Can You Rank on Google Without Backlinks? Legendary SEO David Quaid Answers

Edward Sturm

Abstract

Seventy-five minutes answering the question everybody wants a formula for: can a site keep ranking by scaling content if its backlinks stay flat. The answer is that there is no fixed ratio, and the rest of the conversation is what to do in the absence of one — which turns out to be a long-tail climb, a model of authority as current rather than stock, and a set of honest limits about where that climb stops.

• A site scales by climbing from one low-difficulty keyword to the next, and machine-built encyclopedias and listing sites have done it without any link programme.
• Link authority becomes decisive only against incumbents with near-infinite history — the wall is history rather than quality.
• An internal link works only from a page that already ranks and gets traffic, not merely one that is indexed.
• Organic traffic arriving at a page is direct proof it is not penalised.
• Duplicate content is not penalised, because roughly a fifth of the index is duplicate — spam is filtered rather than punished.
• A blog on a subdomain is a separate domain and shares no topical authority with the main site.

Everything below follows the conversation in order, through the scaling question, the authority-as-current model, the link-buying answer, and a long run of structural questions — subdomains, duplicates, siloing, exact-match domains — and out to why interest in the channel is rising. For anyone building a personal brand the useful part is the register: several answers are explicitly "I don't know how they detect that", and the practice is built around the uncertainty rather than around a claim about it.

Chapter summaries

00:28 - There is no fixed ratio between backlinks and content

There is no fixed ratio between backlinks and content; whether scaled content is flagged as spam is a genuine black box nobody can see into.

"That's a really good question, and I think we're seeing."

— 00:58

There is no fixed ratio between backlinks and content, and whether scaled content gets flagged is not decided by a number on either side. The question opens the episode because it is the one everybody wants a formula for, and the answer is that no formula exists.

01:53 - A site scales by climbing from one low-difficulty keyword to the next

You can scale by climbing from one low-difficulty keyword to the next, ranking, getting clicks, and using that to reach the next — slow on its own, faster with links, or years without.

"It right. I think that as they have reduced the."

— 01:51

A site scales by climbing from one low-difficulty keyword to the next — rank, earn clicks, and use what that page now carries to reach the next one. It is the cornerstone model stated as a growth path rather than as a linking tactic.

02:31 - A machine-built encyclopedia scaled without any link programme at all

A machine-built encyclopedia scaled without a backlink programme, holding and rotating positions, but is running out of steam.

"What's interesting about groipedia, which a lot of SEOs have been talking about."

— 02:31

It earns links passively and rides easy uncontested keywords, which is exactly where it runs out: the ranking report goes steady and then flattens at the point authority is needed. Its link profile is also new and uncoordinated, so it is not proof that a plan is unnecessary — only that a plan is not a precondition.

04:25 - Listing and job sites grow to millions of clicks without buying links

Job, e-commerce and listing sites routinely grow from thousands to millions of clicks with no new links, because programmatic pages ride the site's topical authority.

"I don't know what — I don't know how Google detects machine scale content."

— 04:25

Listing, job and commerce sites routinely grow from thousands to millions of clicks without buying links. The pattern is programmatic rather than editorial, and it is common enough that any theory of ranking has to accommodate it.

10:51 - Link authority becomes decisive against incumbents with near-infinite history

Link authority becomes decisive when you rank against sites with near-infinite ranking and click-through history; you win the uncontested easy keywords, then run out of steam as volume and required authority climb.

"A lot of SEOs expected it to drop and saw traffic."

— 10:51

Link authority becomes decisive when the competition is an incumbent with near-infinite ranking history. The wall is history rather than quality, which is why the long-tail climb exists.

12:37 - What counts as machine-scaled content is a black box

What counts as machine-scaled content, and whether it trips a helpful-content-type score, is the real unknown of SEO; doubts that backlinks can rescue a machine-scaled site.

"Of steam. And Red Gaga spoke about this, or put."

— 12:35

What counts as machine-scaled content, and whether it trips a quality score, is a black box. The honesty is the useful part — the mechanism is not known, so the practice is built around the uncertainty rather than around a claim about it.

14:44 - Answers placed on one page gained traffic and then lost every impression

A first-hand test: FAQ answers placed on ONE page gained traffic, then lost all impressions and rankings within about six days — repeated three times.

