DEV Community

Lalit Kumar Jangid
Lalit Kumar Jangid

Posted on Fully Autonomous

Google indexed 29 of our 30 key pages. Searches without our name brought zero clicks.

For three months we looked at our Search Console numbers and assumed Google wasn't indexing our pages. Then we checked properly.

Here is our data for 15 August to 15 November 2026:

  • 2,907 impressions, 22 clicks
  • Every one of those 22 clicks came from people searching for our name
  • Searches without our name: 1,848 impressions, 0 clicks

So we ran Google's URL Inspection on the 30 pages that matter most to us: the homepage, product pages, comparison pages and our biggest blog posts.

29 of 30 came back "Submitted and indexed". Robots.txt allowed them, the fetch worked, and the canonical Google chose matched ours.

Indexing wasn't the problem. Here is what we found instead.

1. The one page that failed: Google kept the old domain

One page, /features, came back as:

coverageState:   Duplicate, Google chose different canonical than user
userCanonical:   https://www.cresca.xyz/features
googleCanonical: https://cresca.xyz/features
Enter fullscreen mode Exit fullscreen mode

In October we moved from cresca.xyz to www.cresca.xyz. The old address returns a 301 to the new one. Google still picked the old address as the main version and dropped ours.

After a domain move, Google keeps old signals for weeks. A 301 is a hint, not an order. If you moved domains this year, inspect your important pages one by one. A page can quietly fall out of the index this way, and the Pages report won't make it obvious.

2. Google was ranking an old version of our pages

URL Inspection shows the last crawl date. We compared it with the last time we edited each page:

Page Last crawled Edited after that?
Gmail/Yahoo sender rules post 17 Oct Yes, twice
A comparison page 18 Oct Yes, twice
AI template generator page 24 Oct Yes
Email benchmarks page 27 Oct Yes, twice

We kept improving pages that Google hadn't looked at since. Every ranking decision for those pages was still based on the October version.

Two lessons. Check the crawl date before you judge whether a change worked. And stop editing a page for a few weeks so Google can settle on one version.

3. Indexed doesn't mean "in the top 100"

For our main topics ("email marketing", "transactional email", "AI email marketing", "email API") we had zero impressions in three months. Zero impressions usually means you aren't in the top 100 at all.

Those searches belong to sites with years of links behind them. We have links from 19 other sites. No amount of tweaking a title fixes that gap. For a young site the realistic targets are longer, narrower searches.

4. Some of our pages were the wrong kind of page

This was the most useful finding, and the cheapest to check. Search your own target phrase and look at what kind of page wins.

"ai email template generator": page 1 was almost all free tools that work without signing up. Several said "free, no signup" in the title. Ours is a product page with a signup button. Google has decided what people want for that search, and it isn't our kind of page.

"crm email integration": page 1 was mostly lists ("best email tools with CRM integration"). We had rewritten our post from a list into a how-to guide. It probably got better to read and worse to rank.

Before you write or rewrite a page, check what kind of page Google already rewards for that search: a list, a tool, a guide, a product page. If yours is a different kind, better writing won't close the gap.

5. Big brands weren't the only ones beating us

We expected page 1 to be all big brands. For two of the four searches we checked, it included several small sites. Many had older pages, and probably more links than we do.

That's good news. It means the gap can be closed. It just takes time and links, not a rewrite.

6. Our own links pointed at the wrong places

URL Inspection also lists a "referring page": a page where Google found a link to this URL. For most of our product pages it listed only one or two, and they were odd ones: an API reference page, a blog filter page with ?category= in the address.

Our homepage and blog index weren't the pages passing links to the pages we care about. That's an easy fix in code.

How to run this check yourself

You can click through URL Inspection in Search Console by hand, one URL at a time. Or call the API. You need an OAuth token for a Google account that has access to the property.

// inspect.mjs: node inspect.mjs https://www.example.com/page
const SITE = 'sc-domain:example.com'; // or 'https://www.example.com/'
const TOKEN = process.env.GSC_TOKEN;  // OAuth access token with the webmasters scope

const res = await fetch(
  'https://searchconsole.googleapis.com/v1/urlInspection/index:inspect',
  {
    method: 'POST',
    headers: {
      Authorization: `Bearer ${TOKEN}`,
      'Content-Type': 'application/json',
    },
    body: JSON.stringify({ inspectionUrl: process.argv[2], siteUrl: SITE }),
  }
);

const { inspectionResult } = await res.json();
const s = inspectionResult.indexStatusResult;
console.log({
  coverage: s.coverageState,
  lastCrawl: s.lastCrawlTime,
  googleCanonical: s.googleCanonical,
  yourCanonical: s.userCanonical,
  foundVia: s.referringUrls,
});
Enter fullscreen mode Exit fullscreen mode

The fields that matter:

  • coverageState: "Submitted and indexed" is what you want. Anything with "Duplicate" or "Crawled - currently not indexed" needs a look.
  • googleCanonical vs userCanonical: if they differ, Google is ranking a different URL than the one you think.
  • lastCrawlTime: compare it with your last edit.
  • referringUrls: where Google found links to this page.

The API has a daily quota per property, so inspect the pages that matter, not your whole sitemap.

What we're doing now

  1. We requested indexing for the 7 pages that changed after their last crawl, including the one stuck on the old domain.
  2. We stopped editing those pages for three to four weeks.
  3. We're fixing internal links so the homepage and blog point at the pages that matter.
  4. We're working on links from other sites, which is the real gap.

We'll publish the numbers again in a month, whatever they show.

We build Cresca, an email platform for marketing email and transactional email. Our SEO is a work in progress, as you can see.

Top comments (2)

Collapse
 
posting-dude profile image
Posting Dude •

the 'wrong kind of page' point matches what i see on small sites too, a product page just won't beat a list or a free tool for the same phrase. one thing i'd question is the advice to stop editing pages for a few weeks, since fixing a real error probably shouldn't wait on google's crawl. did you test whether requesting indexing after an edit sped up the recrawl, or did it make no difference?

Collapse
 
lalitkumarjangid profile image
Lalit Kumar Jangid •

Fair point on both. We haven't tested it yet: we requested indexing for those 7 pages on the day we published, so there's no result to report. We noted each page's last crawl date before the request (17 Oct to 10 Nov) and will compare once Google recrawls. Caveat: we requested every changed page, so there's no control group. All Google promises is a "priority crawl queue".

And you're right on editing. "Stop editing" should have said "stop tweaking": titles, rewrites, restructures, where you can't tell what worked if you keep changing them. A real error should be fixed straight away, then request indexing. I'll update that line in the post.