Every client believes they know who their competitors are. They are usually wrong, and the way they are wrong is predictable: they name the businesses they meet at trade events and lose tenders to. Search does not care about any of that. In the SERP, your competitor is whoever occupies the result your customer clicks, and that is frequently an aggregator, a marketplace, a directory or a media outlet that does not compete with the business at all in the real world.
This article covers the competitive and link intelligence side of the Semalt platform: how to identify your real search competitors, how to read a backlink profile without deceiving yourself, and how to convert both into a prospect list that produces links rather than a spreadsheet nobody opens.
Key takeaways
- A search competitor is any domain ranking on your queries — not the businesses you meet at trade events.
- Separate direct competitors from aggregators and media sites. Treating a marketplace as a rival burns budgets.
- In a backlink profile, shape beats size: growth curve, link types, anchor distribution and topical relevance.
- Track share of visibility, not just your own curve — rising visibility with falling share means you are growing slower than the market.
Identifying your actual search competitors
The correct definition is mechanical: a search competitor is any domain that ranks on the queries you want. Not any domain in your industry. The distinction sounds pedantic until you run the analysis and discover that four of the top ten results for your primary commercial term are comparison sites.
Semalt builds the competitor set from live SERP overlap: it takes your tracked keyword set, records who ranks on each query, and ranks domains by how much of your keyword universe they cover. The resulting list nearly always contains two or three names the client has never mentioned.
What you do with those names depends on what kind of competitor they are, and there are three kinds worth separating:
| Type | How to recognise it | Correct response |
|---|---|---|
| Direct commercial competitor | Sells what you sell, similar business model | Beat them — their content and link strategy is directly instructive |
| Aggregator or marketplace | Ranks on almost every head term in the category | Get listed on them, then attack the long tail they cover badly |
| Content or media site | Ranks on informational queries adjacent to yours | Not a competitor — a link opportunity and partnership target |
Misclassifying the second group as the first is the most expensive strategic mistake in this whole area. Spending six months trying to outrank a national marketplace on a head term burns a budget that would have produced results on two hundred long-tail queries.
Working parameters from the routine described in this article.
Gap analysis: the highest-value screen in the platform
The gap report answers one question: which queries do the competitors rank for that you do not? Sorted by estimated value, that list is effectively your content roadmap, and it is derived from evidence rather than brainstorming.
Reading it well requires filtering in a specific order. Start by removing queries where the ranking competitor is a marketplace or the intent is clearly navigational to their brand — those are not winnable. Then group what remains by topic rather than treating each query as a separate page, because twenty variations of the same question are one article, not twenty.
Then apply the only prioritisation that matters: how close is the query to a transaction, and how realistic is the position given the competition on that specific SERP. A high-volume informational query with three national media outlets ranking is a worse investment than a modest commercial query where the current results are thin and old.
A pattern worth looking for. Queries where a competitor ranks on page one with an obviously weak page — thin content, published years ago, no depth. Those are the cheapest wins in any gap report, and they are visible in seconds once you start clicking through to the ranking URL instead of only reading the table.
Reading a backlink profile honestly
Backlink analysis attracts more self-deception than any other part of SEO, mostly because the headline numbers are so easy to misread. Total backlinks is close to meaningless — one directory can generate ten thousand of them. Referring domains is better. What actually matters is the shape of the profile.
Four things to look at, in order:
Referring domain growth over time. A steady curve suggests a business that people genuinely reference. A vertical spike followed by a flat line usually means a purchased batch, and it tells you the domain's authority is more fragile than its numbers suggest.
Link type distribution. Editorial links inside article text behave differently from footer links, directory listings and profile links. A profile that is 80% directories has a large number that means very little.
Anchor text distribution. Natural profiles are dominated by brand names, URLs and generic phrases. A profile where 40% of anchors are exact-match commercial keywords was built deliberately, and not carefully.
Topical relevance of the linking domains. Fifty links from sites in the same field beat five hundred from unrelated ones. This is the dimension that raw authority metrics miss entirely.
Run this analysis on your competitors before running it on yourself. It tells you what it actually took to earn their position, which is usually more sobering and more useful than any audit of your own site.
