James Oliver | Profit Focused SEO

Founder of @SEOTalos it's GSC for for agencies. Co-Founder of @ParasiteSEOcom & @AnswerSocrates. Sharing SEO, SaaS, Affiliate, and CRO strategies

[James Oliver | Profit Focused SEO @oliver__com]

How to Find Keywords That Rank Fast (In 11 Steps)

After dropping that Black Friday guide, everyone’s asking: 'How do YOU find the best keywords?'

Well, here’s the deal: I use a simple, 11-step process that takes me from broad keywords to ranking gold—all with @parasiteseocom

Think:

- Keywords your competition doesn’t even know about.
- Insights on how easy it’ll be to rank.
- A way to dominate the SERPs faster than ever.

And the best part? You can do it all in minutes using the same tools I use—without all the fluff.

Want the exact steps?

[Watch the video] and get the SOP in the comments!

[James Oliver | Profit Focused SEO @oliver__com]

Played around with @Remotion in Claude Code

Made a short demo for https://t.co/Zb5gkpxr0X 👇

Less than 30 mins. Genuinely impressed.

[James Oliver | Profit Focused SEO @oliver__com]

🚨DR 94 backlink🚨

here's how it works:

join Stripe Climate and commit 0.5% of your revenue to carbon removal. in return, Stripe gives you a public commitment page on climate(.)stripe that links to your site.

the catch:

it's not free like a directory link. you're spending 0.5% of revenue. but on low or pre-revenue MRR, 0.5% is pocket change for a link from one of the most trusted domains on the internet.

two ways to think about it:

→ cheap insurance if you're early. tiny revenue = tiny cost = same DR 94 link to start your SEO efforts.
→ brand + trust signal. a Stripe-hosted page vouching for you isn't a spammy directory.

Bonus: it's an actual carbon removal commitment. You get the link AND you're not greenwashing, the contribution is real.

Full list of backlinks below 👇

[James Oliver | Profit Focused SEO @oliver__com]

https://t.co/Zb5gkpxr0X - tactics + directory submissions. it's there. use it.

[James Oliver | Profit Focused SEO @oliver__com]

Free 81 DR backlink:

GitHub

No-follow, yes. But strong foundational no-follow links still diversify your profile and get you crawled fast.

The play:
"Build me a public GitHub README listicle of the [X best Y]. Make it genuinely useful, well-formatted with markdown. Include [yoursite com] as one of the credited sources/entries with a short honest description."

Bonus:
→ Parasite SEO - GitHub ranks so well you can rank content on it directly. Great for affiliate.
→ Cloud stacking - bundle it with other cloud links to pass trust as a group.

Full list of free backlinks 👇

[James Oliver | Profit Focused SEO @oliver__com]

@tibo_maker Backlink

[James Oliver | Profit Focused SEO @oliver__com]

@PaulVolynskiSEO True. You need to ask perplexity for stats and get the sources

[James Oliver | Profit Focused SEO @oliver__com]

"In today's fast-paced digital landscape..."

You've read that opener a thousand times. Google has too.

Here's a partial list of what my Claude Code skill bans from every article it writes:
"delve"
"tapestry"
"navigate the landscape"
"not only... but also"
"as mentioned above"
"in today's ever-evolving world"
"let's dive in"
"it's worth noting"
"without further ado"
"serves as a testament"
tricolons ("faster, smarter, better")
repetitive openers (3+ sentences with the same structure)

427 banned words. 59 banned phrases. Zero em dashes. Total.

That's 1 of 27 gates every article has to pass before it outputs.

The other 26 check reading level (FK grade 7 max), keyword placement, paragraph length, section structure, content score against the live SERP, and whether every stat in the article actually exists in the research data.

453/453 fabricated claims caught in testing. Every invented price. Every "studies show 73%..." the AI pulled from nowhere. Caught and killed.

I built this because I got tired of editing the same AI tells out of every article by hand. Now Claude strips them before I ever see the draft.

It pulls the real top-10 Google results via Exa. Extracts entities, prices, keywords from competitors. Targets 10% above their average word count. Picks the right structure from 6 article types. Writes section-by-section at grade 7 reading level.

