How James Dooley Starts a New SEO Site (650 Industries Deep)

Edward Sturm

Abstract

Twenty minutes with an operator running sites across more than six hundred industries, answering the most common question in the field — how do I start a new site — by refusing the question as posed. What kind of site it is and how hard the market is decide everything downstream, and without both the answer is a guess wearing confidence.

• In a hard niche an exact-match domain still works, aimed at the long tails rather than the head terms.
• New markets are scouted by comparing search volume against cost-per-click before any commitment is made.
• Most aged domains are bought wrong, because the headline authority score is a vendor's product rather than the engine's opinion.
• The gambling churn-and-burn model is described in full, including why he left it: takedown notices within weeks.
• Pages rank on two or three sentences where the intent is matched, and fifteen years of results contradict the people predicting penalties.
• A brandable business starts with a stack of twenty to fifty citations, then an award applied for and won.

Everything below follows the conversation in order, from the refused question through market scouting and domain buying, into the two contrasting business models, and out to how a brandable site is actually assembled. For anyone building a personal brand the transferable part is the last third: the award is manufactured in the honest sense — the application is real, the award is real, and almost nobody bothers.

Chapter summaries

01:01 - The site type and the market's difficulty come before any answer about starting

You cannot answer "how do I start" without context — site type (local lead-gen, affiliate) and how competitive the niche is; the plan branches on that.

"I need a little bit more context around the new website."

— 01:20

The question — how do I start a new site — is the most common one asked, and it cannot be answered as posed. What kind of site (local lead generation, affiliate) and how hard the market is decide everything downstream, and without both the answer is a guess wearing confidence.

01:53 - A hard niche still allows an exact-match domain aimed at the long tails

In a hard niche you can still use an exact-match domain if you go after the long tails on a micro-site; how far depends on how big you want to go.

"How come you wouldn't do an EMD in finance?"

— 01:53

In a hard niche an exact-match domain is still usable if it is aimed at the long tails rather than the head terms. The domain is not doing the work the head term would need; it is marking a narrow position precisely.

02:10 - The narrowest, highest-intent slice converts at the highest rate

Target the narrowest, highest-intent slice: a very specific buyer situation converts at high commission even on a tiny site.

"So prime example, let's say in the debt market, we might go."

— 02:10

The narrowest, highest-intent slice converts at the highest rate. A very specific buyer situation named in the domain and the pages is worth more than a broad category with more searches in it. The narrow claim is the strong one.

02:35 - Competing with the established players is a different job that takes time

Going after the competitive players is different: they all hold very high domain-authority sites, so you must build topical authority and it takes time.

"You're going to try and go after the insolvency practitioners within the."

— 02:35

Competing with the established players is a different job. They hold very high authority scores, and topical authority takes time rather than budget to accumulate — which is why the long-tail route exists at all. The operator is active across more than six hundred industries, having freed his own time by inventing a role and handing the day-to-day work on.

03:47 - New markets are scouted by comparing search volume against cost-per-click

New markets are found by scouting demand against cost: comparing search volume versus cost-per-click across directories and keyword tools to spot industries with enough demand and commercial value to enter.

"So we've got like an ideation team who might be going and."

— 03:47

New markets are scouted by comparing search volume against cost-per-click: high demand with low competitive spend marks an industry nobody has worked properly. It is a two-column check that can be run before any commitment.

04:15 - The industry list grew by referral into adjacent niches rather than by plan

The industry list was never planned — it grew by referral from one niche into adjacent ones, all of them the same lead-generation service.

"Also, like one we never started to."

— 04:15

It grew by referral from one niche into adjacent ones, which is a description of how a portfolio actually forms rather than a strategy anyone could execute from a standing start.

05:34 - The playbook is rinse-and-repeat, tuned to how hard the market is

The playbook is rinse-and-repeat, tuned to market difficulty; for hard finance markets, buy an aged domain that already carries many referring domains and trust.

"We've got a rinse and repeat strategy here, depending on how."

— 05:34

The playbook is rinse-and-repeat, tuned to how hard the market is — the hard finance markets get aged domains, the easier ones get built fresh. The difficulty is the input that selects the method rather than something to be overcome by effort.

05:52 - Buying a site that ranks but is not monetised is the other route

You can also buy a site that already ranks and gets traffic but whose owner cannot monetise it, then take it over — improve conversion, republish unindexed content, fill topical gaps, strengthen categories with backlinks.

"There's certain places still, and places like that, where."

— 05:52

The other route is buying a site that already ranks and gets traffic but whose owner cannot monetise it. The asset and the ability to use it are separable, and the gap between them is where the price sits.

