The link building course

About this course

This is a nine-module, fifty-lesson course on link building, written to be worked through in order. Lesson one assumes you know nothing beyond what a hyperlink is. The last assumes you can run a program, brief a team and defend the results to whoever controls the budget. Everything in between is the actual job: judging whether a link is worth having, reading the link graph, finding prospects, writing outreach that gets answered, running digital PR, choosing tactics that suit your site, and recognizing risk before it becomes a problem.

Who this is for

Four kinds of reader, in roughly the order I had them in mind while writing:

• Somebody who owns a site and has been told they need links. You do not work in search, you have no vocabulary for this, and every article you have found so far was written to sell you something. Start at lesson 1.1 and read forward.
• An in-house marketer who inherited SEO. You can run a campaign, you understand a funnel, and links are the part nobody handed over properly. Modules 1 through 3 will be quick; modules 4 through 8 are the ones you were missing.
• A freelancer or agency practitioner who wants a defensible method. You already build links. What you may not have is a written rubric for quality, a risk position you can state out loud to a client, and a report that survives being questioned by somebody senior. Modules 2, 7 and 8 are written for that.
• An experienced SEO who came for the data. The evidence base is an analysis of my own profile — 1,301,839 links from 22,260 referring domains, none of them bought — and it says several things the industry repeats confidently that are not true. Module 3 and module 8 carry most of it.

Who it is not for

I would rather you leave now than read forty lessons and find out at the end.

• Anyone who wants a list of sites that will link to them. There is no such list here and there is not going to be. A published list of link opportunities stops working the week it is published, because everybody who reads it sends the same email to the same person.
• Anyone who wants to buy links efficiently and be told it is fine. The tactic reference grades nine of the thirty-four tactics as high risk or obsolete, and several of those are sold commercially by real companies with good websites. If you want reassurance rather than an assessment, you will not enjoy this.
• Anyone who needs a result this month. Nothing described here produces links in a week. The fastest tactics in the reference are measured in weeks to a first link, and most are measured in months.
• Anyone looking for tool tutorials. This course teaches you what to ask a tool for and how to read the answer. It does not walk you through an interface that will be redesigned before the screenshot ages.

What you need before you start

Nothing. That is not a marketing line, it is a design constraint I held myself to.

You need no paid tool subscription, no budget, no team, no existing traffic and no technical background. Where a lesson uses a tool, it names what the tool is doing so you can follow the reasoning without owning it, and the manual version of the method is described alongside the automated one. There is no math beyond percentages. There is no code. You do not need a website to understand the material, although you will get considerably more out of it if you have one to apply it to, because most of the checklist items are small pieces of work on a real site.

How long it takes

Two answers, because there are two questions hiding in that one.

Reading it is a week of evenings. Most lessons run somewhere between twelve hundred and fifteen hundred words, which is well under ten minutes each; a module is an hour or so; the fifty lessons together are a long weekend if you are determined, or two weeks at a lesson a day if you are sensible. The tactic profiles, guides and glossary are reference — you are not meant to read those front to back, and I would think slightly less of you if you did.

Doing it is a quarter. The checklists are the honest measure: nine of them, one per module, each a short list of things to do to your own site. Working them properly means running an audit, building a prospect list, sending real outreach and waiting for real replies, and that timeline is set by other people's inboxes rather than by your reading speed. Plan for a quarter to work through it with a live site, and expect the first links from module 5 or 6 work to arrive somewhere near the end of it.

The order the modules are in, and why

The sequence is not arbitrary and it is not the order these subjects usually get taught in. Each module exists because the one after it is unteachable without it.

• Module 1, foundations. What a link does, what PageRank was, what Penguin changed, and the vocabulary. Everything downstream is unintelligible without this, and most bad link building decisions trace back to a wrong model of what a link is for.
• Module 2, judging link quality. Before you can acquire anything you have to be able to tell whether it is worth acquiring. Relevance first, then authority metrics and their limits, then whether the page is actually alive. This module is where the scoring rubric gets built.
• Module 3, reading the link graph. Quality judgment applied at scale, and one step further out: tiers, Trust Flow and Topical Trust Flow, network footprints, co-citation. You need module 2 to read a single site and module 3 to read a thousand.
• Module 4, prospecting. Now that you can recognize a good link, go and find the sites that might give you one. Target sets, competitor gap analysis, search operators, finding the actual human, and scoring what you find.
• Module 5, outreach. Prospecting produces a list; outreach turns the list into replies. It comes after prospecting because the most common cause of failed outreach is a bad list rather than bad writing.
• Module 6, digital PR and earned media. Outreach asks one person for one link. Digital PR builds something a newsroom wants and lets the links follow. It comes after module 5 because pitching a journalist badly is more expensive than pitching a blogger badly.
• Module 7, choosing and running tactics. Only now, with quality judgment, prospecting and two acquisition disciplines in hand, does tactic selection become a real decision rather than a guess. This module builds the ninety-day plan and the workflow around it.
• Module 8, measurement, risk and recovery. What to measure, how to report it so it survives scrutiny, how to recognize an unnatural profile, and what to do about manual actions and disavow. It sits at the end because you cannot audit a program you have not learned to run.
• Module 9, the future of link building. Mentions, entities, AI search and what happens to the value of a followed link. Last, deliberately, for the reason below.

