How I Verified Build With Less AI in Google Search Console

AI Export Lab · June 18, 2026

Last night was not a big AI breakthrough.

It was one of those boring website setup nights where the work looks small from the outside but decides whether Google can actually see the site.

I verified Build With Less AI in Google Search Console, submitted the sitemap, checked the basic static-site files, and cleaned up old TCM URLs that no longer fit the direction of this website.

It sounds simple when written in one sentence.

In reality, I still managed to lose time on one of the most basic problems possible: the verification file name.

That is why I am writing this as a public build log. Not because the task is advanced, but because this is the real work behind a small website. You do the simple things, you check them, you get one detail wrong, and then you fix it without pretending it was more elegant than it was.

This site is about building with less AI and more evidence. So here is the evidence trail.

What I Was Trying to Verify

The goal was not to rank overnight.

The goal was much smaller:

prove that I control the domain in Google Search Console
make sure Google can find the sitemap
make sure important pages use clean canonical URLs
redirect old content paths instead of leaving broken URLs
confirm that this static blog has a stable structure before I publish more build notes

This matters because a new site does not need more decoration in the first week. It needs boring reliability.

For Build With Less AI, the structure is simple:

static HTML pages
folder-style URLs such as /ai-export-lab/
a sitemap at /sitemap.xml
a robots file at /robots.txt
a small contact flow
AI Export Lab articles as the main content section

I do not want to keep rebuilding the homepage every time I feel impatient. I want a stable base, then consistent publishing.

Step 1: HTML File Verification

Google Search Console gives several verification methods. I used the HTML file method.

The process should be very simple:

Download the verification file from Google.
Upload it to the root of the website.
Visit the file URL in the browser.
Click Verify in Search Console.

The important detail is that Google does not only check the file content. It also checks the exact location and exact file name.

That is where I made the mistake.

At first, the file was not being detected. The site was live, the hosting was working, and the homepage was loading. But Search Console still could not verify ownership.

The problem was not a deep technical issue. It was the file name.

The verification file must keep the exact google...html file name that Google gives you. If the name changes, if the file is placed in the wrong folder, or if the upload tool silently changes something, Google will not accept it.

The fix was boring:

create the exact verification HTML file
put it in the website root
upload it again
open the verification URL directly in the browser
only then click Verify

Once the direct URL loaded, verification worked.

That little mistake is a useful reminder: before blaming hosting, DNS, Google, AI, or "technical SEO," first check the file path.

Step 2: Sitemap Submission

After verification, I submitted the sitemap:

https://buildwithlessai.com/sitemap.xml

The sitemap is not a ranking trick. It is a map. It tells Google which URLs are meant to exist.

For this site, I want the sitemap to contain only real canonical pages:

homepage
About
Start Here
Contact
Privacy Policy
AI Export Lab
Live Stats
Updates
Services
individual AI Export Lab articles

I do not want old experimental URLs, duplicate paths, or draft sections inside the sitemap.

The sitemap being accepted does not mean the pages are indexed. That distinction matters.

Search Console can show a sitemap as successful while individual pages still wait for crawling or indexing. So the correct conclusion is:

Google can read the sitemap.

Not:

Google has indexed everything.

That difference prevents bad decision-making.

Step 3: Robots.txt Check

The robots file is intentionally plain:

User-agent: *
Allow: /

Sitemap: https://buildwithlessai.com/sitemap.xml

For a small public blog, I do not need complicated crawl rules.

I want Google to crawl the site. I want the sitemap location to be obvious. I do not want to accidentally block the section I am trying to grow.

This is one of those files where "clever" is usually worse than simple.

Step 4: Old TCM Links Needed a Decision

The first version of this website mixed two ideas:

AI-assisted export website building
TCM food therapy content

Both are real parts of my background, but they do not belong in the same early-stage SEO structure.

The current direction of Build With Less AI is clear:

A non-programmer in China uses AI to build, improve, and measure a real B2B auto parts export website.

That means the TCM section should not remain as a separate content vertical on this domain.

So I had to decide what to do with old TCM URLs.

Deleting them without a redirect would create unnecessary dead ends. Keeping them would confuse the site direction. The clean choice was a 301 redirect.