06:23 - Dozens of buyer-intent pages is the alternative to answering questions

Most SEO puts up articles answering how/what/when questions; an alternative puts up dozens of short pages that sell to searchers already looking to buy, and those rank and convert far better with less work.

"This method of marketing is so effective I had to make."

— 06:23

Most optimisation puts up articles answering how, what and when. The alternative puts up dozens of buyer-intent pages instead — fewer questions answered, more situations named.

07:47 - Most aged domains are bought wrong, because the headline score is the wrong signal

Most people buy aged domains wrong because they are obsessed with the headline authority score — the same mistake as buying backlinks; sellers manipulate that number so it looks great on a backlink tool while the site is actually terrible.

"So, that's a brilliant question to ask, because."

— 07:47

Most people buy aged domains wrong because they fixate on the headline authority score. The vendor's number is a product rather than the engine's opinion, which is the single most expensive misunderstanding in this part of the field.

08:20 - The real checks are trust signals rather than a number

The real checks are trust signals: a local business profile (which implies a place identity), any branded search, and a clean backlink profile — no mass exact-match anchors, no foreign-language or adult anchor text.

"Want trust. So we're looking at certain things, like does it."

— 08:20

The real checks are trust signals: a local business profile implying a place identity, and the other marks of something that was once a real operation. Trust and score are different properties, and only one of them survives inspection.

09:05 - Too many backlinks to review by hand means triangulating across tools

With too many backlinks to review by hand, the answer is "it depends" and you triangulate: pull every tool's data — trust and citation scores, authority and spam scores, toxicity — and combine it.

"So that's a good question overall, cuz obviously it."

— 09:05

With too many backlinks to review by hand the answer is that it depends, and the method is triangulation — pull every tool's estimate and combine them, because no single vendor's number is reliable on its own.

09:46 - The anchor-text percentages have to read as a natural profile

For the links themselves, read the anchor-text percentages: ideally the top three or four anchors are branded anchors and naked URLs; a profile full of random anchors is a reason to walk away.

"And we're going to get all the information, we'll put it all."

— 09:46

For the links themselves, read the anchor-text percentages: the top few anchors should look like a natural profile rather than a targeted one. A clean profile is legible at a glance once you know what shape to expect.

10:11 - A sabotaged marketplace listing is a tactic worth recognising

A shady tactic exists: spam a marketplace listing with junk links so the seller panics, then negotiate the price far down; the spammer can spoof or remove the links afterward.

"And this is a little shady technique that some people are doing."

— 10:11

A shady tactic exists and is worth recognising: spam a marketplace listing with junk links so the seller panics and sells cheap. The genuinely premium aged domains — high trust, an encyclopedia link, major-news links — are priced accordingly and do not come up that way.

11:35 - Gambling operators buy these domains, convert them and burn them

In the gambling space operators buy these domains up fast, convert them to casino sites, and can earn back the cost almost immediately; that demand has pushed prices up roughly tenfold.

"I mean, in the iGaming space people are buying them up left."

— 11:35

In the gambling space these domains get bought up fast, converted to casino sites and burned. Rankings arrive in about forty-eight hours and last roughly six to eight weeks, which is the whole economic model stated as a duration.

12:37 - The mechanic is a spoofed anchor profile on any aged domain

The mechanic: buy any aged domain regardless of its old niche, add thousands of blog-roll homepage links to restructure the anchor ratios toward the new gambling terms, and the old domain's trust carries it into the rankings.

"It's a very churn and burn market, cuz they're normally doing it."

— 12:37

The mechanic is a spoofed anchor profile: buy any aged domain regardless of its previous niche, add thousands of blogroll links, and restructure what the profile appears to be about. It works because the signal being read is cheap to fake at volume.

13:22 - Takedown notices arrive within weeks on the big terms

The reason to leave that market: within a couple of weeks on big terms you get takedown notices, negative SEO and constant attacks — that blowback is why they are exiting gambling.

"It to come in and rank. The issue is then normally within."

— 13:21

The reason to leave that market is what arrives next — within a couple of weeks on the big terms, takedown notices and the rest of the blowback. The exit is described as a business decision rather than an ethical one, which is more useful than a sermon.

13:44 - A local exact-match site can be built in a week

By contrast, a local exact-match site can be built within a week with good content — maybe only 50 pages, the main service across 50 local locations; an existing business profile makes it stronger.

"Yeah. But in a local, you."

— 13:44

By contrast a local exact-match site can be built inside a week with good content. The comparison is doing work: the same operator describes both, and the second is slower to start and does not end in takedown notices.

