BUY WEBSITES
BUY DOMAINS
LOGIN
SIGNUP
PBN Network Size: How Many Sites You Need, the Sizing Matrix by Keyword Competition, and the Domain Quality That Lets You Build Fewer
There is no fixed number of sites a private blog network needs. The honest answer to how big a PBN has to be is set by one variable outside your control: the referring-domain gap between your page and the pages already ranking for your target keyword. The network is sized to close that gap, and not one site larger.
Done badly, network sizing is a vanity number. Operators stack a hundred junk domains at one target, and the outbound-link ratio and synchronized footprint are exactly what Google’s link-graph systems read. Done well, the count is small, the domains are real, and each links once. This guide gives the sizing model both reads share, without telling anyone whether to build a network at all.
It also draws the line the field blurs. The number of sites you need falls as the quality of each domain rises, because a clean, strong aged domain carries more earned authority than three junk drops combined. SEO Domains operates the curated marketplace where that raw material is screened before it is priced, so a network built from vetted inventory needs fewer sites, not more.
PBN network size: how many sites do you actually need?
A private blog network has no universal size. The number of sites you need equals the referring-domain gap between your page and the pages already ranking for your target keyword, divided by the contribution of each site, where one strong clean domain contributes more than a stack of junk ones. Recognised authorities reject any absolute count and benchmark against competitors instead.
The search behind this question wants a single figure. The accurate answer is that the figure is a function, not a constant, and the inputs are the competition for the keyword and the quality of each domain. A low-competition phrase can move on a handful of links. An entrenched commercial term can resist hundreds.
Why the field gives one number and why that number is wrong
The one ranking page that commits to a figure, the EasyBlogNetworks guide on the PBN link count it takes to rank, states that ten links suit an average-competition keyword and works an example down to about ten sites linking once each. That figure is a reasonable midpoint, and it is also a trap when copied without its context, because the same page lists five inputs that change the answer: niche competition, keyword competition, on-page strength, the natural backlinks already present, and the competitor backlink profile.
Recognised authorities go further and decline a number altogether. The Ahrefs position on link benchmarking, applied across this hub, is that there is no one-size-fits-all count and the only honest benchmark is the growth pattern of the pages you are trying to outrank. Size is therefore read off the search results page, not chosen in advance.
The two variables that set the count
Strip the question to its core and two variables remain. The first is keyword competition, measured as the referring-domain count of the pages already ranking, which sets the total authority the page has to acquire. The second is domain quality, which sets how much authority each site in the network contributes toward that total.
Raise the second variable and the first one is satisfied with fewer sites. A network of strong, clean aged domains reaches a referring-domain target with a smaller count than a network of weak drops, which is the central reason quality decides size. The remaining sections turn this into a tiered matrix, a measurement procedure, and a footprint rule.
Three counts, not one: network minimum, ranking minimum, and sites per money site
The single-number question hides three separate counts. The first is the minimum to be a network at all, which is structural. The second is the minimum to move one keyword, which is the competitor-gap count. The third is the count of network sites that can safely feed one money site, which is a footprint limit. Conflating them is the field’s core error.
The ranking pages treat “the number of sites” as one question. It is three, and the answer to each is different. The EasyBlogNetworks guide, the only competitor with figures, blurs the link count into the site count throughout, which is the gap this section closes.
Count 1: the network minimum (structural)
Two or more owned sites pointing at one target is already a network in policy terms. The minimum to BE a PBN is structural, not a performance figure. A single domain is not a network; the moment a second owned site links to the same money site, the network footprint exists.
Count 2: the ranking minimum (the competitor gap)
The sites needed to move one keyword equals the referring-domain gap to the ranking pages, divided by per-site contribution. This is the only count tied to a result, and it is read from the search results, not picked. It rises with competition and falls with domain quality.
Count 3: sites per money site (the footprint limit)
However large the network, the count that safely feeds one target is capped by the outbound-link ratio detection reads. A spike of links from a network at one page is the signal a link-graph system separates from editorial linking, so this count is bounded by footprint, not budget.
The conflation error
The field collapses links, sites, and targets into one number. Ten links is not ten sites if a site links twice, and ten sites is not safe if all ten hit one page at once. Separating the three counts is what turns a guessed figure into a defensible one.
