bother verifying, and Keyword Planner and Trends cover demand. Paid tools genuinely help for competitor backlink depth, and we will say when the job needs one. What we will never do is run your data through a shared group-buy tool account, which is account sharing, against the terms, and puts your competitive research somewhere neither of us controls.

How is this different from keyword research?

Keyword research starts from your customer and asks what they type. A gap analysis starts from the result page and asks who is already being paid for those searches and why. They overlap in the middle and the outputs look similar, but the gap version finds demand you would never have thought to ask for, because someone else is already serving it and you can see the page they are serving it with.

SEO & Organic Growth

Finding what your competitors rank for and you do not

The queries, pages and links that send buyers to them instead of you, sorted by which gaps you can realistically take and which ones are not worth the year they would cost.

From: $200–$1,500
Typical time: 5–10 days
Usually bought by: Any marketer

What this actually solves

There are two versions of the person who lands here, and they need different things from the same piece of work.

The first has watched a rival appear everywhere. Their name comes up in searches, in roundups, in the answers that sit above the results, and the feeling is that they know something you do not. That feeling is usually right, but not in the way it seems. Rarely is there a secret. What there usually is, is a set of pages built for questions you never thought to build for, and a handful of links earned three years ago that keep paying out.

The second is spending on content already and cannot tell whether any of it is aimed anywhere. Posts go up, nothing happens, and the next month's plan is written the same way the last one was — from a meeting rather than from evidence. The expensive part is not the content budget. It is the twelve months spent proving that a topic had no demand, which anyone could have seen in an afternoon by looking at who ranks for it.

What this work produces is a list of specific gaps with an argument attached to each one. Not a data dump. Anyone can export a rival's keywords; the export is not the deliverable and on its own it is worse than nothing, because it invites you to chase the biggest numbers on it. The deliverable is a shortlist of gaps you can realistically take, an explicit list of the ones we are telling you to ignore, and the reason for both.

What we do, and in what order

The order is deliberate. Looking at competitors before looking at your own data is how people end up rebuilding a site around someone else's business model.

Work out who the competitors actually are. Your named rivals go on the list, then we open the result pages for the queries that matter and write down who is really there. Directories, marketplaces, publisher roundups and forum threads count, because they are taking the click whether or not they are taking the customer. Half a day to a day.
Establish your own baseline first, from real data. Search Console, three months minimum, filtered to non-branded queries — your own brand name inflates every chart and belongs nowhere near a gap analysis. We pull out anything sitting between position eight and twenty, because those are already earning impressions and are the cheapest ground in the whole exercise. Half a day.
Build the keyword gap and put an intent on every row. Competitor rankings are exported, brand terms stripped out, and the survivors sorted. Then the slow part — opening the result page for anything ambiguous, because a query's intent is decided by what Google is choosing to show, not by how the words read. Two minutes per query, and it changes the fate of the plan. One to three days.
Map the gap to pages, not to keywords. Queries that a single page would satisfy get grouped together, and each group is matched to a page type — service page, comparison, guide, location page — and to an existing page of yours where one already half-covers it. One group is one page. Treating every keyword as its own page is how a site ends up competing with itself. One day.
Look at the link and source gap. Who links to them, and why did those links happen — a piece of original data, a free tool, a story, or simply being older. This is where we are honest about cost, because some gaps are only closable with a year of outreach and should be labelled as such rather than quietly listed next to a title-tag fix. Half a day to a day.
Turn it into a build order. Three phases — closest to money and easiest to rank first, supporting content second, top of funnel last — plus the internal links each phase-one page should receive. Alongside it, the explicit ignore list. Half a day to a day.

The tools, so you know what is behind the numbers rather than a black box — Google Search Console for anything true about your own site, Ahrefs or Semrush for competitor rankings and backlinks, Ahrefs Webmaster Tools and Bing Webmaster Tools where a free route covers it, Keyword Planner and Trends for demand, Screaming Frog to read a competitor's site structure, and the result pages themselves, in an incognito window, by hand. Language models are used to group and reconcile large exports, which they are very good at, and never asked for a volume or a ranking, which they will invent with total confidence.

What moves the price

How many competitors and how many markets. One local rival in one country is a different job from five competitors across three countries and two languages, where every query has to be checked against a different result page and half the intents change on the way.

