There is no safe number. Google publishes no limit on how many backlinks a site can gain per month, so every figure you see is a heuristic. The workable rule for deciding how many backlinks per month to add is to pace new links as a plausible, slightly faster version of what your site and its competitors already earn, and to spread them over weeks instead of landing them in one day.
New or small sites should start low, mix link types, and scale only after the links are indexed and nothing looks off. Below is a four-step method for working out that pace for your own site.
No official number exists, so measure the pattern instead
What Google does and doesn’t publish
Google’s Search Central documentation gives no threshold for how much link growth is acceptable in a month. What it does publish is a definition of link spam: links created mainly to manipulate rankings, including buying or selling links that pass ranking credit. The policy targets the intent and the pattern, not a particular monthly count. So “20 a month is fine, 200 is not” is guesswork, however confidently it is stated.
Every range in this article is the same kind of thing: an unsourced rule of thumb that can be wrong in your niche. Use them as starting points to test. For why links matter in the first place, see our article on why backlinks are so important; we won’t repeat that here.
Link velocity: a shape, not a count
Link velocity is how quickly new links, or new referring domains, appear over time. The useful question is not “how many?” but “what does the curve look like next to this site’s history and its competitors?”
Links vs. referring domains
A referring domain is a unique website that links to you. Thirty links from one site count as one referring domain. Because a single site can add hundreds of links through a sidebar or footer, raw link counts are noisy. Track your pace in new referring domains per month, and treat total links as a secondary number.
Steps 1 and 2: Baseline your own site and your competitors in an afternoon
You need two reference points before choosing a pace. Both take a few hours with free or paid tools.
- Pull your own referring-domain growth by month. Note your typical monthly gain and any past spikes.
- Check 3-5 ranking competitors over 12 months. Record each one’s monthly gain and take the median.
Step 1: Pull your own referring-domain growth by month
Use the Links report in Search Console, or Ahrefs, Semrush or a similar tool that charts referring domains over time. Write down the usual monthly gain, even if it is zero or one. Then mark every spike and find the cause: a press mention, a directory listing, an old campaign, a scraper site. Spikes with no explanation are worth investigating before you add anything.
Step 2: Check 3-5 ranking competitors over 12 months
Pick competitors that actually rank for your target terms, not just big brands in your industry. Chart their referring domains over the last 12 months and take the median monthly gain. Treat it as a ceiling reference, not a target. A new site that matches an established competitor’s pace is already moving faster than its own history supports.
Why tool numbers are a sample, not the truth
Google says the Search Console Links report is a sample, not a complete list of your links. Ahrefs, Semrush, Moz and Majestic each crawl a different subset of the web, so the same site shows different counts in each. Pick one tool and use it for every comparison, so you are at least measuring with the same ruler.
Competitor data has a second problem. Some of those competitors buy links, and others had a one-off PR spike. Copying their pace can copy their risk. Look at the chart’s shape as well as the median, and discount months with obvious jumps.
Steps 3 and 4: Adjust for site age and for link type
Step 3: The same 30 links on a new site and an established one
Thirty links are background noise on a site with 500 referring domains. On a site with six, they rewrite the profile. The younger the site, the less history there is to dilute a batch, so be conservative. Both scenarios below are hypothetical illustrations, not benchmarks.
| Scenario A: local plumber site | Scenario B: affiliate site | |
|---|---|---|
| Starting point | 4 months old, 6 referring domains | 150 referring domains; top three competitors gain roughly 5-10 new referring domains a month (per the tool’s chart) |
| Hard to defend | 80 links in one week, changing the profile by more than ten times | 100 links in one week |
| Easier to defend | A few links per week across directories, a Google Stack/citation entity and a couple of Web 2.0 profiles, reviewed after each batch | A plan near the competitor median, spread across link types |
Step 4: Which link types can be added faster, and which need restraint
Links are not interchangeable units. Profile, citation, directory, social and other nofollow-type links can generally be added faster than editorial-looking contextual links and guest posts. They carry less weight individually and look more like what a normal business collects. Our piece on Web 2.0 backlinks covers the lower-risk end of that range.
Heavily paid or sponsored placements deserve the slowest pace. Also watch the mix: if a whole month’s links come from one type or one network, the pattern looks engineered even when the total is modest.
Slow is not automatically safe, either. A trickle of poor links does little, and relevance and quality matter more than the schedule. In finance, health, gambling and other competitive or YMYL niches, scrutiny is closer, so lean cautious.
Starting ranges by scenario (rules of thumb to test, not benchmarks)
These are unsourced heuristics. They can be wrong for your niche, and they are starting points to test, not targets to hit. Scale only after links are indexed and nothing looks off. None of this promises any ranking or timeframe.
| Scenario | Starting pace idea | Link-type mix | When to scale |
|---|---|---|---|
| Brand-new site, near zero links | A handful of links per month | Varied low-risk types: directories, citations, social, a few Web 2.0 profiles; no exact-match anchors | After most links are live and indexed and Search Console shows no issues |
| Young site, a few dozen referring domains | Modestly above your own baseline | Low-risk types plus one or two contextual placements | After a clean review of the previous batch |
| Established site, hundreds of referring domains | Move toward the competitor median | Broader mix, including guest posts where relevant | Gradually, if indexing and impressions hold steady |
| YMYL or highly competitive niche | At or below the competitor median | Favour relevance over volume | Slowly, with a review after every batch |
Spike or drip: what the curve looks like to a spam system
The flat line, the cliff and the gentle slope
Picture two referring-domain graphs. The first is flat for months, jumps vertically, then goes flat again. The second rises gently. If you had to review one by hand, you would probably open the first. We don’t know Google’s internals, so don’t read this as a rule about how its systems work. It is a reasonable assumption that abrupt, unexplained changes get more attention than gradual ones.