The sizing matrix: referring-domain targets by keyword competition
Network size scales with keyword competition. A low-competition keyword is moved by a small contribution, a medium one needs more, and an entrenched commercial term resists a large network of weak sites entirely. The matrix below tiers the referring-domain target by competition band, derived from the competitor-baseline method, with the per-site count falling as domain quality rises.
The figures here are reference ranges, not promises. They fuse the EasyBlogNetworks midpoint (about ten links for an average keyword) with the Ahrefs benchmarking principle (read the competitor referring-domain count, do not assume an absolute), and they assume one editorial link per site. Treat the per-site count as an upper bound that clean domains pull downward.
| Competition band | Referring-domain target (read from competitors) | Sites with junk domains | Sites with clean strong domains |
|---|---|---|---|
| Low (long-tail, local, new niche) | Match the lowest-ranked competitor on page one | 8 to 15, often deindexed before they rank | 3 to 6 quality sites, one link each |
| Medium (average commercial keyword) | Around the EasyBlogNetworks 10-link midpoint | 15 to 30, footprint-heavy and fragile | 6 to 12 quality sites, one link each |
| High (competitive commercial term) | Match the median page-one referring-domain count | 30 to 100, the classic over-built network | 12 to 25 quality sites, plus earned links |
| Enterprise (entrenched head term) | Exceeds what any network alone supplies | Any count fails; the gap is too large to fake | A network cannot close this alone; brand and earned links are required |
Two reads of the matrix matter. First, the count climbs steeply with competition, so a single number copied from a low band is useless in a high one. Second, the clean-domain column is roughly half the junk column at every level, because a real earned profile contributes more per site. The enterprise row is the field’s blind spot: no network alone moves an entrenched head term, and the largest over-built networks are the ones that tried.
How to size a network from the competitor baseline, step by step
Sizing is a five-step procedure: read the competitor referring-domain count for the target keyword, subtract what your page already holds, divide the gap by realistic per-site contribution, cap the result at the footprint limit, and round down for clean domains. The done-right move at every step is to size to the measured gap; the mistake is to pick a round number and build to it.
The EasyBlogNetworks guide names the competitor backlink profile as a factor without showing how to read it. The Ahrefs benchmarking principle says to read it without giving the steps. The sequence below makes the measurement operational, so the count comes from data instead of habit.
-
Read the competitor referring-domain count
Take the keyword you are targeting and record the referring-domain count of the pages ranking on page one, using an authority tool. The metrics that matter and how to read them are set out in the Domain Authority & Metrics hub. The page-one median is your authority target, the level the page has to reach.
The mistake: skipping the read and starting from a round number such as ten or fifty. A figure with no competitor basis is a guess, and a guess is what over-builds or under-builds the network.
-
Subtract the authority your page already holds
Your money site rarely starts at zero. Count the referring domains it already earns editorially and subtract them from the target. The done-right move is to size only the gap that remains, because existing natural links do part of the work and reduce the network you need.
The mistake: ignoring existing links and building to the full target. That over-builds the network, stacks footprint you did not need, and wastes clean domains on authority you already had.
-
Divide the gap by realistic per-site contribution
Each network site contributes one link from a domain of a given strength. Divide the remaining gap by a conservative per-site contribution to get the raw site count. The done-right move is to assume each site does less than you hope, so the estimate errs small instead of large.
The mistake: assuming each junk domain contributes as much as a clean one. Inflated per-site assumptions produce a count that looks small on paper and fails in practice, prompting a second, footprint-heavy build.
-
Cap the count at the footprint limit
Whatever the division returns, cap it at the number of sites that can point at one money site without a detectable spike. The done-right move is to respect the outbound-ratio cap from PBN footprints: the complete list and the pacing in PBN link velocity: how fast is too fast, spreading links over time instead of firing them at once.
The mistake: pointing the full network at one target in a single window. A synchronized burst from a block of owned sites is the network-to-money ratio detection reads first.
