There is no safe number of backlinks per month, because Google does not police speed. Its spam policies define link spam by intent and source: links bought or sold for ranking purposes, excessive link exchanges, links created by automated programs or services. None of those rules mentions a monthly cap. Ten links from a private blog network (a set of sites built only to sell links) in a month is a problem. Two hundred links from newspapers covering a study you published is not.
What does matter is the rate of growth relative to your existing profile, and who is doing the linking. Ahrefs’ 2018 study of 200,000 ranking pages found that pages ranking first on Google gain followed links (links that pass ranking credit) from new referring domains at roughly 5 to 14.5 percent per month. A site with 2,000 referring domains that adds 150 a month is growing 7.5 percent, squarely in that range. A site with 200 referring domains adding the same 150 is growing 75 percent, fine if those links are real coverage and a red flag if they are all paid posts with the same anchor text. The supply of sites worth being on will cap you long before Google does.
Set your monthly link target by growth rate and site quality, not by a number someone else called safe: the pages that already rank first tend to add 5 to 15 percent new referring domains a month, and every one of yours should pass a human review.
What does Google actually say about link velocity?
Nothing, directly. The spam policies describe link spam as links intended to manipulate rankings, and every example is about how a link came to exist: money changed hands, a reciprocal deal was struck, a tool generated it, a contract required it. There is no clause about how many arrived in a given month.
Enforcement follows the same logic. Google’s webspam report for 2022 describes SpamBrain, its machine-learning spam system, being trained to detect sites that build spammy links and sites that exist to pass them, catching 50 times more link spam sites than the previous link spam update. The target is the network, not the calendar.
Real companies have real spikes. A funding round, a product launch, a data study picked up by trade press, a Reddit thread that takes off: each can bring more links in a week than the previous six months. If speed were the signal, Google would be penalizing every brand that does PR well. So the honest answer to “how many is safe” is: as many as you can earn from sites that would pass a human review. The useful questions are how many your site can absorb without looking lopsided, and how many qualifying sites exist in your market.
Why does growth rate matter more than the raw count?
Because 50 new referring domains means something different to a site with 200 of them and a site with 20,000. Ahrefs’ 2018 study pulled the top 20 US results for 10,000 non-branded keywords, giving 200,000 pages, and measured how many followed referring domains each had gained in the previous three months. Most pages ranking first grew by 5 to 14.5 percent a month. Pages ranking lower gained fewer new domains.
Two caveats. The study measures pages, not whole domains, and it is correlational: Ahrefs found no clear link between acquisition pace and later ranking movement, and suggested link quality explains the gap. It is still the only published growth-rate benchmark we know of.
The table shows what three common monthly targets mean for sites of different sizes. The percentages are plain arithmetic (new links divided by existing referring domains), not a measurement of any real site.
| Existing referring domains | 50 new a month | 150 new a month | 500 new a month |
|---|---|---|---|
| 200 | 25.0% | 75.0% | 250.0% |
| 2,000 | 2.5% | 7.5% | 25.0% |
| 20,000 | 0.25% | 0.75% | 2.5% |
| 100,000 | 0.05% | 0.15% | 0.5% |
Five hundred links a month is explosive growth for a 2,000-domain site and a rounding error for a publisher with 100,000. The 150 column shows why small sites get into trouble: a regional law firm or a new e-commerce store adding 150 paid links a month is tripling its profile every quarter, which no comparable business does naturally.
What makes a month of links look unnatural?
Pattern, not pace. When we audit a site hit by a manual action, the monthly count is rarely what stands out. These do:
- Anchor text. A large share of links using the exact commercial phrase the site wants to rank for. Natural profiles are dominated by brand names, bare URLs and generic phrases.
- Linking sites with no readers. Domains with a high Domain Rating (DR, Ahrefs’ 0 to 100 score of a site’s link profile) but almost no organic traffic.
