Put out your weekly post. Put some money behind it with paid traffic and be a presence on LinkedIn. You can do all of that and still have the traffic chart go nowhere for months on end. It is maddening, but then you realize the trouble is not what it seems. Content is an easy target for blame because you can see it, yet the actual culprit is often something more subdued: Google is unable to crawl the important pages, the site is sluggish on a real phone, or your URLs are so at odds with one another that search engines can’t make head or tail of what is significant.
A technical SEO checklist is meant to address this. We are not talking about a 200-page PDF that gets sent to a developer and promptly forgotten. This is a no-nonsense health check a founder can put together in an hour, complete with notes for whoever has their hands in the code. If you have ever handed an engineer a vague instruction to “look into SEO” and been left with nothing, consider this the remedy.
Why a Technical SEO Checklist Matters More in 2026
There have been two changes in as many years that any founder ought to be aware of, since they dictate what your site needs to be good at.
Then there is the matter of Core Web Vitals. Back in March 2024, Google did away with First Input Delay and made Interaction to Next Paint the standard. INP is a truer measure of how responsive your page is to every tap and keypress, not just the first. You might have been fine before March and now you are not, thanks to some heavy third-party scripts or a slow event handler.
Google has also made plain that if a page gives them anything other than a 200 response, it can be pulled from the rendering queue. For those of you with JavaScript-heavy sites where content only appears after rendering, a minor status-code error on deploy is enough to get a page de-indexed until the next crawl.
It is not theoretical. Sure, 91% of web traffic is HTTPS and most pages have a canonical tag, but plenty of those are set up wrong. The point is not that technical SEO has become more difficult, per se. The failure modes have just moved, and an old checklist will not catch them.
You don’t need more SEO theory. What you need is a quick way to size up whether the site is in good shape for growth and a clean brief for the person who has to put it right.
How To Use This Checklist
We have written this for the business side of things, not the coder. You can work through it in an hour with Google Search Console, PageSpeed Insights and the like; no paid tools required.
Every check is broken down into three parts:
- The why: The business case, sans jargon.
- The how: Something you can do today.
- Handoff notes: Ready to paste into an email or ticket.
Go through the list and flag the issues by impact. Then put the short version in front of your developer in one go. Don’t go chasing nice-to-haves like extra schema when you should be dealing with speed, mobile and crawling.
Check 1: Can Google Find and Read Your Pages
A page that Google can’t get to won’t rank. It is the gatekeeper for everything else.
You want to look in three spots: your robots.txt, the XML sitemap and the Pages report in Search Console. As long as they are in agreement, you are fine. When they are not, put off writing new content and fix it.
How to check it
- Open robots.txt: Go to
yourdomain.com/robots.txtand see if anyDisallowrules are encroaching on your product or blog sections. People often leave a staging-era disallow in place after launch. - Review the sitemap: Usually you will find it at
/sitemap.xml. Make sure it is not listing thin or duplicate URLs and does include the ones you want indexed. - Read the Pages report: Under Indexing in Search Console, you will want to watch for “Crawled — currently not indexed” or “Duplicate, Google chose different canonical.” These are telltale signs of weak internal linking or conflicting signals.
For a wider view before you talk to anyone, our website audit guide shows you how to put what you find in a form a developer can use.
Here is a rule of thumb: if a money page isn’t indexed, stop producing content until you know why. Ignoring it is expensive. The fix is usually not.
Handoff notes for your engineer
Please take a look at crawlability and indexation. I want you to validate the robots.txt for any accidental blocks, make sure the XML sitemap is only pointing to 200s that we want indexed and run through the Pages report for any excluded URLs we should be seeing. Return a list of the priority URLs, and for each one include the reason for its exclusion along with what we would recommend to fix it.
Check 2: Is the Site Actually Fast on Real Phones
You will lose people and conversions to a slow site, and in doing so you are feeding Google quality signals that have a way of compounding over a few months. Google’s Core Web Vitals give you three thresholds to work with:
| Metric | What it is | Good |
|---|---|---|
| LCP | Speed of the main content appearing | Less than 2.5s |
| INP | Response time to a tap or click | Under 200ms |
| CLS | Any layout jumping during load | Below 0.1 |
These are all gauged at the 75th percentile from real Chrome users in the field, on mobile. That distinction is important. You might see a page turn green in a PageSpeed lab test only to have it fail out there in the wild because the field data picks up your ad scripts, analytics, chat widgets and third-party tags. The lab doesn’t.
Heavy tag managers and analytics running on every interaction are the usual culprits for a bad INP score, as are mega-menus that rebuild the DOM on hover and chat widgets present on every page. If your team has put in five new tools this past year without culling any old ones, I would bet your INP has regressed and no one has made the connection.
We have a guide for WordPress sites on how to improve website loading speed that makes the case for fixes that actually move the field score versus those that just put a good face on the lab numbers.
For your engineer
Go to the Search Console Core Web Vitals report and pull the field CWV for the homepage and our top templates. We want you to focus on INP and LCP at the 75th percentile on mobile. Have a look at unsized media, long event handlers and third-party scripts. Give us a ranked list of the top offenders and what delta we can expect from fixing them.
Check 3: Clean up your URLs, Redirects and Internal Links
This isn’t about cosmetics. Your internal structure tells Google which pages are of consequence, where the authority goes and how often it will come to crawl new material. Any page of import ought to be no more than three clicks from the front door. Left without an internal link, a page can languish in “Crawled — currently not indexed” for months on end.