-
Round down for clean domains, and source them screened
A clean, strong domain contributes more than the conservative estimate, so the final count rounds down when the raw material is real. The done-right move is to source screened aged or expired domains so the count shrinks honestly. Browse vetted inventory on the SEO Domains marketplace, and read the build sequence in Setting up your first PBN: full walkthrough.
The mistake: padding the count with cheap drops to hit a target faster. Junk domains are already devalued in the link graph, so they raise the count and the risk at the same time.
Why one link per site is a sizing rule, not folklore
The one-link-per-site rule is a footprint-control mechanism, not superstition. Outbound-link concentration and the network-to-money-site ratio are precisely what link-graph detection reads, so the number of links a site sends, and where they point, is a sizing variable. EasyBlogNetworks reports that linking each site once keeps the pattern organic and warns against any site carrying fifty to a hundred outbound links.
The field repeats “link each site once” as a rule of thumb without the reason. The reason is structural. A page that exists to send a single editorial link reads like a publisher referencing a source. A page sending its links to one commercial target reads like a node in a coordinated network, and the ratio between those two patterns is computable.
The outbound-link math detection reads
Consider the two extremes EasyBlogNetworks describes. A site linking out once, to one relevant target, sits inside the normal range of how real sites behave. A site carrying fifty to a hundred outbound commercial links, which is what certain sellers advertise, sits far outside it. That guide states plainly that a buyer has reason to be cautious when a seller claims a high outbound-link count, because the concentration itself is the risk.
The sizing consequence is direct. If each site links once, a network needs as great a site count as the gap requires. If sites link multiple times to cut the count, the outbound concentration rises and the footprint with it. The rule trades site count against footprint, and the one-link version keeps the footprint low at the cost of needing more clean domains.
The network-to-money-site ratio
The second ratio is the whole network against the single target. The deeper the inbound links concentrate on one money site from sites that link nowhere else meaningful, the clearer the coordination. This is the structural signal that makes the model, explained in What Is a PBN (Private Blog Network), fragile by design, and it is read by the SpamBrain link-graph system Google deployed in its December 2022 link-spam update. Site count, links per site, and link targets are three dials on the same footprint, and sizing controls all three.
Over-building: what too many sites looks like, and how it is detected
Over-building is the dominant sizing failure. A network larger than the competitor gap adds footprint without adding rank, and the surplus sites are pure detection surface. Google reads the over-built signature through the network-to-money ratio, synchronized linking, and shared infrastructure, increasingly with the SpamBrain machine-learning system. The cost when it fails is documented, not abstract.
The instinct the field encourages is “build more, hide better.” The sizing reality is the reverse. Past the point where the gap is closed, each additional site is a liability that raises the chance of recognition while doing nothing for rank. Over-building is the leading way a network that would otherwise have held gets caught.
The over-built signature
An over-built network leaves a recognisable shape. These are the mistakes that turn surplus size into a penalty, listed so the signature is identifiable:
- A site count far above the competitor referring-domain gap, with surplus sites adding only risk.
- A high proportion of network sites pointing at one money site, raising the network-to-money ratio.
- Synchronized linking, where a large block of sites links to the target inside one short window.
- Shared hosting or registration across the surplus sites, since a bigger network is harder to diversify.
- Thin or duplicated content on the extra sites, built fast to reach a count rather than to read as real.
Each surplus site multiplies these signals. A network sized to the gap leaves a small surface. A network built to a vanity number leaves a large one, and the detection systems below scale with that surface.
How the over-built network is detected
Detection reads size through pattern, not through a counter. The Penguin update of 24 April 2012 first targeted manipulative link patterns directly, and the December 2022 link-spam update deployed SpamBrain, Google’s machine-learning spam system, to evaluate the link graph for coordinated structures. A larger network gives these systems more nodes to correlate, so over-building accelerates the recognition it was meant to delay. Ownership ties through registration data, now read via RDAP after it replaced WHOIS as the standard ICANN lookup on 28 January 2025, add a further correlation layer across a bigger network.
What over-building costs when it fails
The downside is financial. DomCop, an expired-domain data platform selling into the same supply as this market, puts published recovery costs in the range of 312 to 9,380 US dollars per penalised property, with revenue losses on hit sites reported as high as 80 percent. Treat those as cited reference figures, not a guarantee. A larger network multiplies the exposure, because every surplus site is another property that can be hit, and the recovery work in Recovering a deindexed PBN site scales with the count.