14:13 - The compounding runs by reinvesting into a second and third site

Then you compound: grow the site to a few hundred a month, reinvest into second and third sites and micro-sites, and in time build bigger, more brandable sites covering multiple services.

"And yeah, you can then start getting that to earn, let's say."

— 14:13

Then it compounds: grow the site to a few hundred a month, reinvest into a second and third, and repeat. The compounding is the strategy rather than any individual site.

14:55 - Pages rank on two or three sentences where the intent is matched

"Good quality content" is contested, but pages rank with only two or three sentences as long as they meet the exact intent of the H1 question.

"So there is obviously a lot of debate within."

— 14:55

Pages rank on two or three sentences where the intent is matched. Good quality content is contested territory, and this is the sharpest version of the contest — the page answers the situation and stops.

15:13 - Fifteen years of results contradict the people predicting penalties

The people who call that thin content and predict penalties are contradicted by 15 years of it working; judge by what actually happens, not by what is asserted.

"Now, some people would turn around."

— 15:13

The people who call that thin content and predict penalties are contradicted by fifteen years of results. It is evidence against dogma, offered from a position of having run the test repeatedly rather than having reasoned about it.

15:36 - The method is to work out what the searcher in that situation wants

The method is to think of the user: work out what a searcher in that situation actually wants to know (price, speed, services) and answer those questions concisely, not playing hide-and-seek with the answer.

"No, I wouldn't. In an ideal world, if someone wants."

— 15:36

The method is to think of the user: work out what a searcher in that specific situation actually wants and give them that. Much of the industry used to write five to seven thousand words and bury the answer, which is the failure this replaces.

16:27 - Content should be as short as possible and as long as necessary

The maxim: content should be "as short as possible, but as long as necessary" — cut the fluff, be concise, give the answer, but cover all the different answers a searcher might want.

"Mental, mental. So I'd say it needs to be."

— 16:27

Content should be as short as possible but as long as necessary. The maxim does the work a word count cannot, because it sets the target as coverage of the intent rather than a number to reach.

17:03 - A brandable business is built on a stack of off-page citations

A brandable business is built on off-page signals: a stack of citations (20-50 main business listings), presence on review platforms, a business profile carrying reviews, and case studies.

"So, brandable business, I would then start going down the road."

— 17:03

A brandable business is built on off-page signals — a stack of twenty to fifty main business citations before the pages are worried about. The off-site layer is the foundation rather than the finishing touch.

17:33 - An award is applied for, won, and then claimed everywhere

Make it an award-winning brand: find the awards in your industry, apply to win them, then claim "award-winning" and broadcast it across every social channel — a "branded social fortress".

"I'm going to try and make it an award-winning brand."

— 17:33

An award is a hard-to-dispute credential that can be manufactured demand around and then repeated everywhere — applied for, won, and shared across every social and Web 2.0 property the brand holds. The manufacture is honest in the only sense that matters: the application is real, the award is real, and almost nobody bothers to apply.

18:00 - One award becomes photos, a video and images, syndicated everywhere

AI makes the distribution trivial: turn one award into photos, a video and images, then syndicate it — one award plus a press release can become 50-plus articles saying you are award-winning, building reputation and firsthand-experience signals.

"And now with AI, especially if you've got any autonomous AI agents."

— 18:00

Distribution of that award is trivial now: one award becomes photos, a video and images, syndicated across every surface. The close flags a challenge to the experience-and-expertise framework, consistent with everything else in the episode — what can be claimed cheaply cannot be load-bearing.

Personal Branding Lessons

An operator describing what he actually does across several hundred markets. The moves below are mostly about deciding what kind of problem you have before deciding what to do about it.

Establish the site type and the market's difficulty before choosing a tactic

How do I start is the most common question in the field and it cannot be answered as asked. Local lead generation and affiliate are different businesses; a hard market and an easy one call for opposite methods. Both variables come first, because every tactic downstream is selected by them. An answer given without either is a guess delivered confidently. 01:01

Target the narrowest, highest-intent slice you can name

A very specific buyer situation converts at a far higher rate than the broad category it sits inside, even though the category has more searches in it. The narrow position is what makes an exact-match domain workable in a hard niche — it is not competing for the head term, it is naming a situation precisely enough that nobody else has bothered. 02:10

Scout new markets by volume against cost-per-click

High demand with low competitive spend marks an industry nobody has worked properly. It is a two-column check, it runs before any commitment, and it answers the question most people answer by instinct. The comparison is the whole method — neither number tells you anything useful on its own. 03:47