The founder’s way of checking is simple enough. Put in your domain four ways – with and without the www, http and https – and you should arrive at the same canonical version in a single hop. Then run through your main nav and read the URLs. They need to be stable, readable and without tracking parameters.
Things to check
- A single primary domain: Bots and users alike should always hit the canonical version.
- No redirect chains: Don’t make a 301 take two or three hops to get to the final URL.
- No broken links internally: Whether in the footer, menu or body copy, they should all be returning 200.
- Use HTML, not JS alone: A crawler can’t see navigation built on onClick handlers. Stick to proper <a href> tags.
- Orphans don’t belong here: Link anything of value from an already-indexed page.
If you are running a JavaScript app, this is where your rendering choices will be exposed. See our JS single page application guide for when client-only rendering is acceptable and when it is quietly doing damage to your SEO.
As a rule, most of the mess you find in your URLs is just the residue of some past decision, be it a migration, a CMS switch, a redesign or a marketing campaign that left its tracking parameters in the canonical link. By themselves, none of these are fatal. Put them together and you have Google guessing.
For your engineer
Have a look at the site structure for any canonical domain inconsistencies, broken internal links or redirect chains. You will also want to check the JS-only navigation. Make sure your key pages are no more than three clicks from the home page and that an indexed page has at least one link pointing to them. If you come across any orphan templates or redirect loops (anything over one hop), put a flag on it.
Check 4: The Mobile Experience has to work
Google is indexing your mobile site, not the desktop version. Which means if mobile is the weaker of the two, that is what will rank. The usual problem is content that is there on desktop but gets removed, collapsed or otherwise buried when you are on a small screen. As far as rankings are concerned, that content has vanished.
Put aside the lab test and pick up your phone. Go to your homepage, a blog post and a key landing page and read them. Can you do it without zooming? Do you hit the right link every time you tap? Is the menu easy to find? Are you seeing the same essentials as a desktop user would? If you can’t say yes to all of that, your users are running into the same friction.
Handoff notes for the engineer
Audit the mobile side for both usability and indexing. I want you to confirm the main content is in parity with the desktop template and check your tap-target sizing. Look for any hidden or collapsed material that might be impacting keyword coverage and flag anything where the mobile version could be hurting search visibility.
Check 5: Does Google really get your content?
There is a difference between crawling a page and knowing what it is about. That is where structured data comes in. With JSON-LD you give Google the explicit label for whether a page is an article, a local business, an FAQ or a product. These days it does double duty: you still get rich results in classic search, but it is also a primary signal for AI search engines to determine what they should cite.
Try the Rich Results Test with some of your top URLs. What you are after is detected structured data and zero errors. The typical error is schema that doesn’t square with the page itself. Say your Product schema has a price but the page doesn’t; Google will simply disregard the markup, often without making a fuss.
And if you have a variety of page types, keep the schema in the template layer. Don’t put it in individual posts or it will drift as soon as your editors lose track of it.
Notes for the engineer
Run through the structured data on our principal templates. Validate each one in the Rich Results Test. We need Organization on site-wide items, Article on our posts, Product for commerce and so on. Let me know if you see any inconsistency in how it is implemented or any fields that don’t match what is visible on the page.
What a Checklist Cannot Do
You can get by with a founder’s checklist for a routine health check. But do not expect it to see you through a platform migration, a redesign, or a programmatic SEO buildout of thousands of templated pages that need to be indexed at scale. Nor will it suffice if your traffic has taken an unexplained nosedive. Those are matters of engineering judgment, systems problems that call for more than a few screenshots.
Take our work with Teton Gravity Research. On the surface we were in for a redesign. In reality, we had to relocate ten thousand articles from an old CMS and make sure the team did not lose their publishing workflows, search rankings or referral traffic in the process. You would not have seen most of the technical SEO; that was done months ahead of launch in the form of content model decisions, redirect planning and URL mapping.
We see the same thing at scale. The Keck School of Medicine at USC project meant dealing with almost ten years of newsroom material spread over microsites and subdomains, any one of which could be cited where it counts. Making sure the canonicals, sitemaps and redirects were in order was not a box to tick, it was the whole project.
So when you are faced with a domain consolidation or a CMS switch, remember that the work to safeguard your SEO is done before you go live. Our pre-migration checklist tells you what to put in place first.
Your Action Plan
What keeps a site from drifting for years is not how long your checklist is, but whether someone is on the hook for it and the fixes actually ship.
- Put aside an hour. Put your homepage, a couple of recent blog posts and your top three landing pages through all five checks.
- One line per issue. Note the page, the problem and what the impact is likely to be.
- Keep the handoff brief. Refer to the notes here. One ticket per check, not for every little issue.
- Do not assume it was done right. Go back and re-run the check yourself after a week.
- Make it a quarterly habit. New marketing tags, a plugin update or a deploy will bring most issues back around. That is the cadence seasoned teams use.
And if you are still deciding on keywords, sort that out before you put money into content. It is upstream of everything else. We have a guide to popular SEO keywords for finding terms with qualified traffic rather than just volume.
For the bigger situations – a migration you have penciled in, a programmatic site with unindexed pages, or a traffic drop you cannot put your finger on – there is no substitute for Refact’s SEO audit and optimization. An audit and this checklist have the same aim: to take that hunch that something is off and make it a short list of things to fix.