"I have seen, where I was testing FAQs on single pages."

— 14:44

A first-hand test: answers placed on one page gained traffic and then lost every impression. It is the sharpest evidence in the episode for the one-question-one-page structure, and it comes from having run it rather than from reasoning about it.

16:11 - Going broader into new topics is what keeps the growth going

You eventually hit a wall; going broader — new topics, new low-difficulty keywords — keeps growth alive, whereas scaling up in one topic means fighting older, entrenched pages.

"As long as you can avoid."

— 16:11

You eventually hit a wall, and going broader — new topics, new low-difficulty keywords — is what keeps the growth going. The expansion is forced by the ceiling rather than chosen.

16:48 - Authority behaves like electrical current carried by pages with traffic

Authority behaves like electrical current: pages with traffic carry it; you may grow without links if you truly master shaping authority, but links accelerate it.

"Authority is like current, right? Like if you think about."

— 16:48

Authority behaves like electrical current: pages carrying traffic carry it. Flow rather than stock is the model the whole pillar keeps arriving at, and it is why routing matters more than accumulating.

18:13 - An internal link works only from a page that already ranks and gets traffic

Internal-link a new page from a page already ranking and getting traffic — not merely one that has a backlink; ranking plus clicks alone builds authority.

"Because you hear a lot of people saying."

— 18:13

An internal link works only from a page that already ranks and gets traffic — not merely one that is indexed. The distinction is the whole of the internal-linking argument, and it disposes of most automated linking in a sentence.

19:20 - Outranking a nearest seed in the link graph needs links rather than cleverness

To outrank a top national newspaper — a nearest seed in the link graph — you would need links from roughly three sites its size; earned authority alone will not do it.

"Easier to analyse if you compare, for example, let's say you."

— 19:20

Outranking a nearest seed in the link graph — a major national publisher — needs links rather than cleverness. The honest limit: there are positions the long-tail climb does not reach.

19:53 - A pogo-sticking reader can overcome a page's link advantage

If users read a page, dislike it and bounce back to click competitors (pogo-sticking), click-through should overcome the link graph and derank even a strong-domain page — otherwise the system is flawed.

"In terms of user experience and authority, which."

— 19:53

A reader who lands, dislikes the page and bounces back to click a competitor can overcome that page's link advantage. Behaviour beats accumulated authority at the margin, which is what makes the click quality worth protecting.

23:29 - Years in one niche where real businesses transact produce links without buying

No need to buy links: years spent in one niche where real businesses transact and link to each other; acting as a broker who arranges genuine guest posts between non-competing clients.

"So I don't need to, in the simple sense."

— 23:29

Years spent in one niche where real businesses transact produce links without buying any. The claim is about time and position rather than about technique — the links arrive because the operation is real.

24:36 - Co-writing partner-network articles with a large software firm is the cross-link play

The partner cross-link play: co-writing partner-network articles with a large software platform's team and cross-linking, showing the engine real businesses transacting; an authoritative partner link is powerful.

"Having one company write a post, or contribute a post to."

— 24:36

The partner cross-link play is co-writing partner-network articles with a large software firm. Both sides get a page they wanted, and neither had to ask for a link.

25:51 - A page's value is its traffic rather than its brand

PR often builds pages that fill space but get no traffic and contribute nothing; a page's value is not its brand but whether it has traffic or backlinks — the only universal, objective measure.

"Absolutely. And it's a couple of projects I've worked."

— 25:51

A page's value is its traffic rather than its brand. Public-relations work often produces pages that fill space and get no visits, contributing nothing — which is a hard measure and a fair one.

27:26 - Organic traffic arriving at a page proves it is not penalised

Self-correcting system: if the engine sends organic traffic to a page it is not penalised, so judge a link by whether its host page has traffic, not by how "real" the site looks — many PR sites are dead link farms.

"Like organic traffic. Yeah. So if you think about a."

— 27:26

If the engine sends organic traffic to a page, that page is not penalised. It is a self-correcting read: the thing people worry about is directly observable, and the observation is free.

29:03 - Each outbound link loses most of what it carries, so authority gets spread

Spread authority rather than concentrate it: each outbound link loses about 85%, a page cannot emit a thousand links, so 10-20 authoritative domains pointing to different parts beats piling everything on one page; if you must concentrate, point at the blog root so new posts auto-index.