- A shared footprint. Dozens of linking sites that all have a “write for us” page, a sponsored-post price list, and outbound links to casino, CBD or essay-writing sites.
- Every link to a money page. Natural coverage mostly points at homepages and useful content. Paid links point at the pricing page.
Compare that with a digital PR spike. Forty links in ten days from news and trade sites, all pointing to the page that holds the data, with anchors that vary because each journalist wrote their own sentence. The velocity is extreme and nothing about it is unnatural.
What is the real limit on how many links you can build?
Supply. Set a quality bar, count the sites that clear it, and the number of links you can responsibly place each month falls out of the arithmetic.
Our bar is DR 50 or higher, at least 10,000 monthly organic visits that are steady or growing, most of that traffic from the target market, real editors publishing regularly, no “write for us” page or sponsored-post price list, no outbound links to casino, CBD or essay sites, and topical fit. Every article and link then gets a second peer review, and a ninth check asks how often the site is already cited by ChatGPT, Perplexity, Gemini, Claude and Google AI Mode. Our guide to vetting a site before buying a link walks through each check in Ahrefs or Semrush.
Run that filter on a niche like industrial polymers or specialist insurance and you may find a few hundred qualifying sites worldwide, many of which rarely publish outside contributions. Consumer apps, fintech and e-commerce have more candidates, but the same editors are approached by everyone. In our experience, a brand can place roughly five to fifteen links a month that pass all nine checks. Beyond that, the only way to hit a bigger number is to lower the bar, and that is where the risk begins.
The bar matters more now because of AI search. Profound’s analysis of 11.8 billion AI citation data points (Profound, 2026) found that brand websites account for 57 percent of citations globally, leaving about 43 percent for third-party pages such as media, communities and institutions; on ChatGPT, brand sites take 47 percent, so over half of its citations are third-party (WARC summary). Engines recommend the brands the rest of the web vouches for, and a link on a site with no readers is never retrieved. In engagements we track, brands we have placed in third-party articles, guest posts and listicles have seen AI citations rise by roughly 70 to 80 percent on the prompts where those articles are cited.
How should you set your own monthly link target?
- Measure your baseline. Pull your referring-domain count and the last twelve months of new domains in Ahrefs or Semrush. That is your natural rate.
- Pick a growth rate, not a number. Five to fifteen percent a month is the benchmark from pages that already rank. For 800 referring domains that is 40 to 120 new domains a month. For 15,000 it is 750 to 2,250.
- Cap it by supply. Compare step 2 with how many qualifying sites you can realistically place on. The smaller number wins. For most brands the paid, vetted portion lands at five to fifteen links a month; the rest has to be earned through PR and community.
- Mix the sources. Guest posts, link insertions, press mentions and Reddit threads each leave a different footprint. A month made only of templated guest posts looks like what it is.
- Keep anchors boring. Brand names and URLs for the majority.
- Budget for spikes. A data-led PR campaign typically produces eight to twelve DR 60+ linking domains in four to eight weeks. That burst is healthy; plan for it.
- Price it honestly. Our 2026 link building cost guide puts a real editorial link at $200 to $500 and a digital PR link at $1,250 to $1,500. Anything far below that means step 3 was skipped.
How Model Marketing helps
We build links at the pace the market allows, not the pace a spreadsheet wants. Every placement passes the nine-check vetting bar described above. Our link building service covers guest posts at $250 to $320 and niche link insertions at $200 to $300 depending on the site’s DR, with a 12-month free replacement if a link is removed.
When a brand needs more authority than vetted placements alone can supply, our digital PR campaigns earn coverage from publications that do not sell links at any price: a data-led campaign is $8,000 with a floor of six unique DR 60+ linking domains, typically eight to twelve. There is no retainer. You are billed per live unit twice a month, month to month, with 30 days’ notice. Full details are on the pricing page.
For a second opinion on your current velocity, send us your domain and we will tell you how fast your profile is growing and how many qualifying sites exist in your market.