Your first backlinks: a 50-link plan for a brand-new site

Your first backlinks are the ones most new sites waste. The usual playbook is to pitch every blog in the niche in week one, collect a few replies, and then wonder why the domain doesn’t move. The effort isn’t the problem. The sequence is. A brand-new site with a thin homepage and no indexed pages can’t make good use of a link: the visitor lands on nothing, the editor who placed it finds nothing worth citing, and Google has little reason to treat the domain as a real destination. This is a realistic plan for the first 50 backlinks on a new domain. It covers what to fix before you ask for anything, which placements are worth the time, what to skip, and how to tell when you’ve moved from building links to attracting them.

Why your first backlinks do a different job

Every backlink does two things. It passes a signal, and it creates a path. On an established site the signal dominates. Google already knows the domain, has crawled it for years, and has a rough picture of what it covers, so a new link is one more vote on a large existing record. A new domain has no record. Its first links have to establish that the site exists, that it is about something specific, and that a third party considers it worth pointing at.

The path matters more than most people expect. Google finds new URLs through sitemaps and by following links from pages it has already crawled. A new domain can be crawled perfectly well with a submitted sitemap and zero inbound links, but discovery tends to be slower and less predictable. A link from a crawled, indexed page on a relevant site often gets your new URLs found sooner. “Sooner” is the honest word. I won’t give you a number of days, because it depends on how often the linking site is crawled.

That’s why the 50 is not really a volume target. It’s a planning unit: roughly the number of credible, relevant placements a careful builder can land over a few months from several source types. Treat it as a checkpoint, not a threshold. A site in a narrow niche with little competition can start ranking for specific terms on a handful of good links. A site chasing a crowded commercial term may need far more, and 50 weak links won’t close that gap. What decides whether your 50 work is where the links come from and what they point to. For the broader argument, see relevance over volume.

Fix the site before you ask for a single link

Most new sites have the same problem. They ask for links to a homepage that describes the product in marketing language and links to nothing else. An editor who clicks through sees a sales page, and editors don’t cite sales pages. Before you send a single email, check five things.

1. Indexing is clean. Your key pages return a 200 status, carry no leftover noindex tag from staging, and appear in a sitemap you’ve submitted in Search Console. A staging noindex that survives launch is common. It means the pages your links point to can’t rank, no matter how many links arrive.
2. You have three to five pages worth linking to. Each one should answer a specific question a buyer or practitioner actually asks, with real substance: a method, numbers, a worked example. Three thin pages look like a brochure. Five substantive ones look like a site that plans to stay.
3. The site says who runs it. An About page with named people, a real company description, a contact route and a privacy policy. Editors check this, and a site that could belong to anyone is hard to justify linking to.
4. The name and description match everywhere. Same brand name, same one-line description, same URL form. This is what lets listings, profiles and mentions be attributed to one entity rather than three half-related ones.
5. Each page has one address. One canonical version of the domain, https everywhere, and no redirect chains left over from a placeholder domain. Chains slow crawling and make it harder to tell which URL an editor meant to link.

The last gap is content, and it’s the real one. A new site rarely has anything an editor would cite. Your homepage can’t carry link building on its own. What gets cited is a resource: a calculator for a number your niche argues about, a template with a worked example, a glossary entry more precise than the generic one, or a teardown with real figures from your own product. You don’t need five of these before you start. You need one, and it’s the centre of phase two below.

The order to build your first backlinks

Here is the sequence. Treat the timings as rough. The stages overlap in practice, and the order matters more than the dates.

Phase one: foundation profiles

Start with the listings any legitimate business should have anyway. These aren’t link building in the classic sense. Their job is to confirm the entity: that the business exists, what it’s called and what it does. Examples that fit most products:

• A Google Business Profile, if you serve a local area or a physical location.
• A company page on LinkedIn and a profile on Crunchbase.
• A GitHub organization, if you’re developer-facing, with a README that links to your site.
• Your product on the marketplaces for any platform you integrate with.

Most of these are nofollow, and that’s fine. Check each one anyway, because the attribute can change and you should know what you have. At this stage their value is consistency and referral traffic, not ranking equity. Build them in week one because they cost little and anchor the entity. Then keep them out of the target count, so the number doesn’t inflate and you don’t mistake a profile for progress.

Phase two: one asset that other sites can cite

This is the engine for everything after. People link to resources when they need something specific and can’t find a good version elsewhere. A new site with one sharp, honest resource gives every later outreach email something concrete to point at.

Good first assets are narrow and useful. Think of a calculator that shows its formula, a template that works in a real workflow, a small public dataset from your own product with the method explained, or a comparison of options with the criteria stated and applied honestly. Avoid the generic “ultimate guide.” It’s the most common asset on the web, which makes it the least likely to be cited.

An asset does not earn links just by existing. It needs introducing. Find 20 to 40 pages that already link to similar resources: resource pages, tool roundups, niche guides that cite a calculator or template on the same subject. Then write to the people who maintain them with a specific reason. Your resource covers a gap their page has, and the number on it is one they can’t get elsewhere. The outreach email piece covers the wording. The point here is targeting. Pitch the 20 pages that already cite the thing you’ve built, not the 200 blogs that merely share your topic.

Phase three: partner, directory and editorial placements

With the foundation and an asset in place, you can fill the rest of the 50. The mix below is a plan, not a quota. Your niche will push it one way or another.

Two points on the table. Directories are worth the time only when a person reviews them and the category fits. A directory that accepts anything for a fee is a link list with a logo on it. Partner and integration pages are often the best early source for a product, because the partner has a reason to link to you that isn’t about SEO. You work together, and they want their customers to find the integration.