Whether you have history. A site with three years of Search Console data lets us compare a modelled competitor number against something measured, which sharpens everything. A new domain with no data means the entire analysis is external estimate and manual result-page reading, and it takes longer to reach a conclusion we would defend.

Where the deliverable stops. Findings and a prioritised gap list sit at the lower end. A full page-by-page build order, mapped onto your existing URLs with cannibalisation flagged and internal linking specified, is more work and sits at the upper end.

What usually goes wrong

The wrong competitors get analysed. You name the three firms you lose pitches to, and those three turn out to be just as invisible in search as you are. A gap report built on them is full of terms nobody ranks for, and it reads as reassuring right up until it produces nothing. The competitor that matters here is whoever occupies the result page, which is frequently a directory or a publisher's list article — not a business you would ever recognise as a rival.

Branded queries are counted as gaps. A competitor's largest traffic terms are almost always their own name and their product names. Leave those in the export and the headline gap is that people search for their company more than yours, which is true, unactionable, and not what you paid for. Every serious report separates branded from non-branded before anything is concluded, and the ones that do not are usually the ones with the most impressive-looking chart.

Modelled visibility is quoted as measured fact, especially across September 2025. Third-party traffic and visibility figures for a site you do not own are estimates, and their history has a seam in it. When Google removed the &num=100 search parameter, rank-tracking tools could no longer pull a hundred results in one request, and the phantom impressions that used to accrue to everything ranking below page one disappeared. Charts spanning that date show collapses that were never real and average positions that "improved" for no reason. Reading a competitor's decline off one of those charts, and building a plan around a weakness that does not exist, is the most expensive mistake available in this service.

See where the site stands before you brief anyone.

If the site is already live, start from what it scores rather than from a guess. The audit reports what it checked, what it found, and — the part most tools leave out — what it could not check, and why.

See the website audit Talk about this work

Questions people ask before they commit.

Can you tell me exactly how much traffic my competitor gets?: No, and no honest tool can. Real clicks and impressions come from Search Console, and Search Console only ever shows you a property you own. Every number anyone quotes about a rival's traffic is a third-party index modelling what it thinks happens. We use those estimates for direction and relative size, never as fact, and we label them that way in the report so you do not repeat a modelled number in a board meeting as if it were measured.
Which competitors should I give you?: Give us the businesses you lose deals to, because that tells us what a buyer compares you against. Then expect the list to change. In search your competitor is whoever holds the result page for the queries that matter, and that is often a directory, a marketplace listing, a national publisher's roundup or a Reddit thread rather than a company you have ever pitched against. We reconcile both lists and tell you which of your named rivals are as invisible as you are.
Will this tell me how to overtake them?: It tells you where the gaps are, which of them are reachable and in what order to attack them. What it will not do is promise a position. Rankings sit with Google's algorithm, not with us. We commit to the analysis, the evidence and the build order, and we write those as scope. Overtaking a named competitor is a goal, and anyone selling it as a deliverable is selling something they cannot control.
We are a new site with almost no history. Is this still worth doing?: Yes, but it changes shape. With no Search Console history there is no first-party baseline to compare against, so the whole picture comes from external estimates and from reading result pages by hand. That makes the output less about closing measured gaps and more about deciding what to build first, which is usually exactly what a new site needs anyway. We say up front which half of the normal method is unavailable to you.
Do I need to buy Ahrefs or Semrush for this?: Not necessarily. Ahrefs Webmaster Tools is free on a site you have verified, Bing Webmaster Tools gives away both keyword and backlink data that goes unclaimed because the traffic share looks too small to bother verifying, and Keyword Planner and Trends cover demand. Paid tools genuinely help for competitor backlink depth, and we will say when the job needs one. What we will never do is run your data through a shared group-buy tool account, which is account sharing, against the terms, and puts your competitive research somewhere neither of us controls.
How is this different from keyword research?: Keyword research starts from your customer and asks what they type. A gap analysis starts from the result page and asks who is already being paid for those searches and why. They overlap in the middle and the outputs look similar, but the gap version finds demand you would never have thought to ask for, because someone else is already serving it and you can see the page they are serving it with.

Often taken together.

Start with the brief, not the invoice.

Send what you are trying to achieve. You get a written scope, a price and a date — or an honest “this is not the right service for that”.

Talk about Competitor gap analysis All SEO & Organic Growth services