The most expensive mistake in this area is treating an aggregator as a direct competitor. Spending six months trying to outrank a national marketplace on a head term burns a budget that would have produced results on two hundred long-tail queries.
Turning competitor links into a prospect list
The classic workflow still works, and the platform mainly makes it faster. Take three to five direct competitors, pull their referring domains, and find the domains that link to several of them but not to you. A site that has linked to three of your competitors demonstrably publishes links in your field and has no relationship with you — that is the definition of a warm prospect.
There is a full walkthrough in The New Semalt: One Platform That Runs Your Entire SEO Workflow.
Filter that intersection list on four criteria before anyone writes an email:
| Criterion | What to check | Reject if |
|---|---|---|
| Relevance | Does the site actually cover your topic? | Generic site publishing anything for a fee |
| Real traffic | Does it have search visibility of its own? | High authority score, no visible traffic |
| Link pattern | Where do its outbound links sit? | Every link in a "sponsored posts" ghetto |
| Editorial signal | Is there a real author and real content? | Automated aggregation, no bylines |
The second row deserves emphasis. Authority metrics can be manipulated, and there is an entire economy built on domains that score well and are read by nobody. Search visibility is much harder to fake. If a domain has a strong score and no organic traffic, something is wrong with it, and a link from it is at best worthless.
What actually earns links in a small market
Outreach advice written for the English-speaking web transfers badly to markets like Hungary, where the pool of relevant publishers is small enough that everyone notices templated emails. Three approaches work reliably here.
Original local data. Publishing genuine numbers about your market — pricing surveys, demand analysis, industry statistics nobody else compiles — earns links because journalists and bloggers need citable sources and there are very few in smaller languages. This is the single most reliable link asset for a Hungarian business.
Practical tools and calculators. A working calculator that solves a real problem gets referenced for years. It also survives content refresh cycles that kill ordinary articles.
Genuine expert commentary. Local journalists need quotable specialists and struggle to find ones who answer quickly. Being reliably available to a handful of relevant publications produces better links than any outreach campaign, and the relationship compounds.
What does not work, reliably: mass guest posting on sites that exist to sell guest posts, comment links, and any arrangement where the price list arrives before the editorial conversation.
Link sources worth pursuing
- Original local data nobody else compiles
- Working tools and calculators that solve a real problem
- Expert commentary for journalists who need quotable specialists
- Industry bodies, associations and partners you already work with
Sources that reliably waste budget
- Sites that exist to sell guest posts
- High authority score with no organic traffic
- Every outbound link parked in a "sponsored" section
- Any arrangement where the price list arrives before the editorial conversation
Monitoring: lost links and competitor movement
The part of link work everyone skips is watching what disappears. Links are lost constantly — sites redesign, articles get pruned, pages 404, publishers switch CMS and drop old content. A profile that looks flat may be gaining and losing at the same rate, which is a very different situation from stability.
The platform reports new and lost referring domains per period, and the lost list is worth ten minutes a month. Many losses are recoverable with one polite email, particularly when the cause was a URL change on their side rather than a decision to remove the link. Recovering an existing link is dramatically cheaper than earning a new one.
See also: The Semalt Website Audit.
The same monitoring applied to competitors answers a strategic question: what are they acquiring, and from where? A competitor that suddenly gains twenty referring domains in a month is running a campaign, and the pattern of those domains tells you what kind. If they are all from one publisher network, you have learned something about the quality of their growth. If they are scattered across genuine industry publications, you have learned that they have a real PR function and you should plan accordingly.
The disavow question
Clients ask about toxic links more than the topic deserves. The current reality is that search engines discount most low-quality links rather than penalising for them, and aggressive disavowing has caused more harm than the links themselves in a fair number of cases we have seen.
The defensible position: disavow when there is evidence of a deliberate negative campaign or when the profile contains a large volume of links from a previous provider you know were purchased in bulk. Otherwise, monitor and leave it alone. Spend the effort on earning links rather than curating the ones you already have, because the second activity has a much lower ceiling.
