Technical SEO and indexing
Before any content can rank it has to be crawlable, canonical, structured and indexed. Check the pages that take money first, and confirm indexing in the search engines rather than assuming it.
19 steps, about 14 hours of work · 1 plays · 1 checked claims
The concept, explained in full: SEO.
Do it
The steps from the launch plan that belong to this topic, in plan order.
Make the site crawlable and citable
- One H1 per page, sequential heading hierarchy45m
Sequential headings appear on 68.7% of ChatGPT-cited pages against 23.9% of Google’s top results, and 87% of cited pages have a single H1 (AirOps, 2025). It is a correlation, not proof that headings cause citation.
- Generate sitemap.xml automatically30m
Hand-maintained sitemaps rot within a month. Generate it from your routes.
- Allow crawling in robots.txt and name the sitemap10m
A staging Disallow shipped to production is the single most common launch-week disaster.
- Add canonical tags to every page30m
You will syndicate to Medium and Dev.to in Phase 2. Canonicals are what stop that hurting.
- Add OG and Twitter card images45m
Every share for the next two years renders through these. Make them readable at thumbnail size.
- Pass Core Web Vitals on mobile2h
Test on a throttled 4G profile, not your laptop.
- Read your own page the way a crawler does30m
Fetch it with JavaScript disabled: one audited site was serving hundreds of Lorem Ipsum pages to crawlers while looking finished in a browser.
- Confirm Google has indexed every page that takes money15m
Homepage, pricing, signup and your best feature page. A page Google has not indexed cannot rank and cannot appear in an AI Overview, and a nosnippet tag keeps it out of the AI answers even when it is indexed. Until these four pass, a deeper technical audit can wait.
- Add Organization schema30m
Name, logo, URL, sameAs to every social profile. This is what entity resolution keys on.
- Add SoftwareApplication or Product schema30m
Category, operating system, price. Review sites and AI answers both read it.
- Add FAQPage schema to the homepage and pricing page30m
The highest-leverage schema for answer extraction, and you already wrote the FAQs.
- Add BreadcrumbList to nested pages30m
Matters once you have hundreds of programmatic pages. Add it before you do.
Get indexed at scale
- Submit the sitemap to Google and Bing15m
Then check the coverage report for what was excluded and why.
- Set up IndexNow45m
Push new URLs on publish instead of waiting for a crawl. One ping reaches Bing, Yandex, Naver, Seznam and Yep — and, through Bing, Yahoo, DuckDuckGo and Copilot.
- Make every page reachable in three clicks2h
Orphan pages do not get crawled. Internal links are the cheapest ranking factor you control.
- Clear the crawl errors in Search Console2h
Especially “Discovered — currently not indexed”, which usually means thin or duplicate.
- Give every page four links from the body of other pages2h
Nav and footer links sit on every page and separate nothing. Links written into the prose are the ones that carry — and vary the words you link on, because the same anchor repeated reads as a template. Read each anchor inside its own sentence: one that promises something the destination does not contain costs you the click and the reader.
- Track indexed as a share of submitted, not as a count30m
A hundred indexed pages reads as progress until you notice you asked for five hundred. One audited site had roughly one sitemap URL in five indexed.
- Claim the national consoles wherever a local engine competes with Google45m
Read your country split first, then claim only where buyers are: Yandex (Russia, Turkey, CIS), Naver and Daum (Korea), Seznam (Czechia), Cốc Cốc (Vietnam), and Baidu, then Sogou, 360 and Shenma (China, which also wants Chinese pages and an ICP-filed host). All free.
Plays
Reusable patterns that explain how: backed by read sources first, then practitioner advice, then our own ideas.
- Turn the "crawled, not indexed" list into a merge-or-improve sprintBacked · 1 source
Search Console lists pages Google crawled and chose not to index. Export that list and sort it: improve the pages that matter, merge the ones that repeat another page, and remove the rest. Google says the status is common and not always a problem, so judge by whether the page earns its place, then re-inspect the ones you fixed.
What the evidence says
Sources
Related topics
Where it sits in the plan
- 1.3.1Clear the technical baselinein Ship-ready
- 1.3.2Ship structured datain Ship-ready
- 3.3Get indexed at scalein Authority baseline