"We didn't start with Microsoft. We started with much smaller partners in the UK."

— 29:03

Each outbound link loses most of what it carries, so authority gets spread rather than concentrated. The arithmetic decides the structure, and it is the reason a page with many links out is a page giving away very little each time.

31:58 - Good content is not the lever, and it is too subjective to be one

"Good content" is not the ranking lever and is too subjective to be one; telling people to write good content will not stop spam, and ranking positions get taken daily regardless of content merit.

"The good content problem is that it's very subjective, right?"

— 31:58

Telling people to write better is not advice that can be acted on, which is the practical objection rather than the theoretical one.

33:47 - Relevancy is a gate rather than a limit or a form of authority

Linking gives the power to shape authority; relevancy is a gate, not a limit or a form of authority in itself — you can pivot, but a new segment's first pages start at the bottom and must be grown.

"I think it's a good place to start."

— 33:47

Relevancy is a gate rather than a limit or a form of authority. It decides eligibility and nothing else, which is why starting narrow does not box a site in.

34:38 - Brand-new human-coined keywords are the system's weak spot

The system's weak spot: target brand-new, human-coined low-difficulty keywords the engine cannot get ahead of, rank and get clicks to earn topical authority, then link the new cluster from a middle common-denominator keyword.

"So two ways. One, you start at the bottom."

— 34:38

The system's weak spot is brand-new, human-coined low-difficulty keywords the engine has not seen before. Nothing has accumulated against them yet, which is the only kind of position that is genuinely uncontested.

36:25 - Bridging through an intermediate topic moves a site laterally

You move laterally as well as vertically by bridging through an intermediate topic; each new index starts with little authority and climbs from scratch.

"Happen is, if you are writing and very successful in one."

— 36:25

A site moves laterally as well as vertically by bridging through an intermediate topic. Each bridge is an adjacency rather than a leap, which is how a narrow start becomes a broad one without ever competing head-on.

39:17 - Sitewide topical authority and trust exist above a certain threshold

Yes — sitewide topical authority and trust exist (visible above roughly a domain-authority of 10); macro pogo-sticking erodes it, and after core updates a cascade drops one page then the next until the site cannot rank even its old easy keywords.

"I do, and having seen a lot."

— 39:17

Sitewide topical authority and trust exist, and become visible above a certain authority threshold. It is a qualified yes rather than a flat one, which matters because the question is usually asked as a binary.

41:31 - Machine-scaled content risks missing intent and drawing readers back

A core risk of machine-scaled content: it may miss search intent, drawing pogo-sticking that then hurts the rest of the site.

"This is also one of the major risks with machine."

— 41:31

A core risk of machine-scaled content is that it misses search intent, drawing readers who bounce straight back. The risk is behavioural rather than about detection, which makes it the more serious of the two.

42:10 - Quality can be siloed to a page or a subfolder

Quality can be siloed to individual pages or subfolders; a folder with a strong engagement profile can stay insulated, but effects leak between "firewalls" once other content comes to depend on the bad content.

"What level do you think — do you think Google silos quality scores within subfolders?"

— 42:10

Quality can be siloed to individual pages or subfolders — a folder with a strong engagement record is read differently from the rest of the site. That is what makes a contained experiment survivable.

44:25 - Near-duplicate keywords start as one page, and the clicks decide

For near-duplicate keywords (90-inch, 91-inch, 92-inch curtains) try one page first; the engine decides whether variants are semantically the same from user click behaviour, which you cannot control — build separate pages only if forced, and expect cannibalisation otherwise.

"Yeah. All right. So this was a."

— 44:24

For near-duplicate keywords — ninety-inch, ninety-one-inch, ninety-two-inch — start with one page and let the clicks decide whether they are the same thing. The engine's own behaviour answers a question the operator cannot.

48:24 - Duplicate content is not penalised, because a fifth of the index is duplicate

Duplicate content is not penalised — about 20% of the index is duplicate; spam is simply what people do not want, which is subjective, so being specific about products is fine and most "penalty" beliefs are misread cannibalisation.

"Talking about duplicate content. They can't care about duplicate content because."

