Keyword research for link building: a practical guide

Link building gets blamed for a lot of things it did not do. When a campaign produces links but no movement, the problem is usually not the links; it is the pages they point to. If you build links to a page that cannot rank, or that targets a keyword nobody searches, the links do their job and nothing happens. This guide is about the step that comes before outreach: deciding exactly which pages deserve links, and why.

Keyword research for link building is different from keyword research for content. For content, you are looking for topics worth writing about. For links, you are looking for pages worth investing in, which means pages with commercial value, a realistic chance of ranking, and a gap that links can close. Get this step right and everything downstream gets easier.

If you have read our guide to how many backlinks you need, you already know that the answer depends on your competitors. This guide is the companion piece: how to work out which pages those links should point to in the first place.

Why link campaigns fail at the keyword stage

Most failed link campaigns fail for one of three reasons, and all three are keyword problems in disguise. The first is targeting a page that cannot rank no matter how many links it gets, because the competition is too strong or the page is too thin. The second is targeting a keyword with no commercial value, so even a page-one ranking produces no revenue. The third is spreading links across too many pages, so no single page accumulates enough authority to move.

Each of these is avoidable with a little research up front. The cost of skipping it is not just wasted links; it is wasted months, because link building compounds slowly and a misdirected campaign only reveals itself after a quarter of effort. The research takes a few hours and saves a few months, which is the best trade in SEO.

There is also a strategic reason to start with keywords. Links are an investment, and like any investment they should be allocated where the return is highest. A page that is already ranking in positions eleven to twenty is a far better target than a page stuck on page five, because the links needed to push it onto page one are far fewer. Keyword research is how you find those near-miss pages.

The mistake beginners make is to treat link building as a volume game: build as many links as possible to the site and hope. That approach worked, roughly, a decade ago. Today it produces a scattered profile that helps nothing in particular. The modern approach is concentrated: fewer links, aimed precisely at pages that can convert them into rankings and revenue.

Start with search intent, not volume

The first filter is intent. A keyword with 10,000 monthly searches is worthless if the people searching it are not going to buy, and a keyword with 200 searches is gold if they are ready to convert. For link building, you want commercial and transactional intent, because those are the pages where rankings translate into money.

Sort your keyword list into four buckets: informational, navigational, commercial and transactional. Informational keywords are for blog content and rarely need links. Navigational keywords are for your brand and need none. Commercial and transactional keywords are where link investment pays, because the searcher is evaluating or ready to act.

Within the commercial bucket, look for keywords that signal evaluation: "best", "vs", "review", "alternatives", "pricing". These are the terms where a well-linked page can capture high-value traffic. A page ranking for "best project management software" is worth more than one ranking for "what is project management", even though the second may have more searches.

Intent also tells you what kind of page to build links to. A transactional keyword needs a product or service page. A commercial keyword might need a comparison or category page. Matching the page type to the intent is part of the research, and getting it wrong means the links point at a page that does not satisfy the searcher.

Mapping keywords to pages

Once you have your commercial keywords, map each one to a specific page. The rule is one primary keyword per page, with two to four closely related secondary keywords. If two pages target the same primary keyword, you have cannibalisation, and the fix is to merge them or differentiate them.

Build a simple table: page URL, primary keyword, secondary keywords, current ranking, and search volume. This table becomes your link-building plan. Every link you build should point to a row in this table, and every row should have a reason to exist.

For each page, note whether it already ranks and where. Pages ranking in positions four to twenty are your best targets, because they are close enough that a handful of good links can push them up. Pages ranking below position fifty usually need content work first, not links. Pages already in the top three need maintenance, not investment.

This mapping also reveals gaps. If you have a commercial keyword with no page targeting it, that is a content task before it is a link task. If you have a page with no keyword, it is probably not worth linking to. The table makes both problems visible, which is the point.

Reading your competitors' link profiles

Your competitors have already done the research, whether they know it or not. Their link profiles tell you which pages they consider worth investing in, and which keywords they are trying to win. Reading those profiles is one of the fastest ways to prioritise your own targets.

Take the top three results for each of your target keywords and look at which pages are ranking. Then look at which of those pages have the most links. The pattern tells you what it takes to compete: if the top results all have hundreds of referring domains to their category pages, you know the bar. If they rank with few links, you know the opportunity.

Look, too, at the anchors competitors use. If a competitor ranks for a commercial term with mostly branded anchors, their brand is doing the work and you may struggle to match it with links alone. If they rank with a spread of partial-match anchors, there is room for a well-linked page to compete.

