Link or rewrite: how to tell what a struggling page actually needs

I’ve spent money building links to pages that were never going to rank no matter how many links they got. I’ve also spent hours rewriting pages that were fine and just needed a push. Both mistakes cost time, and on a portfolio of sites, time is the resource you run out of first.

The question “does this page need a link or a rewrite” sounds simple, but most people answer it by gut feeling. They see a page stuck on page two and assume it needs links, because links are what everyone talks about. Or they see a page with low traffic and assume the content is bad, because that’s the easier story to tell yourself. Neither instinct is reliable on its own. You need a diagnostic, not a hunch.

Start with what the page is actually competing against

Before touching links or content, pull up the top five results for your target keyword and read them. Not skim, read. You’re checking three things: what format they use (list, guide, comparison, tool), what the page actually answers, and how deep they go on the subtopics a searcher would expect.

If your page is a 600-word overview and the top five are 2,000-word guides with comparison tables and examples, you have a content gap. No number of links closes that gap, because the page doesn’t satisfy what the searcher came for. Links can push a page that’s already competitive on substance. They can’t manufacture substance that isn’t there.

If your page matches the depth and format of what’s ranking, and it’s still stuck below the top results, that’s a different signal. At that point content parity is met and the gap is more likely about how many other pages point to the ranking pages versus yours, and how established those domains are. That’s a link problem, or at least a problem links are more likely to move.

Check the on-page basics before you decide anything

A lot of pages that look like “needs links” are actually “needs the on-page fixed first.” I check these before making any decision:

Title tag and H1 actually match the query. If someone searches “best budget tripod for travel” and your title says “Tripod reviews,” you’re leaving relevance on the table before a single link gets built.

The page answers the query in the first 100-150 words. Readers and crawlers both use the opening to judge whether the page is worth the rest of the scroll. If your page spends three paragraphs on background before getting to the point, that’s a rewrite issue, not a links issue.

Internal links pointing at the page use relevant anchor text. I’ve found pages on my own sites where I’d linked to them with anchor text like “click here” or the bare URL. That’s a five-minute fix that costs nothing and I check it before I ever consider paying for external links.

If any of these are broken, fix them first. It’s free, it’s fast, and it removes a variable. You don’t want to build links to a page and then go back and change the title tag two weeks later. That just makes it harder to know what actually worked.

The pattern I use to sort pages

After running this check across enough pages, I’ve settled into four buckets.

Good content, few or weak links, ranking somewhere between position 8 and 25. This is the clearest “build links” case. The page satisfies intent, it’s just underexposed relative to what’s ranking above it. This is where link building has the best odds of moving the needle, because you’re not fighting the content itself.

Thin content, few links, ranking anywhere. Rewrite first. Don’t build a single link until the content is at least on par with what’s ranking. I’ve seen pages jump several positions just from expanding thin content and adding the sections competitors have, with zero new links involved.

Good content, decent links, but stuck. This is the frustrating one. Sometimes it’s a format mismatch you missed on the first pass (they want a table, you have prose). Sometimes it’s that the page targets a keyword where the intent has shifted and you’re now competing against a different kind of result than you think. Reread the top results again, specifically for format and intent, not just word count.

Thin content, lots of links. This happens when someone builds links to a page before checking if the content deserved them. If this is one of your own pages, the fix is usually to rewrite it, not to keep adding links, because you’re pouring effort into a page that can’t hold the position even if it gets one.

Why I don’t split the difference

The temptation is to do a little of both at once, a light content refresh and a handful of links, and call it covered. I’ve done this and it makes it hard to learn anything. If the page moves, you don’t know if it was the content change or the links. If it doesn’t move, you don’t know which lever to pull next. On a single page that might not matter. Across a portfolio where you’re making this call dozens of times a month, not knowing which lever works wastes the next round of decisions too.

My rule now: pick one lever, make the change, wait long enough to see a real signal (not a day or two, more like several weeks, since rankings move slowly and noisily), then decide on the next move. If the content fix alone gets the page from position 15 to position 9, links are now worth trying. If it does nothing, the content probably wasn’t the bottleneck and I look at links or reconsider the keyword entirely.

What links actually need to look like

I’m not going to pretend there’s a shortcut here. Relevant links from sites that have their own real audience and their own reason to exist are what move pages that already deserve to rank. Buying volume from networks built purely to pass links, or trading links in bulk with unrelated sites, is the kind of thing that’s cost me clean pages before, either through no effect at all or through the site getting flagged in ways that take far longer to undo than the rewrite would have taken. If a link opportunity only makes sense because it’s cheap and fast, that’s usually the sign to skip it, not the sign you found a deal.

The pages that respond to link building are the ones where I’d feel fine mentioning the link to the site owner honestly, because the content stands on its own. If I wouldn’t want to explain why the link exists, I don’t build it.

What a rewrite actually needs to look like

A rewrite isn’t padding the word count. It’s matching or exceeding the depth of what’s already ranking, in the format the query wants, with the answer up front. Sometimes that means the page triples in length. Sometimes it means cutting a bloated page down to something that actually answers the question instead of stalling for word count of its own. I’ve had rewrites that removed 40% of the content and still ranked better, because what was left matched intent instead of drowning it.

There’s no fixed number of words or sections that guarantees a result, and I’m not going to tell you one, because I don’t know your query or your competition. What I can tell you is that skipping the “read the top five results properly” step is the single most common reason people build links to pages that were never going to hold a ranking, or rewrite pages that didn’t need it.

If you want more breakdowns like this on deciding where to actually spend SEO effort, you can find the rest of what we’ve written at The SEO Desk.