Backlinks are still one of the strongest signals in search, and they are the hardest thing on this list to fake convincingly. That combination is why link building attracts so much bad advice: the incentive to shortcut it is enormous, and the shortcuts are easy to sell.
This guide is about earning links you would still want if search engines vanished tomorrow. That is not a moral position, it is a practical one. A link that a real person chose to place is the only kind that keeps working after the next algorithm update.
What makes a backlink worth having
Not all links carry weight, and the difference is not subtle. Before you spend a week chasing one, check it against four things.
- Relevance: the linking page is about something adjacent to your topic, not a general directory of everything
- Editorial choice: a human decided to link, rather than a form letting you add yourself
- Placement: inside the body copy beats a footer, a sidebar, or an author bio
- Traffic: the page has readers, which means the link can send you people as well as signal
The tactics that still work
Almost everything that works reduces to the same trade: you give someone something genuinely useful, and the link is how they credit it. The variations are just different forms of the useful thing.
Original data is the strongest. Publish a number nobody else has, from your own product, your own survey, or your own analysis of a public dataset, and writers who need that number will cite you. You only have to do the work once and the citations accumulate for years.
Being a source is the fastest. Journalists and bloggers need quotes on deadline. Answering three requests a week with a specific, quotable opinion in your actual field will out-earn a month of outreach emails, because you are solving their problem rather than asking them to solve yours.
Free tools work when they are genuinely useful and free without a signup wall. A calculator, a checker, a generator: something a writer can point a reader at without apology. This is expensive to build and cheap to maintain, which is the right shape for a link asset.
Genuine partnerships work at small scale. Guest posts on sites you would read anyway, co-written pieces, podcast appearances, community answers where you are actually helpful. Slow, but every one is defensible.
The tactics that stopped working
Some of these still get sold. None of them are worth your budget.
- Buying links from a marketplace: detectable, and the risk sits entirely with you
- Mass guest posting on sites that exist only to host guest posts
- Link exchanges beyond the occasional natural one between two real sites
- Comment and forum signature links, which have been discounted for over a decade
- Private blog networks, which are the same purchase with more steps and more risk
Broken link building, done honestly
One classic tactic still earns its keep. Find a page in your field that links to something now dead, write the replacement, and tell the owner both facts in the same email: their link is broken, and here is a live page that covers it.
It works because you are reporting a real defect on their site and handing them the fix in one message. It stops working the moment your replacement is not actually as good as the thing that died, which is the part most people skip.
How to write outreach that gets answered
Most outreach fails on the first line. The recipient can tell within a few words whether a person read their work or a tool merged their domain into a template.
Keep it short enough to read on a phone. Name the specific page and what in it prompted you to write. Say plainly what you are proposing. Make the ask small and easy to refuse, because a fast no is better for both of you than silence.
Send fewer, better emails. Twenty considered messages will beat five hundred merged ones, and they will not burn your domain's reputation in the process.
How many links do you actually need
Fewer than the tools imply. Difficulty scores are computed from the link profiles of pages currently ranking, which flattens a lot of important detail: age, brand, internal linking and topical depth all contribute, and none of them show up in that number.
A practical read is to look at the pages ranking in positions five to ten rather than one to three. Those are the ones you are realistically displacing. Count the referring domains pointing at those specific pages, not at the whole site, and you will usually find the number is smaller and more achievable than the headline difficulty suggested.
Measuring link building without fooling yourself
Count referring domains, not links. A hundred links from one site is one relationship. Ten links from ten sites is ten.
Watch the pages that gained links and check whether their rankings moved for the terms you care about. If links arrive and nothing moves, links were not the constraint, and continuing to buy them will not change that.
Search Console shows which sites link to you and which of your pages they point at. It is free, it is Google's own view rather than a third party crawl, and it is the honest baseline to measure any tool against.
A realistic first ninety days
If you are starting from nothing, this is a sane sequence.
- Weeks 1 to 2: publish one piece of original data or one genuinely free tool
- Weeks 3 to 6: answer source requests consistently, three a week, in your real area of expertise
- Weeks 5 to 8: find ten broken links in your field and write the two best replacements
- Weeks 7 to 12: two guest pieces on sites you would read anyway, written to be worth reading
- Throughout: log every referring domain and which page it points at, so you can tell what worked
The part nobody sells you
The single most reliable way to earn links is to publish something worth linking to and then tell the right people it exists. That is unglamorous and it does not scale on a spreadsheet, which is exactly why it is still available.
If you are choosing between another round of outreach on a thin page and one week spent making the page genuinely the best answer to its question, spend the week on the page. Outreach on a page that deserves the link is a different activity from outreach on one that does not.
Common questions
- How many backlinks do I need to rank?
- Fewer than difficulty scores imply. Count the referring domains pointing at the specific pages ranking fifth to tenth, not at their whole site, because those are the pages you are realistically displacing. That number is usually far smaller than the headline difficulty suggested.
- Is buying backlinks safe if the site looks legitimate?
- No. Paid links violate Google's guidelines regardless of how the selling site presents itself, and the risk sits entirely with the buyer. The link also stops working the moment the arrangement ends, which makes it a rental rather than an asset.
- How long does link building take to affect rankings?
- Expect weeks to months between earning a link and seeing movement, and longer on a new domain. If links arrive and nothing moves after several months, links were not the constraint and buying more will not change that.
Keep reading