— 48:24

Duplicate content is not penalised, because roughly a fifth of the index is duplicate. Spam is simply filtered rather than punished, which is a materially different mechanism from the one people fear.

50:36 - A blog on a subdomain is a separate domain that shares no authority

A blog on a subdomain is a separate domain that does not share topical authority with the parent; host it as a /blog subfolder instead — a subdomain quadruples the effort for nothing.

"Yeah. All right. Should a blog be on."

— 50:36

A blog on a subdomain is a separate domain and does not share topical authority with the main site. It is a common and expensive mistake, made for reasons that are usually about the content management system.

53:33 - A forum on a subdomain is a human-scaled content engine

Forums on subdomains are a human-scaled content engine (user-generated, not machine-scaled) producing endless pages that can link up to the parent; a forum is effectively a standalone site, which is why it ranks.

"So what about forums? People like put."

— 53:33

A forum on a subdomain is a human-scaled content engine — user-generated rather than machine-generated — which is the one case where the split is doing something useful.

55:13 - Topical authority is an array, and some of it leaks upward

Topical authority is an array — a topic plus a value, not one number; some of it leaks from a subdomain up to the parent domain but not back down, so authority tends to flow up.

"Is there a difference between raw authority and topical authority?"

— 55:13

Topical authority is an array of topics and values rather than one number, and some of it leaks upward rather than down. Authority is local before it is global.

57:51 - An exact-match domain used as a subdomain behaves like a folder

Q5: an exact-match domain used as a subdomain behaves like a folder but must be built as its own entity, so it is not worth it — you only change the nature of the work.

"All right, the last question is, it's funny."

— 57:51

An exact-match domain used as a subdomain behaves like a folder but has to be built as its own site. The behaviour and the obligation point in different directions, which is the trap.

1:00:03 - Keyword-in-domain names are becoming close to mandatory as listicles revive

Exact-match and keyword-in-domain names are becoming close to mandatory: as listicles are de-grounded, a multi-domain strategy resuscitates them and spreads ranking, lowering the effort-to-return ratio.

"Such a great idea. It's going to become more."

— 1:00:03

Keyword-in-domain names are becoming close to mandatory as listicles revive: when the listing is what ranks, the name in the listing is doing the work.

1:01:14 - A second keyword domain firewalls the risky work from the brand

A second keyword domain firewalls risky programmatic SEO from a valuable brand domain; worst case it loses, best case two domains reinforce each other — and because large language models quote entities, not domains, a branded keyword domain reads as the brand.

"Protect your silo, or firewall against penalties, right?"

— 1:01:14

A second keyword domain firewalls risky programmatic work from a valuable brand domain. The separation is insurance rather than strategy, and it costs the price of a domain.

1:03:33 - A branded exact-match domain can earn a top paid quality score

A branded exact-match domain can earn a top paid-search quality score, saving tens of thousands a month on bids for a roughly $20 domain; consumers may trust a brandable keyword domain over a bare "best-x" one, and inbound links that use the brand or domain name add keyword context.

"That'll be good enough, and it'll also be good enough for."

— 1:03:33

A branded exact-match domain can earn a top paid-search quality score, saving a great deal on advertising. It is a second return on the same decision, and it is measurable in a way most domain arguments are not.

1:05:28 - The exact-match devaluation is real and neither permanent nor absolute

The leak's exact-match "devaluation" is real but neither permanent nor absolute — a demotion that a few links and clicks overcome; it is a democratisation mechanism letting other sites rank for a phrase an incumbent domain would otherwise own outright.

"Again on the leak showing exact match domains."

— 1:05:28

The exact-match devaluation described in the leak is real and neither permanent nor absolute. Both halves are held at once, which is the register the whole episode operates in.

1:07:15 - Optimisation is machine-proof, because a model cannot deliver the truth

SEO is "AI-proof": large language models carry disinformation and cannot deliver "the truth" — the goal is the best answer, not the truth, so critical thinking and experiment still win.

"I'm glad. I just think like SEO is."

— 1:07:15

Optimisation is machine-proof, because a model carries disinformation and cannot deliver the truth. It is the strongest claim in the episode and it is made as a claim rather than as a finding.