Modules 1 to 8 describe the job. Module 9 argues.

I want to be straight about the difference, because it changes how you should read it.

Modules 1 through 8 describe link building as it works today, and almost none of it is contested. The mechanisms are documented, the failure modes are observable, and where the industry genuinely disagrees I say so on the page. If a search engineer read those eight modules they might quibble with emphasis; they would not find much to call wrong.

Module 9 is a position, not a description. It argues that the useful unit is drifting from the link toward the citation — that unlinked mentions, entity recognition and how AI systems select sources are changing what a link is worth and what counts as a result. I believe that, I have reasons, and I have written them out so you can check them. But it is an argument about where things are going, made in advance of the evidence that would settle it, and the honest label for that is informed opinion rather than instruction. Read modules 1 through 8 as the job. Read module 9 as a colleague telling you what he thinks is coming and showing his working.

It is free, and free means free

There is no signup. There is no email gate on module three, no free trial that expires, no certificate to buy and no upsell at the end. You can read all fifty lessons, all thirty-four tactic profiles, every guide and the whole glossary without telling me who you are. I did not build a lead magnet and then dress it up as education.

How to use it

Read it in sequence the first time. The modules build: you cannot judge link quality sensibly until you understand what links do, and you cannot choose a tactic sensibly until you can judge quality. Each lesson opens with what you should be able to do when you finish it, and ends with a link to the next one. Where a lesson teaches a method that a specific tactic depends on, it links to that tactic's profile so you can see the method applied end to end. If you already work in search and want only the parts you do not have, the syllabus below is the index.

There is no progress tracking, and I am not going to pretend there is

This site is static HTML. There are no accounts, no logins, no database and no cookies set by the site. That is a deliberate design decision and it has a consequence I would rather state than hide: nothing here remembers where you got to. There is no progress bar, no completion percentage and no green tick next to lesson 4.2. Building one would mean identifying you, storing that identity somewhere, and asking you to sign in — which would make every claim on this page about signup and email gates untrue.

So the course uses three mechanisms instead, all of which work without knowing anything about you:

• The previous and next chain. Every lesson links to the one before it and the one after it, in order, all the way from lesson 1.1 to the end of module 9. Your browser history is the progress bar.
• The printable syllabus. The full list of modules and lessons is on this page. Print it, or keep the tab open, and mark it as you go.
• The downloadable checklists. One per module, in Markdown and plain text, each with the module's lesson URLs at the bottom. They are the practical record of what you have done and what you still owe your own site.

A record that lives in your own notes outlasts one that lives in my database anyway. I have watched people lose years of progress when a platform shut down.

What is deliberately not covered

This is a course about links. It is not a course about SEO, and pretending otherwise would have made it worse at the thing it is for. Not covered, on purpose:

• Technical SEO. Crawling, indexation, site speed, structured data and migrations appear only where they touch links directly — a migration that drops redirects loses link equity, so that gets covered; core web vitals do not.
• Keyword research and content strategy. Both are prerequisites for a sensible link program and both are large enough to deserve their own course. Where a tactic needs a linkable asset, the course tells you what makes an asset linkable, not how to plan a content calendar.
• On-page optimization and conversion. A link sends people to a page. What that page does with them is somebody else's chapter.
• Paid media and social media management. Digital PR touches both at the edges. Neither is taught.
• Local SEO as a discipline. Local link building has a tactic profile of its own. Listings management, review generation and map pack mechanics do not appear.
• Tool walkthroughs and vendor comparisons. The tools guide explains what each category of tool is for and what its metrics can and cannot tell you. It does not rank products, and no product paid to be in it.

What it will not do

It will not make link building fast. It will not give you a list of sites that will link to you, because such a list is worth nothing the moment it is published. And it will not tell you every tactic is fine, because several of them are not. The tactic reference grades all thirty-four by risk, and nine of those grades are unflattering to things that are actively sold.