Quality over quantity: why a small clean network beats a large junk one, and the single-site alternative
The quality inversion is the lever the field omits. A small network of strong, clean domains outperforms a large network of junk, because junk domains are already devalued in the link graph and contribute little while adding maximum footprint. The durable alternative the field rarely names is to rebuild one strong aged domain into a single owned authority site, with zero network footprint.
The whole sizing conversation assumes that more sites is more authority. It is true only when each site is real. A flagged drop in the link graph contributes near nothing yet counts fully toward the network surface detection reads, so padding the count with junk raises risk faster than rank.
Why junk raises the count and the risk together
A junk domain fails twice. Its toxic or spam-inflated profile is already discounted, so it adds little authority and the count has to rise to compensate. At the same time it still occupies a slot in the network, so the footprint grows with every weak site added. Quantity built on junk is the worst of both: a high count and a low return, which is the exact profile the largest penalised networks share.
The single-site alternative the field omits
The single durable answer to the site-count question is sometimes one. Rebuilding a single strong aged or expired domain into a real, owned authority site keeps the inherited authority while removing the interlinking footprint, the shared-ownership signal, and the network-to-money ratio entirely, because there is no network. The authority is the same raw material the network borrows; the structure is the opposite, and it cannot be withdrawn in a spam-update refresh. This is the path the underlying search behind network sizing usually reaches for.
The sizing checklist: every mistake, why it is caught, and the fix
The mistakes that mis-size a network are a short, repeatable list. Each one either over-builds the count, raises the footprint, or both, and each has a documented fix. The fix points back to the same place every time: size to the measured competitor gap, link each site once, and start from clean domains so the count stays low. Use this as the scannable reference.
The table consolidates the sizing errors scattered through the matrix, the step sequence, and the over-building section into one place. The left column is the mistake, the centre column is why it is detectable or ineffective, and the right column is the done-right fix. Read top to bottom, the fixes describe a network sized to the gap, built from clean material, with no surplus surface.
| The sizing mistake | Why it fails or is caught | The fix (done-right move) |
|---|---|---|
| Picking a round number (ten, fifty, a hundred) | A count with no competitor basis over-builds or under-builds the gap | Read the page-one referring-domain count and size to that gap |
| Ignoring the authority the page already holds | Building to the full target stacks sites you did not need | Subtract existing editorial links, size only the remaining gap |
| Assuming junk contributes like a clean domain | Inflated per-site estimates fail in practice and prompt a second build | Assume conservative per-site contribution, round down for clean domains |
| Over-building past the gap | Surplus sites add detection surface without adding rank | Stop at the count that closes the gap, treat every extra site as risk |
| Pointing the whole network at one target at once | A synchronized burst raises the network-to-money ratio SpamBrain reads | Spread links over time within the velocity cap, link selectively |
| Multiple outbound links per site to cut the count | Outbound-link concentration is a link-graph footprint | One editorial link per site, accept the higher clean-domain count |
| Padding the count with cheap drops | Flagged domains are already devalued, raising count and risk together | Source screened domains so quality pulls the count down |
| Shared hosting and registration across a big network | A larger network is harder to diversify, tying sites to one owner | Keep the count low enough to diversify infrastructure fully |
| Sizing a network against an enterprise head term | No network alone closes an entrenched gap; the attempt over-builds | Recognise the ceiling, pair earned links and brand with the network |
| Treating size as a budget question, not a footprint one | The cap on sites per target is detection, not cost | Size to the gap and the footprint cap, never to the budget alone |
One pattern runs down the fix column. The recurring move is to size to the measured gap instead of to a round number, and to raise domain quality so the gap closes with fewer sites. A junk domain breaks this in the first row and inflates every count after it, because a devalued profile cannot be sized away. That is why sourcing the right raw material is the practical starting point of sizing, not an afterthought, and it is the foundation the final section returns to.
PBN network size frequently asked questions
The five questions buyers and SEOs raise when they search for the site count a PBN needs, answered against the competitor-baseline method and the quality-over-quantity inversion this guide draws.