Finally, look for the pages competitors have neglected. Every niche has commercial keywords where the top results are weak, thin, or outdated. Those are the easiest wins, because you can outrank them with a better page and a modest link investment. Finding them is the whole game.

Prioritising which pages to build links to

You cannot build links to everything at once, so prioritise. Score each page on three factors: commercial value, current ranking, and competition. High value, close to page one, and beatable competition is your first tier. Work down from there.

A simple scoring model works well. Give each page a score out of five for value, a score out of five for how close it is to ranking, and a score out of five for how beatable the competition is. Add them up and sort. The top of the list is where your first month of links goes.

Resist the temptation to spread links evenly. Concentrated investment in a few pages produces visible movement, which builds confidence and informs the next round. Spread investment produces nothing visible, which makes it impossible to learn what works.

Revisit the priority list every quarter. As pages move up, their priority falls, and new near-miss pages rise. The list is a living document, not a one-off exercise, and keeping it current is what keeps your link budget pointed at the best opportunities.

Choosing anchors from your keyword set

Your keyword research also informs your anchor text. The primary keyword is the obvious exact-match anchor, but exact-match anchors should be rare. Most of your anchors should be branded, generic, or partial-match, drawn from your secondary keywords and natural phrasing.

Use the keyword set to generate anchor variety. For a page targeting "enterprise CRM software", natural anchors include the brand name, "this CRM guide", "enterprise CRM options", "their CRM comparison", and the URL. A spread like that looks organic; a page where every anchor is the exact keyword looks bought.

Keep a running list of anchors used per page, so you can see the distribution and avoid over-optimising. If one anchor dominates, vary it. Our guide to anchor text mistakes covers the ratios in detail, and it is worth reading before you build your first link.

The keyword set is a source of anchors, not a script. The best anchors are the ones a real writer would use, which usually means a mix of the brand, a descriptive phrase, and the occasional keyword. Let the content dictate the anchor, and use the keyword list to check you are covering the range.

Tracking whether it worked

Finally, track the result. For each target page, record its ranking for the primary keyword before you build links, and check it monthly. If a page moves after links are built, the targeting worked. If it does not, the problem is either the links or the page, and you need to diagnose which.

Track referring domains per target page, not just site-wide. A site-wide count hides the fact that all your links point at the homepage. Per-page tracking shows you where the authority is actually going, which is what determines rankings.

Give it time. Links take weeks to be indexed and months to compound, so judge a campaign over a quarter. If a page has not moved after three months of relevant links, revisit the keyword research: the page may be targeting a term it cannot win, or the competition may be stronger than you estimated.

Do this research once, properly, and link building stops being a guess. You will know which pages to target, why, and what success looks like. If you would rather have it done for you, our custom campaigns start with exactly this mapping exercise, and our packages apply it to every order.

Common keyword research mistakes

The mistakes in link-building keyword research are predictable, and avoiding them saves months. The first is chasing volume over intent. A high-volume informational keyword looks attractive in a spreadsheet, but links pointed at a page targeting it rarely produce revenue, because the searcher is not buying.

The second is targeting keywords you cannot win. If the top three results are national brands with thousands of referring domains, a modest link budget will not close the gap. Be honest about the competition, and pick terms where a well-linked page can realistically reach page one.

The third is spreading links too thin. Building one link to each of twenty pages produces no movement anywhere. Concentrate on a handful of near-miss pages, move them, then move on. Concentration is what makes link building visible.

The fourth is ignoring the page itself. Links amplify a page that already deserves to rank; they cannot rescue a thin or poorly structured one. If a target page is weak, fix the page before you build links to it, or the links will be wasted.

The fifth is never revisiting the plan. Keywords, competitors and rankings all change, and a map built once and never updated drifts out of date. Review it quarterly, retire targets that have been won, and add new opportunities as they appear.

Avoid these five and your link budget will go where it can actually do something. The research is not glamorous, but it is the difference between a campaign that compounds and one that quietly produces nothing.

Frequently asked questions

Do I need paid tools for keyword research?

Not to start. Google's autocomplete, related searches and free keyword tools will get you a working list. Paid tools help with volume estimates and competitor data, but the intent and mapping work is free.

How often should I revisit my keyword map?

Quarterly is enough for most sites. Rankings shift, competitors change, and new opportunities appear. A quarterly review keeps the map current without turning it into a full-time job.

What if my target page has no content yet?

Build the content first. Links to a thin page waste authority. Create a page that genuinely serves the keyword, then build links to it once it is worth ranking.