Five ways a new site wastes its first links

Most new-site link failures aren’t dramatic. They’re small decisions that look productive in a spreadsheet and add nothing to the index.

• Day-one velocity. Going from zero to 40 links in a week from a domain with no history looks manufactured, because it is. A steady trickle reads as natural. Spread placements over months and let the pace follow your asset, not a deadline.
• Exact-match anchors. Your first anchors should be your brand name, your URL or a plain descriptive phrase. An anchor like “best project management software” on a six-week-old domain reads as engineered, and it’s the easiest pattern for a filter to notice.
• Partner footers and sitewide blogroll links. A link in the footer of every page on a partner’s site adds little, and sitewide placements look arranged. Ask for an in-content link next to the sentence it supports. If the partner won’t do that, take the placement that isn’t there.
• Reciprocal swaps. “You link to me and I’ll link to you” groups are easy to detect and add little. A reciprocal link that is relevant and editorial is simply a normal link. One that exists only because of the swap is not worth the page it sits on.
• Chasing authority scores. A DR 20 blog in your exact niche with a real readership beats a high-authority general news site with nothing to do with your product. Third-party authority scores estimate a link profile. They don’t measure how much a given link helps you. Relevance and placement decide far more, which is the whole argument of relevance over volume.

There’s a sixth trap that catches careful people: links on pages that never get indexed. Some placements look fine and count for nothing. After a few weeks, search Google for the exact URL of each placement. If it doesn’t come back, the link probably isn’t doing anything. The fixes are covered in our piece on backlink indexing.

How to count progress without fooling yourself

Count referring domains, not links. Fifty links from fifty separate sites is a very different profile from fifty links from three sites, and only the first looks like a real audience. Keep a simple log with the date, the linking domain, the page, the anchor, whether the link is dofollow, whether the page is indexed, and how you got it: outreach, directory, partner or unsolicited. Keep foundation profiles on a separate tab.

The log shows the mix, and the mix matters more than the total. A log that is 90 percent directories after three months tells you the plan drifted. A log where half the domains came from outreach around your asset tells you the asset is working.

Use Search Console as the verdict. Its Links report shows which sites Google has found linking to you, which is the most reliable list of what has arrived. Third-party tools will show you links Google hasn’t picked up yet, which is useful for the log. Just don’t count a link as real until the search engine agrees it exists.

How to tell you’ve shifted to earned links

The goal of the first 50 is to make the 51st unnecessary. Earned links come from the same place as your first ones, only later: a page good enough that someone chooses to cite it without being asked. Outreach buys the first set of people a look. Once a resource is genuinely useful, it does part of the work itself.

Watch for four signals.

• Unsolicited referring domains. In your log, the share of new domains you didn’t pitch is the cleanest measure. Early on it’s close to zero. When most of a quarter’s new domains came to you, you’ve made the shift.
• Unlinked mentions. People write about your product or cite your resource without linking to it. Search for your brand name and your asset’s title every month. A short, polite request to add the link converts a useful share of these, and they’re cheap to earn, because the writer has already decided to mention you.
• Inbound requests. A journalist, newsletter writer or site owner asks for a quote, a number from your dataset or permission to reuse a chart. These arrive once the resource is known, and they’re the closest thing to a link you can’t take credit for planning.
• Citations you didn’t chase. Your calculator turns up in a forum thread, a course reading list or a competitor’s footnote.

One practical caveat: resources decay. A calculator with outdated rates or a template built for last year’s tool stops getting cited. Plan one new linkable asset per quarter, or refresh an existing one with a visible, dated update. Earned links follow relevance, and relevance has to be maintained.

What to expect, and what not to promise

In the first month you’ll mostly be building the foundation and the asset, with few links to show. Placements tend to accumulate over several months, and the pace will be uneven. Links alone rarely lift a new site into rankings. They help search engines find and trust the domain, and they bring referral traffic. Rankings still depend on whether your content matches what people search for, whether the site is technically sound, and how crowded the terms are. Be wary of anyone who sells a ranking as a link count.

The short version: fix the site, build one thing worth citing, place links in the order a real audience would find them, and count domains rather than totals. The 50th link should be one you didn’t have to ask for.

This is the part of the plan that’s tedious to run by hand. Finding relevant placements, placing them, checking that each one is indexed and dofollow, and tracking results week by week is repetitive work, and it’s where a new site’s momentum tends to stall. OptiLink finds relevant placements across millions of blogs, hand-places them, gets every one indexed, and tracks it live, so your time can go into the resource that earns the links. See how it works for your site.

Keep reading

Playbook

Anchor text optimization without tripping a filter

A hundred random links can do less than five relevant ones. Here's what makes a backlink pass real ranking value in 2026 — and what makes it worthless or worse.

Sofia Meyer September 24, 20262 min read

Playbook

Broken link building, step by step: find, fix, and pitch

A hundred random links can do less than five relevant ones. Here's what makes a backlink pass real ranking value in 2026 — and what makes it worthless or worse.

Sofia Meyer October 9, 20262 min read

Playbook

Competitor backlink analysis that finds real opportunities

A hundred random links can do less than five relevant ones. Here's what makes a backlink pass real ranking value in 2026 — and what makes it worthless or worse.

Sofia Meyer September 26, 20262 min read

Explainer

Dofollow vs nofollow in 2026: what still passes value

The link attributes changed, Google's treatment changed, and a lot of advice is out of date. Here's what dofollow, nofollow, sponsored and ugc actually do now.

The OptiLink Team October 10, 20262 min read