1:09:16 - Interest is at an all-time high because the rival channels are vanishing

Interest in SEO is at an all-time high because rival channels — email and the major social platforms — have become hard, costly or ineffective; few global channels remain as easy, and talk of the engine dying is confirmation bias and social inertia.

"Right? So I think it's going up because."

— 1:09:16

Interest is at an all-time high because the rival channels are vanishing — email and the major social platforms no longer deliver what they did. The growth is relative rather than absolute.

1:11:37 - A new owner notices the brand not showing before anything else

Starting an online business is easier and less intimidating, and new owners immediately ask why the site does not show for its own brand — and the question they are asking is SEO, now extended to "how do I show up in an LLM"; the conversation has moved from design to data, and senior leaders now join the SEO calls.

"I think it's also so much easier now to."

— 1:11:37

Starting an online business is easier and less intimidating than it was, and the first thing a new owner notices is the brand not showing up. That is where the demand comes from, and it is the note the episode closes on.

Personal Branding Lessons

An hour of questions answered by somebody willing to say which parts are unknown. The moves below are what survives that honesty.

Stop looking for a ratio between links and content

There is no fixed ratio between backlinks and content, and whether scaled content gets flagged is not decided by a number on either side. This is the question everybody wants a formula for, and the absence of one is the finding. Machine-built encyclopedias and large listing sites have scaled to millions of clicks without any link programme at all — so whatever the rule is, it is not that links are a precondition. 00:28

Climb from one low-difficulty keyword to the next

Rank for something easy, earn the clicks, and use what that page now carries to reach the next term. That is the growth path rather than a linking tactic, and it is what makes a site without links viable at all. The ceiling arrives eventually, and going broader into new topics is what moves it. 01:53

Internal-link only from a page that already gets traffic

A link works from a page that ranks and receives visits — not merely one that is indexed. That single distinction disposes of most automated internal linking, because automation cannot tell the difference and does not try. Authority behaves like current: pages carrying traffic carry it, and a page carrying none transfers none. 18:13

Read organic traffic as proof the page is not penalised

If the engine is sending organic traffic to a page, that page is not penalised. The thing people spend weeks worrying about is directly observable and the observation costs nothing. It is a self-correcting read, and it replaces a great deal of speculation with a number you already have. 27:26

Spread the authority rather than concentrating it

Each outbound link loses most of what it carries, so the arithmetic decides the structure. A page with many links out is giving away very little on each one. Spreading beats concentrating — and it is a calculation rather than a preference, which is what makes it settleable. 29:03

Bridge laterally through an intermediate topic

A site moves sideways as well as upward by bridging through a topic that touches both. Each bridge is an adjacency rather than a leap, which is how a narrow start becomes a broad presence without ever competing head-on for the terms in the middle. Relevancy is a gate rather than a limit, so starting narrow does not box you in. 36:25

Keep the risky work on a separate keyword domain

A second keyword domain firewalls programmatic or experimental work away from a valuable brand domain. It is insurance rather than strategy and it costs the price of a domain. The same logic explains why a blog on a subdomain is a mistake: it is a separate domain, and it shares no topical authority with the site you actually care about. 1:01:14

Target brand-new human-coined keywords the system has not seen

The system's weak spot is terms that did not exist until somebody coined them. Nothing has accumulated against them, no incumbent holds history there, and the long-tail climb starts from zero for everyone. It is the only category of position that is genuinely uncontested rather than merely under-contested. 34:38

Questions

Each answer ends at the moment in the recording where it is given.

Can a website keep ranking if it scales content but its backlinks are not growing?

Yes. There is no fixed ratio binding the amount of content to the number of links, so a site can keep publishing and keep ranking while its link profile stands still. Pages that rank and get clicked generate their own authority, which is what carries the next tier of pages. The limit is not the ratio; it is the wall you eventually hit against entrenched competitors, and until you meet it you can keep growing, especially if you publish on a dependable rhythm. 00:58

How does a site build authority without buying or building links?

By ranking and getting clicked. Authority is treated as a flow carried by pages that have traffic, not a stock of links, so a page that earns its position feeds standing to the pages it links to internally. The practical rule is to link a new page from one that is already ranking and already getting visitors, rather than from one that merely holds a backlink. Growing without links at all is possible but hard, and reserved for those who truly master shaping authority — a slower cousin of earning trust rather than buying it. 16:48

Why do some SEOs refuse to buy backlinks?

