Technical SEO for Developers: The Checklist We Ship With

By Nikola, founder at ZEROX. We build HubSpot, Shopify, and custom web apps where organic traffic has to survive launch day, so this list runs on every project we ship. Last updated February 2026.

Most technical SEO problems are not strategy problems. They are things a developer did or forgot at build time: a stray noindex left over from staging, a canonical pointing at the wrong URL, a redirect chain three hops deep. This is the technical SEO checklist we run before any site or app leaves our hands, written for developers and agencies who want search baked into the build instead of bolted on after a ranking drop.

It is deliberately code-level: no keyword research, no content calendars, no link building, just the parts in your templates, server config, and head tag. The cheapest time to get them right is before launch, not after a ranking drop three months later. We treat SEO as a build requirement that belongs in the definition of done, not a follow-up ticket. If nobody searches for you yet, this is not for you.

The technical SEO checklist we ship with

Here is the short version, grouped the way we work through it. The sections below explain the parts that trip teams up.

• Crawlability: robots.txt opens real content, blocks admin and cart, with no accidental site-wide Disallow
• Indexing: important pages return 200 and are indexable, staging noindex removed, thin pages noindexed
• Canonicals: a self-referencing rel="canonical" with an absolute URL on every page
• Sitemaps: canonical, indexable URLs only, linked in robots.txt and submitted to Search Console
• Metadata: unique title and meta description per page, from template data
• Structured data: valid JSON-LD matched to the page type
• Semantic HTML: one h1 per page, logical heading order, real landmarks
• Internal linking: every indexable page reachable by a crawlable link, no orphans
• Redirects: 301 for permanent moves, no chains, a full map on any migration
• International: hreflang only if you serve multiple languages or regions
• Performance and mobile: Core Web Vitals in budget, responsive, parity across devices

Crawlability, indexing, and canonicals

These fail quietly: the site looks perfect to a human and is invisible to Google. robots.txt should block only what should not be crawled, such as internal search, cart, account, and admin, and reference your sitemap; the classic disaster is shipping the staging robots.txt, with a site-wide Disallow, to production, so diff it before every launch.

Indexing is a separate lever: a page kept out of the index must stay crawlable and carry a noindex tag, because a page blocked in robots.txt is one where Google never sees the noindex. The most common wound we find on audits is a site-wide noindex that survived staging.

Canonicals name the real URL when several resolve to the same content. Every page should carry a self-referencing rel="canonical" with an absolute URL; get this wrong and you can canonicalize the whole site to the homepage. Watch trailing-slash and http versus https variants, paginated lists, and faceted URLs. Your sitemap should then list only canonical, indexable, 200-status URLs, since one that is noindexed or redirected sends a mixed signal.

Structured data and semantic HTML

Semantic markup is free accessibility as well as free SEO. Use one h1 per page, keep headings in logical order rather than skipping levels for styling, and use real landmarks: header, nav, main, footer.

Structured data is how you qualify for rich results. Add JSON-LD in the head and match the schema to the page: Organization, Article or BlogPosting, Product, BreadcrumbList. Validate every template against Google's Rich Results Test, and only mark up content that is actually visible, because marking up what the user cannot see is a guidelines violation, not a shortcut.

Internal linking is the other half of crawlability: every indexable page should be reachable through a normal, crawlable link, not a button that fires JavaScript or a menu that only appears after a click. A page reachable only through search or the sitemap is orphaned, so link to it from somewhere relevant.

Redirects, hreflang, and migrations

Redirects are where traffic goes to die on a rebuild. The rules we never break: use 301 for permanent moves, redirect every old URL with traffic or backlinks, send each to the closest relevant page rather than the homepage, and never chain them, so it is old URL to final URL in one hop. On our HubSpot and Shopify migrations a complete 301 map is non-negotiable, checked by crawling the new site against the old URL inventory before launch.

hreflang only matters if you serve multiple languages or regions, and it is easy to get wrong. Every version must reference every other version, including itself, with valid language and region codes plus an x-default. If the return tags do not match in both directions, Google ignores the whole set, so skip hreflang on a single-locale site.

Performance and mobile

Google indexes the mobile version of your site, so the mobile experience is the real one. Serve the same content and structured data on mobile as on desktop, where collapsed or in an accordion is fine but missing is not, and test on a real mid-range phone, not a desktop emulator.

Performance is a ranking factor and a conversion factor at once, so we hold every build to a Core Web Vitals budget and treat regressions as bugs. What LCP, INP, and CLS actually require for a B2B site, and how to hit them, is its own topic, covered in Core Web Vitals for B2B. The short version: measure field data, set a budget, and do not let a marketing tag or a hero video quietly blow it.

FAQ

What is a technical SEO checklist?

The build-time, code-level items that decide whether search engines can find, understand, and rank your pages, from crawlability and canonicals to sitemaps, structured data, redirects, and performance. It is separate from content and link building.

Is technical SEO a developer's job or an SEO's job?

Both. Most of the checklist lives in code, so it is the developer's to implement, while SEOs set the strategy and targets. They work from the same list.

Do I need structured data on every page?

No. Add JSON-LD where it matches a supported rich result, such as articles, products, and breadcrumbs. Never invent schema for a page with no rich-result type, and never mark up content that is not visible.

Will a redirect hurt my rankings?

A single clean 301 passes almost all signals and is the right tool for a moved URL. Damage comes from chains, redirecting everything to the homepage, or forgetting URLs that had backlinks.

How often should I re-run the checklist?

On every launch, every migration, and after any major template change. We also re-check the quiet items, robots, canonicals, and sitemap, about a week after go-live, when staging leftovers surface.

Work with us

If you want SEO baked into the build instead of rescued after a traffic drop, that is how we ship every HubSpot, Shopify, and web app project. See our services, or tell us what you are launching on /contact and we will tell you honestly where the technical risks are.