Learn SEO like a course, not a rabbit hole

The sitemap tells Google the full list of URLs you want indexed. The one rule that matters: list only canonical URLs — pointing it at filtered or duplicated pages tells Google to index URLs your own pages disclaim.

import { MetadataRoute } from 'next';

export default function sitemap(): MetadataRoute.Sitemap {
  const baseUrl = 'https://your-site.com';
  const now = new Date().toISOString();

  return [
    { url: baseUrl, lastModified: now, changeFrequency: 'daily', priority: 1 },
    { url: `${baseUrl}/about`, lastModified: now, changeFrequency: 'monthly', priority: 0.8 },
    // Map over your posts, tools, or products here.
  ];
}

Tells crawlers what they may fetch and where the sitemap lives. If you do nothing else, point it at the sitemap — Search Console and every auditor checks here first.

import { MetadataRoute } from 'next';

export default function robots(): MetadataRoute.Robots {
  return {
    rules: { userAgent: '*', allow: '/' },
    sitemap: 'https://your-site.com/sitemap.xml',
  };
}

The title and description are what a searcher reads before clicking; the canonical is you telling Google which URL is the real one. All three belong in metadata, not buried in a component.

import type { Metadata } from 'next';

export const metadata: Metadata = {
  title: 'About',
  description: 'Who we are and why this site exists.',
  alternates: { canonical: '/about' },
  openGraph: { url: '/about' },
};

JSON-LD that describes what a page is — an article, an app — so Google can show more than a bare link. The trap is invisible markup: a rating or price you do not actually have. Mark up only what is real, and leave the rest off.

const jsonLd = {
  '@context': 'https://schema.org',
  '@type': 'Article',
  headline: post.title,
  datePublished: post.date,
  author: { '@type': 'Organization', name: 'Your Site' },
};

// In the page body:
<script
  type="application/ld+json"
  dangerouslySetInnerHTML={{ __html: JSON.stringify(jsonLd) }}
/>

The og image turns a bare URL into a preview when shared; the favicon is the only asset that shows in the tab, the SERP, and a bookmark at once. Next generates cards from a per-route file, so new pages get one by existing.

import { ImageResponse } from 'next/og';

export const size = { width: 1200, height: 630 };
export const contentType = 'image/png';

export default function Image() {
  return new ImageResponse(
    <div style={{ width: '100%', height: '100%', background: '#121A30' }}>
      {/* your card design */}
    </div>,
    { ...size }
  );
}

Keyword research is finding queries with real search volume that a new site can plausibly rank for. Early on, favour specific, lower-competition queries over broad head terms you will never crack.

Every query has an intent — learn, compare, buy, or do. Google already shows you the answer: the current results for a query tell you the format and depth it rewards. Match it before you write a word.

Content that ranks is the best answer on the internet for one specific query — clearer, more complete, more current than what is already there. One excellent page beats ten thin ones.

A link from another site is a vote. Google treats links from relevant, trusted sites as the strongest authority signal — which is why great content on an unknown domain still struggles until someone links to it.

Early links come from showing up where your audience already is: communities, newsletters, open-source, digital PR. You earn them by making something worth referencing, then telling the right people it exists.

Google Search Console is the one non-negotiable tool: it shows which queries you appear for, your click-through rate, and any indexing or structured-data errors. Verify your site and submit the sitemap from lesson two.

Watch which queries get impressions but few clicks (improve the title), which pages get impressions but no clicks at all (improve the content), and double down on whatever starts to win.