Because real relationships produce links as a by-product. Years inside a cluster of related, non-competing businesses mean clients introduce peers, contributions get written between them, and partnerships create genuine cross-links no purchase can imitate. The engine reads real businesses transacting as exactly what they are. The takeaway is that a real operation doing real work needs far less deliberate link-building than the field assumes, because the links arrive on their own. 23:29

How can I tell if a backlink is actually worth anything?

Check whether the engine sends organic traffic to the page the link sits on. The system is self-correcting: if traffic is flowing to a page, that page is not penalised, whatever it looks like, and if no traffic reaches it, the link is inert no matter how convincing or on-topic the site appears. Ignore design cues and relevance-matching; many polished placements are dead link farms. Traffic, not appearance, is the only universal test — the same instinct behind how audiences actually decide who to believe. 25:51

Will targeting one narrow niche box my new site in if I want to pivot later?

No. Relevancy is a gate that decides eligibility, not a cage and not a form of authority in itself, so you can pivot. The catch is that a new segment starts at the bottom and has to be grown; its first pages will not rank on the strength of your old ones. The way to move between distant topics is to bridge through an intermediate one rather than jumping, so each step stays close enough for the engine to follow. 33:47

Is there a sitewide quality score, and can bad pages drag down the whole site?

Yes, above a modest authority threshold. A site-level trust score sits over page-level ranking, and repeated poor engagement across pages erodes it. After core updates the erosion can cascade — one page falls, then the next, until the site cannot rank even the easy terms it once held. It is a real risk of machine-scaled content that misses intent, and it works like a halo running in reverse, where a few weak parts colour the whole. 39:17

Should near-duplicate product keywords get separate pages or one page?

Try one page first. Put the close variants together and see if the single page ranks. Whether the engine forks the searches apart or merges them as semantically the same is set by user click behaviour, which you cannot control, so building a page per variant against a merge only cannibalises them. Duplicate content is not penalised — a large share of the whole index is duplicate — so being specific is fine; build separate pages only when forced, and expect to test and react. 44:25

Should a blog or forum go on a subdomain or a subfolder?

A subfolder. A subdomain is treated as a separate domain that does not share topical authority with its parent, because it could be handed to another owner, so splitting a blog off multiplies the effort for no gain. The only real reason to split is different technology, and even then the fix is to rewrite it back to a subfolder. A forum is the useful exception: as user-generated content it scales indexable pages that link up to the parent, without the machine-scale risk. 50:36

When should a company use a keyword-in-domain name?

When it lowers the effort-to-return ratio, which is increasingly often. It need not be an exact match; a partial keyword match works. A second keyword domain revives de-grounded listicles, spreads ranking, and firewalls risky experiments off a valuable brand — worst case it loses and costs little, best case two domains reinforce each other. It can also earn a top paid-search quality score for the price of a cheap domain, and because language models quote entities, a branded keyword domain reads as the brand — a real piece of leverage from something you own. 1:00:03

Is SEO still worth learning in the age of AI?

The argument made is yes, and tentatively that interest is at an all-time high. Language models reproduce the errors in their inputs and chase the best answer rather than the truth, so judgement and experiment remain the edge. Meanwhile the rival channels — email, the major social platforms — have grown hard, costly or ineffective, leaving few global channels as easy. Every new site owner meets the same puzzle, that it does not show for its own brand, which is the discipline by another name, now extended to showing up inside an assistant. 1:07:15

Sources

Can You Rank on Google Without Backlinks? Legendary SEO David Quaid Answers

The sections above follow the episode's own order. Each timestamp is the point in the recording where the idea is discussed, so the post reads as a map of the conversation rather than a rearrangement of it.

Quotations come from the episode transcript with spoken filler ("okay", "right", "you know", "um", "actually", "kind of", "sort of") and false starts removed. No words are added: square brackets mark anything inserted for sense, and an ellipsis marks any cut. British spellings are restored, and speech-to-text errors were repaired against context only where the intended word is unambiguous. Where a line appears both in a cold-open teaser and in its place in the conversation, the timestamp cites the second — the point at which it is actually said.

Quotes are stamped with the moment they are said rather than with a speaker's name.