A monthly competitive routine
Fifteen minutes on competitor visibility: did anyone in the tracked set move sharply, up or down? A sharp competitor drop is as informative as a rise — it often reveals a penalty, a migration gone wrong, or a strategic retreat you can exploit.
Thirty minutes on the gap report: pull the top ten newly appearing opportunities and add the two or three worth writing to the content queue. Do not add ten. A queue nobody clears is a queue nobody trusts.
Twenty minutes on links: review new and lost referring domains, send one recovery email if there is one to send, and add any domain that has now linked to two or more competitors to the prospect list.
That is roughly an hour a month per client, and it is the hour that keeps the strategy anchored to what is actually happening in the market rather than what was true when the campaign was planned.
Measuring competitive progress without lying to yourself
There is a reporting trap specific to competitive work. You improve, the client is pleased, and six months later the business has lost market share anyway — because two competitors improved faster. Absolute metrics cannot show that. Relative ones can.
We cover this in detail in Semalt Analytics and Rank Tracking.
The measure worth tracking is share of visibility across the tracked keyword set: your visibility divided by the total visibility of you plus your tracked competitors. It is a blunt instrument, and it depends entirely on choosing a stable and honest competitor set, but it answers the question absolute numbers dodge. Rising visibility with falling share means you are growing slower than the market. That is a strategic warning worth surfacing eighteen months before it shows up in revenue.
Two rules keep this metric honest. First, freeze the competitor set for the duration of the campaign, exactly as you freeze the keyword set — swapping in a weaker competitor to improve the ratio is the same self-deception as adding easy keywords. Second, report the underlying curves alongside the ratio, because a share increase caused by a competitor collapsing is not the same achievement as one caused by your own growth, and pretending otherwise will be embarrassing when they recover.
Used properly, this single number reframes the client conversation from "are we doing well" to "are we winning", which are different questions with different answers surprisingly often.
Verdict
Competitive and link analysis is the area where tooling differences matter least and process differences matter most. The data available to any serious platform is broadly similar, because it comes from crawling the same web and measuring the same SERPs. What separates useful work from busywork is whether the data feeds a routine that produces decisions.
Semalt's contribution here is that the competitor set, the keyword set and the link data live inside the same project, so gap analysis and link prospecting are two views of one dataset rather than two exports that someone has to reconcile. That saves the reconciliation work, which is exactly the work that gets skipped when the month gets busy.
If you want a fast, concrete test of whether your assumed competitors are your real ones, that takes about ten minutes. Open the dashboard, add your domain and your primary keywords, and look at the competitor list the SERP data produces. If it matches the list the client gave you, you have confirmed a strategy. If it does not — and it usually does not — you have found the reason the last campaign underperformed.
Frequently asked questions
How do I find my real search competitors?
Build the set from live SERP overlap rather than from the client's assumptions: take your tracked keyword set, record who ranks on each query, and rank domains by how much of your keyword universe they cover. The resulting list nearly always contains two or three names the client has never mentioned — and frequently an aggregator or comparison site that dominates the head terms.
How many backlinks do I need?
It is the wrong question. Total backlinks is close to meaningless — one directory can generate ten thousand of them. What matters is the shape of the profile: a steady growth curve rather than a spike, editorial links rather than footers and directories, brand-dominated anchor text, and topical relevance in the linking domains. Fifty links from sites in your field beat five hundred from unrelated ones.
Should I disavow low-quality links?
Usually not. Search engines discount most low-quality links rather than penalising for them, and aggressive disavowing has caused more harm than the links themselves in a fair number of cases. Disavow when there is evidence of a deliberate negative campaign, or when the profile contains bulk purchases from a previous provider. Otherwise monitor and spend the effort on earning links instead.
Why bother tracking links we have lost?
Because a flat-looking profile may be gaining and losing at the same rate, which is very different from stability. Many losses are recoverable with one polite email, particularly when the cause was a URL change on the publisher's side rather than a decision to remove the link — and recovering an existing link is dramatically cheaper than earning a new one.
Open your Semalt dashboard
Audits, rank tracking, competitor data and reporting in one place. Sign in and you will be looking at real numbers for your own domain within minutes.
Sign in to SemaltOr browse the service overview at semalt.com.