When a spike is legitimate
Spikes happen naturally. Say a small SaaS product is covered by a niche newsletter and launches on a product directory in the same week. Links jump. But the jump comes with referral traffic, more brand searches and varied anchor text, because real people wrote about the product in their own words. A purchased spike usually lacks those companions. If your spike has an explanation you can document, it is less worrying than one you can’t explain.
What drip delivery does and doesn’t do
Spreading links over weeks has practical benefits:
- Links accrue steadily instead of landing as a day-one cluster.
- Crawlers have time to find and index each batch.
- You can stop and check between batches.
Be clear about the limit, though. Drip delivery is not a guarantee of safety. It does not make links bought to manipulate rankings compliant with Google’s spam policies, and such links can still be devalued or, in worse cases, lead to a manual action. What it does is lower pattern risk and make monitoring possible.
What goes wrong when links arrive too fast or too alike
Six failure modes
- An unexplained spike. Links jump with no press, launch or campaign to account for it.
- Identical or exact-match anchors. Many links repeat the same commercial phrase.
- One link type or one network. Fifty profile links from the same kind of site look like one action, not fifty endorsements.
- Irrelevant or low-quality sources. A plumber’s site linked from unrelated foreign-language pages gives a reviewer little reason to trust the profile.
- Links that never get indexed. If search engines don’t store the page, the link can’t help you.
- Wasted budget. Often the most likely outcome: you pay for volume and nothing measurable happens.
Anchor text: a 40-link example
Anchor text is the clickable wording of a link. Take 40 new links to a running-shoe page. In version one, 30 use “best cheap running shoes” exactly. Real writers rarely all describe a page in the same words, so this looks engineered.
In version two, the same 40 links are split roughly like this (an illustration only, no official ratio exists): 12 brand name, 8 bare URL, 8 generic (“this guide”), 8 partial phrases such as “cheap shoes for runners”, and 4 exact-match. That reads like different people linking in different ways.
Ignored links vs. a manual action
The most likely result of low-value links is that Google’s systems ignore or neutralize them. Google has said publicly, in Search Central materials and through its representatives, that its systems are built to discount many spammy links. Don’t read that as a promise for your particular links.
A manual action for unnatural links is still possible. If one is applied, it appears in the Manual Actions report in Search Console.
Risk, rel=”sponsored” and what mitigation can and can’t do
Google’s spam policies treat links bought to manipulate rankings as link spam. Its guidance on qualifying outbound links recommends rel=”sponsored” or rel=”nofollow” for paid links.
Relevance, moderate volume, mixed anchors and quality over quantity reduce the footprint. They don’t change the policy. Results vary by site and niche, and none of this makes a tactic “safe” or approved by Google.
After each batch: monitoring, pause rules and the delivery report
A pause-rule checklist for every batch
Review a batch two to four weeks after it finishes. If any answer is “no”, stop and diagnose before ordering more.
- Is a good share of the links live and indexed within a few weeks? Check a sample with a search for the exact URL.
- Does the Manual Actions report in Search Console show no issue?
- Does the Links report pick up new linking sites? It is a sample, so absence is a hint, not a verdict.
- Have impressions and rankings on the target pages held up over several weeks, not days?
These are signals, not proof. Content changes, algorithm updates and seasonality all move rankings, so note what else changed in the same window.
What a delivery report is for
A campaign without a report can’t be audited. Each line should show:
- Date live
- Source domain
- Link type
- Anchor text
- Target URL
- Indexed (yes/no)
With that, you can spot duplicate domains, count the anchor mix and each type’s share, and track indexing batch by batch. If a source turns out poor, you also know exactly which links to ask to have removed, or to list in a disavow file, which is Google’s tool for telling it to ignore specific links.
Doing it without scheduling by hand
If you’d rather not schedule and mix link types yourself, the Premium Backlink Packages (Starter, Growth, Dominance) combine several link types in one campaign. It starts within 48-72 hours, is delivered gradually over the stated period, and comes with a report listing every live link, so you can check pace, anchors and indexing.
The same rule applies. Pick the package size against your own baseline and competitor numbers, and start smaller if the site is new. Results vary by site and niche, and no ranking is guaranteed.
Your next step: write down one number before you order
Today, pull your referring-domain history and record your median monthly gain. Do the same for two or three competitors. Then set a first-month target slightly above your own median and at or below the competitor median.
Split it across at least three link types, and put a review date in your calendar for after the first batch lands. Treat the number as a test. It won’t guarantee anything, but it gives you something to compare against when the results come in.
Frequently asked questions
How many backlinks per month is too many?
Google publishes no limit, so no number is universally too many. A better test is whether your monthly gain is a plausible step above your own history and your competitors’ typical pace, and whether you can explain it.
Is 100 backlinks a month safe for a new site?
For a site with only a handful of referring domains, 100 links in a month would change the profile sharply, so be cautious. Nothing is guaranteed safe. Start low, mix link types and review after each batch.
Does drip-feeding links make them safe?
No. It lowers pattern risk and lets you monitor between batches. Links bought to manipulate rankings still fall under Google’s link spam policy and can be devalued or, in worse cases, trigger a manual action.
Should I use rel=”sponsored” on paid links?
Google recommends rel=”sponsored” or rel=”nofollow” for paid links. Such links pass less ranking value, but they follow the guidelines and can still bring referral traffic and visibility.
How long should I wait before ordering more links?
Wait until the last batch is indexed and you’ve checked Search Console for manual actions and for drops in impressions on target pages. That usually takes a few weeks, but it varies by site.
Want this done for you?
Web 2.0 Backlinks builds the links described above, delivers them gradually and sends you a report with every live link. See the details, prices and delivery time on the service page. See Web 2.0 Backlinks →