$27 one-time. Lifetime updates. 5 minute setup.

What's the worst AI phrase you keep seeing in content? Drop it below.

Link in comments.

[James Oliver | Profit Focused SEO @oliver__com]

here here https://t.co/Ds87HVLbcW

[James Oliver | Profit Focused SEO @oliver__com]

This is important... and most AI content tools miss it.

The writing isn't the hard part. The research is.

Most AI writers start with nothing. No competitor data. No entity targets. No word count benchmarks. Just a prompt and vibes.

Here's what I built instead.

A Claude Code skill that pulls the actual top-10 results for your keyword via Exa. Downloads each article. Extracts headings, prices, products, technologies. Then writes an article designed to beat them.

27 gates before anything ships. 427 banned AI words. 59 banned phrases. Every stat traced to research data (453/453 fabrications caught in testing). FK grade 7 reading level. Zero em dashes.

Drop your own articles into a brand-voice folder and it mirrors your writing style too.

$27. One-time. Lifetime updates.

P.S. It also prioritizes keywords, refreshes existing articles against current SERP, and captures real brand screenshots via Playwright. Not AI images.

Link in comments.

[James Oliver | Profit Focused SEO @oliver__com]

@shiyam_kashfiq it pulls the live serp and content so works with any niche

[James Oliver | Profit Focused SEO @oliver__com]

427+ words that AI writers love to use.

I banned all of them.

Then I banned 59+ phrases too. And — — —. And "not only... but also." And grade 10+ reading levels.

That's one piece of a Claude Code skill I just released. The full pipeline:

→ Pulls real top-10 SERP results (via Exa, not guesswork)
→ Extracts entities, prices, keywords from competitor articles
→ Writes section-by-section at grade 7 reading level
→ 3 scripts gate every article before output (lint, claim verification, content score)

The article literally won't output until all three pass.

Same system I've used on hundreds of ranked articles. Packaged it so you can install it in 5 minutes and just say "write an article about [keyword]."

It handles roundups, vs, guides, alternatives, reviews, and informational pieces. Picks the right structure based on the keyword.

find it here https://t.co/eC3E6zHEZI

[James Oliver | Profit Focused SEO @oliver__com]

Most AI articles rank nowhere because they have zero information gain over what's already on page 1.

Here's the fix:
→ pull the top-10 SERP via Exa (full articles, not snippets)
→ TL;DR block optimised for AI search snippets
→ "What's New in 2026" with dated facts (freshness signal)
→ "How We Tested" E-E-A-T block on roundups
→ BLUF per H2 — every section defends a thesis
→ ban words + Ai phrases

The goal: give Google a reason to rank yours over theirs.

You can find the Claude Skill here: https://t.co/eC3E6zHEZI

[James Oliver | Profit Focused SEO @oliver__com]

here https://t.co/Ds87HVLbcW

[James Oliver | Profit Focused SEO @oliver__com]

@everestchris6 pool

[James Oliver | Profit Focused SEO @oliver__com]

Most AI SEO articles fail because they're just a prompt with extra steps.

I packaged my full production pipeline as a Claude Code skill:

- Pulls real top-10 SERP data
- Extracts entities + keywords from competitors
- 540+ banned AI words, zero em dashes
- 3 scripts gate every article (lint, claim verification, content score)
- Article won't output until all 3 pass

Same system behind hundreds of ranked articles.

https://t.co/eC3E6zHEZI

[James Oliver | Profit Focused SEO @oliver__com]

@DsMatie Love to try it out… still waiting for a reviewer for my app

[James Oliver | Profit Focused SEO @oliver__com]

@mikefutia Chat

[James Oliver | Profit Focused SEO @oliver__com]

@everestchris6 Pool

[James Oliver | Profit Focused SEO @oliver__com]

@Domain_Catch https://t.co/OYiTQ1oleR

[James Oliver | Profit Focused SEO @oliver__com]

Lacking quality matches, Google defaults to PDFs from high-authority domains.

PDF parasite play seems to be bigger since the recent spam update.