SEO for founders, in plain English

You do not need to become an SEO specialist

You need to understand enough to make good decisions and to know when someone is selling you nonsense. That is a much smaller body of knowledge than the industry would have you believe.

Most of SEO is a handful of fundamentals applied consistently. The rest is folklore, half-remembered rules from 2012, and tactics that stopped working before your company existed. This guide separates the two. By the end you will know what actually moves the needle, what to safely ignore, and why all of it matters more now that AI answers are part of search.

Why SEO still decides whether AI can find you

The instinct, when AI Overviews and ChatGPT started answering questions directly, was to assume the old rules were dead and a new discipline had taken over. They are not, and it has not.

Google's AI features build their answers from the same search index that powers the blue links. They use Retrieval-Augmented Generation over that index and a query fan-out technique that fires off several related searches at once, then assemble an answer from what they retrieve. The plain-English version: if your page is not indexed and legible to ordinary search, it is not a candidate for the AI answer either. Google has said directly that optimizing for its AI features is simply good SEO, not a separate craft.

The wider ecosystem (ChatGPT, Perplexity, Claude, and the growing pile of AI agents) behaves a little differently, and some emerging conventions matter there. We cover that whole picture in getting found by AI search. But the foundation is identical: a site that is crawlable, indexable, and genuinely useful. Everything below is that foundation.

The pipeline: crawl, index, serve

Three steps stand between you and a search result. Google explains them in its SEO starter guide and get-started documentation, and understanding them in order solves most of the confusion founders have about why a page is not showing up.

Crawl. Googlebot, an automated program, follows links and fetches pages. It mostly uses a mobile crawler (Google is largely mobile-first now) and reads roughly the first 2MB of an HTML page. If a page is blocked, broken, or buried with no links pointing to it, Googlebot may never see it. The how-and-why of this stage is worth a read on its own: see how Google crawls, indexes, and ranks and the crawlability glossary entry.

Index. Once fetched, Google decides whether to store the page in its index. Not everything that gets crawled gets indexed. You can check whether a given page made it in by typing site:yourdomain.com/the-page into Google. If it comes back empty, the page is not in the index, and no amount of optimizing will rank it until that is fixed. The indexing entry explains what blocks pages from this stage.

Serve. When someone searches, Google ranks the indexed pages and shows the most relevant ones, increasingly alongside an AI-generated answer. This is the stage everyone obsesses over, but the first two stages are where most real problems live.

If you take one thing from this section: a page must be crawled and indexed before ranking is even a question. Founders waste weeks tuning content for pages that Google has never managed to store.

The handful of things that actually matter

Let Google see the same page your visitors see. Googlebot needs access to your CSS and JavaScript, not a stripped-down version. If your site renders content only after heavy client-side JavaScript, parts of it can go missing from the index. This is the single most common technical failure on modern sites, and it is worth understanding before you ship anything framework-heavy.

Write clear, unique titles and meta descriptions. Each page should have one title that is concise and specific to that page. Descriptions should be short and unique; Google often writes its own snippet from your page content anyway, so the goal is accurate signposting, not keyword cramming. We go deep on this in title tags and meta descriptions.

Use descriptive URLs and link text. A URL like /guides/seo-for-founders tells both people and crawlers what the page is. Anchor text should describe the destination ("read our pricing breakdown"), never "click here."

Make content for people, with clear authorship. Google's helpful-content thinking comes down to who made it, how, and why. Be transparent about your process, including any AI involvement, and write to genuinely help the reader rather than to game a ranking. This connects to E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness), where trust is the thing that matters most.

Be fast and stable. Slow, janky pages get deprioritized in both classic search and AI answers. Page speed is a shared signal, and you do not need an engineering degree to address the common causes.

Get the technical basics right from day one. If you are launching a new site, a short, ordered checklist beats guesswork. Our technical SEO checklist for a new site is built for exactly that moment.

What to stop worrying about

A surprising amount of SEO advice is about things that do not work or never mattered. Stop spending energy on these:

• The keywords meta tag. Google does not use it. It has not for many years. Delete it from your mental model.
• A magic word count. There is no minimum length that makes a page rank. Some queries want 200 words, some want 2,000. Write what the topic needs and stop.
• Heading order as a ranking trick. The order of your headings (h1 then h2 then h3) matters for accessibility and screen readers, which is reason enough to get it right. It does not affect ranking. You will not be penalized for an out-of-order heading.
• Duplicate-content "penalties." There is no penalty for duplicate content. It is just inefficient, since Google has to pick one version to show. Use a rel=canonical tag to point at your preferred version and move on.
• Sitemaps as a requirement. They help large or new sites get discovered, but they are optional, not mandatory.

For Google specifically, also skip the AI-tailoring tactics being sold right now: you do not need an llms.txt file for Google, you should not chop your content into tiny fragments or rewrite it specifically for AI, and you should not treat structured data as a hard requirement or spin up a separate page for every query variation. That last one is "scaled content abuse," and Google treats it as such.

Where to go from here

This pillar is the map. The articles it links to are the territory: the mechanics of how Google crawls, indexes, and ranks, the JavaScript-rendering trap that hides content from the index, how to write titles and descriptions that earn the click, what Core Web Vitals mean for a non-engineer, how helpful-content and E-E-A-T actually translate into pages, and the launch checklist for a brand-new site.

Start with the pipeline. Get crawled, get indexed, then earn the ranking with content made for people. Do that, and the AI answers take care of themselves, because they are reading the same well-built site.