Buy an aged domain on trust signals rather than on the headline score

Most people buy these wrong because they fixate on the vendor's authority number, which is a product rather than the engine's opinion. The real checks are trust signals: a local business profile implying a place identity, and the other marks of something that was once a real operation. Trust and score are different properties and only one survives inspection. 07:47

Read the anchor-text percentages before buying anything

The top few anchors should read as a natural profile rather than a targeted one. Where there are too many links to review by hand the answer is that it depends — pull every tool's estimate and triangulate, because no single vendor's number is reliable alone. A clean profile is legible at a glance once you know the shape to expect. 09:46

Write to the searcher's situation rather than to a word count

Work out what somebody in that specific situation actually wants and give them exactly that. Much of the industry used to write five to seven thousand words and bury the answer, which is the failure this replaces. Pages rank on two or three sentences where the intent is matched — as short as possible, as long as necessary. 15:36

Build the off-page citation stack before worrying about the pages

A brandable business rests on off-site signals: a stack of twenty to fifty main business citations, assembled before the on-page work is fretted over. It is the least interesting part of the job and the foundation the rest sits on, which is a fair description of why it gets skipped. 17:03

Apply for the industry awards and then claim them everywhere

Find the awards in your field, apply, win one, and claim it across every surface you own. The credential is manufactured in the honest sense — the application is real and so is the award — and almost nobody bothers to do it. Distribution costs nothing now: one award becomes photos, a video and images, syndicated everywhere. 17:33

Questions

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

How should I start a brand-new website if I want it to rank?

You cannot choose a tactic until two things are fixed: what kind of site it is, and how competitive the niche is. Everything else branches on that. An easy market can take a fresh, narrow site aimed at a specific search; a hard one needs inherited power before a fresh domain can rank at all. Decide the difficulty first, then pick the method to match it, rather than reaching for a favourite tactic and hoping the market suits it. 01:20

When should I use a fresh domain versus buying an aged one?

Match it to difficulty. In an easy niche a fresh exact-match micro-site aimed at a narrow long-tail search can rank, because specific intent means little competition. In a hard niche a new domain would take months to build topical authority, so the shortcut is an aged domain that already carries referring domains and trust, or a site that already ranks but is not being monetised. Timing your entry against the field is the deciding factor here — the same first-mover versus fast-follower call. 01:53

How do I find a niche worth entering?

Weigh two numbers together: how much a term is searched, and how much advertisers pay for its clicks. Look for spaces you are not yet in where both numbers clear the bar, then treat entry as a judgement call rather than a certainty. Running candidate ideas through a validation scorecard keeps the demand-versus-value read honest. 03:47

What should I actually check before buying an aged domain?

Not the headline authority score — that number is routinely inflated by sellers, so it certifies exactly the manipulated sites you want to avoid. Check for trust that is hard to fake instead: a genuine business listing that implies a real place, some branded search, and a clean backlink profile with no mass of exact-match, foreign-language or adult anchor text. Trust signals a manipulator cannot easily manufacture are the real inspection list, and the score is the trap. 07:47

How do I judge a backlink profile when there are too many links to read?

You do not read them all. Pull every tool's estimate — trust and citation scores, authority and spam scores, toxicity — and put them back together. Then read the anchor-text distribution: ideally the top three or four anchors are branded anchors and naked URLs. A profile full of random keyword anchors signals manipulation and is a reason to walk away from the buy. 09:05

How fast can a local SEO site start making money?

A local exact-match site can be built within a week with modest content — perhaps only fifty pages, the main service repeated across fifty local locations — and an existing business listing on the domain makes it stronger. From there the model is to earn a few hundred a month and reinvest it into second and third sites, then into bigger, more brandable ones. The point of the fast build is that its small recurring income funds the next asset without outside capital. 13:44

What does "good quality content" really mean for SEO?

The working claim is that a page ranks when it meets the exact intent of its H1 question, even in two or three sentences — length is not the lever, intent match is. Pages like that are described as ranking for fifteen years without the penalty their critics predict, so the test is what actually happens, not what the received rules assert. The rule of thumb is to be as short as possible but as long as necessary: answer the real question, cover its follow-ups, and cut the rest. 14:55

How do I turn a small site into a brandable business?

Through off-page reputation built on other people's platforms: a stack of twenty to fifty citations across the main business listings, presence on review platforms, a business listing carrying reviews, and case studies. The strongest single move is a manufactured credential — find the awards in your industry, apply to win them, then broadcast "award-winning" across every channel. An award is a hard-to-fake signal you can multiply with automated content, turning one real event into dozens of syndicated mentions and a halo that colours everything else. 17:03

Sources

How James Dooley Starts a New SEO Site (650 Industries Deep)

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.