Q1What site count does a PBN need to rank a keyword?
There is no fixed number. The count equals the referring-domain gap between your page and the pages already ranking for that keyword, divided by the contribution of each site. EasyBlogNetworks reports about ten links for an average-competition keyword as a midpoint, but recognised authorities benchmark against the competitors instead of assuming a constant, so the figure is read from the search results, not chosen.
Domain quality moves the count. A network of strong, clean domains reaches the same target with roughly half the sites a junk network would need.
Q2What is the minimum number of sites to be a PBN?
Two. The moment a second owned site links to the same money site as the first, the network footprint exists in policy terms. A single owned domain rebuilt into one authority site is not a network, which is why it carries no network-to-money ratio and no interlinking signal. The minimum to BE a network is structural and separate from the count needed to move a keyword.
Q3Is a bigger PBN safer or more dangerous?
More dangerous past the point where the competitor gap is closed. Every surplus site adds detection surface without adding rank, gives the SpamBrain link-graph system more nodes to correlate, and is harder to diversify across hosting and registration. A larger network also multiplies the recovery exposure, since each property can be hit. Size to the gap and stop.
Q4What link count does each PBN site send to the money site?
One, as an editorial reference. Outbound-link concentration is a footprint detection reads, and EasyBlogNetworks warns against any site carrying fifty to a hundred outbound links. Linking once per site keeps the pattern looking organic and trades site count against footprint: one link each means more clean domains, which is the safer side of that trade.
Q5Can a smaller network of strong domains beat a large network of weak ones?
Yes, and this is the central inversion. A junk domain is already discounted in the link graph, so it adds little authority while still occupying a slot in the network surface. A clean, strong aged domain contributes real earned authority, so fewer sites reach the same target with a smaller footprint. The count falls as the quality of each domain rises.
The foundation: clean domains decide how few sites you need
Domain quality is the variable that sets network size. A clean, real, earned-authority domain contributes more per site, so the count falls as quality rises, and a single strong domain can replace a stack of junk ones. Sourcing from a screened catalogue is what keeps the count low and the footprint small. SEO Domains operates that curated marketplace.
Why quality sets the count
Every section of this guide converges on one lever. The sizing matrix halves the count when domains are clean. The step sequence rounds the count down for real material. The over-building section shows junk inflating the count and the risk together. The quality inversion makes the point directly: the cheapest way to need fewer sites is to make each domain stronger.
The asset versus the scheme
A clean aged domain’s earned authority is a legitimate asset you can own under your own name. Only a careless network built around it is the liability. Buying a quality expired domain is not the risky part, and treating size as a budget exercise instead of a quality one is the error the field makes. The right network is small because each domain in it is real.
How to source domains that keep the count low
A domain that contributes real authority survives a profile check before money changes hands. The signals that decide per-site contribution are documented across the authority-metrics hub:
- Referring domains and the quality, not the raw count, of the links pointing in.
- DR and DA, the Ahrefs and Moz authority scores, read together rather than singly.
- Trust Flow and the TF:CF ratio from Majestic, which surface link-spam patterns a single metric hides.
- Link age, organic traffic history, and a clean spam screen with no toxic inheritance.
A junk domain passes none of these, so it raises the count and the footprint at once. A vetted domain passes them and contributes enough to shrink the network it joins.
| Check | Junk domain (count rises) | Vetted domain (count falls) |
|---|---|---|
| Per-site authority contribution | Near zero, already devalued | Real earned authority per site |
| Effect on network size | More sites needed to compensate | Fewer sites reach the same target |
| Footprint impact | Larger network, larger surface | Smaller network, smaller surface |
| Screening | None, sold on raw metric | Multi-signal screen before listing |
| Outcome in sizing | High count, high risk | Low count, durable contribution |
Browse curated aged and expired domains with clean profiles
The legitimate demand behind every network-size search is access to real domain authority you can own openly, in as few sites as possible. That is the product, not a network service, not hosting, and not a done-for-you scheme. SEO Domains operates the curated marketplace where aged and expired domains are screened across their backlink profiles and authority metrics before they are listed and priced, so the count stays low because the quality is high.