The current redirect pattern sends old /tcm/ paths back to the AI Export Lab section, and the old TCM-focused article URL redirects to the updated non-programmer story.

That is the right direction for this site:

the personal TCM background stays in the story
the website does not become a TCM advice site
old paths do not become dead URLs
Google gets a clearer topical signal

This was not only a technical cleanup. It was a positioning cleanup.

Step 5: Canonical URLs

For each public page, I want one clean canonical URL.

For example:

<link rel="canonical" href="https://buildwithlessai.com/ai-export-lab/">

Canonical tags are easy to ignore on a tiny static site, but they matter because I am using folder-style pages.

I do not want Google guessing whether the page should be treated as:

/ai-export-lab
/ai-export-lab/
/ai-export-lab/index.html

The canonical version should be the clean folder URL with a trailing slash.

This is not exciting work. But small canonical inconsistencies can turn into annoying indexing noise later.

Step 6: What I Checked Manually

After the files were uploaded, I checked the key URLs directly.

My manual checks were simple:

Item	What I checked	Why it matters
Homepage	Loads normally	Google and users need a stable entry point
Search Console verification file	Direct URL opens	Ownership verification depends on exact file access
Sitemap	`/sitemap.xml` loads	Google needs a readable URL list
Robots	`/robots.txt` loads	Crawl rules should not block the site
AI Export Lab	Hub page loads	Main content section must be reachable
Old TCM paths	301 redirect works	Removed content should not become broken paths
Contact page	Contact path exists	B2B trust needs a working inquiry route

The important part is not that this checklist is complex. It is that each item has a visible result.

If I cannot open the file, Google probably cannot either.

What Went Wrong

The main mistake was the verification file name.

It is a small mistake, but it is exactly the kind of mistake that happens when you are building a site manually with AI assistance.

AI can generate code. AI can suggest a sitemap. AI can explain Search Console.

But AI cannot magically make Google see a file that was uploaded with the wrong name or wrong path.

The lesson is simple:

In technical SEO, boring exactness beats smart explanations.

The file name has to be exact.

The sitemap URL has to be exact.

The canonical URL has to be exact.

The redirect rule has to match the real old path.

Small static sites are simple, but they are not forgiving.

What This Does Not Prove Yet

I want to be careful here.

This setup proves that the site is ready to be discovered.

It does not prove:

that all pages are indexed
that Google trusts the site yet
that the articles will rank
that traffic will come quickly
that the inquiry funnel is already producing results

Those are different stages.

For now, this was the foundation stage:

verify ownership
submit sitemap
clean up old paths
stabilize structure
start publishing consistently

The next stage is watching Search Console without obsessing over it.

My Search Console Rule for the Next Few Days

I am not going to keep refreshing Search Console all day.

That is a bad use of attention.

For the next few days, I only need to check:

whether pages are discovered
whether any page becomes indexed
whether the sitemap remains successful
whether crawl errors appear
whether queries start showing up

That is enough.

The better use of time is publishing useful build notes, not staring at a dashboard that needs time to collect data.

My current split is:

70% writing
20% checking Search Console
10% small technical fixes

That feels right for a new site.

Why I Am Publishing This

Most website-building content skips this part.

People show the finished homepage, the AI prompts, or the traffic chart after it starts looking good.

But early-stage website work is mostly this:

one file name wrong
one redirect rule missing
one sitemap submitted
one page inspected
one section removed because the positioning was too broad

This is what "build in public" should include.

Not only wins.

Not only screenshots.

The tiny setup details too.

Because for a non-programmer building with AI, these are the exact moments where the project either becomes real or stays as a folder of nice-looking files.

Last night, Build With Less AI became a little more real.

Current Status

As of this build log:

Google Search Console verification is handled
sitemap submission is handled
robots.txt points to the sitemap
old TCM paths have a redirect decision
the AI Export Lab remains the main content direction
the next work is publishing more useful build notes

No victory lap yet.

Just a clean baseline.

That is enough for today.

→ AI Export Lab — the public record of this project.
→ Live Stats — Search Console evidence from the project.