I Crawled My Own Site with the Free Version and Found a Defect in Half the Titles

A crawler is usually demonstrated on an example site nobody gets to check.

I ran it on mine. It is public, it is in Portuguese, it has 42 URLs, and you can repeat the crawl while you read.

The result has a good part and a bad one, and the bad one is more useful.

What the free version does, and where it stops

Screaming Frog SEO Spider has a free version, and it handles most of what this article shows.

Its limit is 500 URLs per crawl. ¹

Left out, among other things: crawl configuration, saving crawls, JavaScript rendering, custom extraction, and API integrations. A license costs $279 a year, or £199 a year. ²

Five hundred URLs covers a brochure site, a service site, and a small blog. It does not cover an e-commerce catalog, and that is where the paid version stops being optional.

One note that matters more than it looks. The test described here was run on version 22.2. The version published on the official site on September 1, 2026 is 24.3. ³ Two versions apart. If your screen looks different from mine, that is why. And it is the reason I state version and date instead of telling you to just click over there.

The crawl, from zero

Paste the site address into the top bar and click Start. That is it.

In my case, the crawl finished with 42 URLs encountered and 42 crawled, one hundred percent complete. The summary screen broke it down like this: ⁴

Total URLs Encountered             42
Total URLs Crawled                 42
Total Internal URLs                14   (33.33%)
Total External URLs                28   (66.67%)
Total Internal Indexable URLs      12   (85.71%)
Total Internal Non-Indexable URLs   2   (14.29%)

Two thirds of the URLs found are external. They are the links leaving the site: stores, profiles, cited sources. The tool lists all of them, and going through that list is the fastest way to find a broken outbound link.

The three tabs that answer almost everything

Internal

Lists what exists inside the site, with content type and response status. It is where you see, in one pass, whether an important page was left out or answers wrong.

H1

In my crawl, 11 URLs with an H1, zero missing, zero duplicated, zero multiple, and zero out of order. ⁵

This tab is usually the first to show a problem on a site built in a hurry. A page with no H1, a page with two, or an H1 that is a logo instead of text.

Meta Description

11 URLs, all filled in. ⁵

Here the comparison with a real case pays off, measured on another site: a description present on 1 of 22 pages. The difference between the two is not the tool, it is the publishing process.

The defect that showed up on my site

Now the bad part, and it sits in the titles tab.

Page Titles · Over 60 Characters    6 of 11    54.55%
Page Titles · Over 561 Pixels       6 of 11    54.55%

⁶

More than half the titles on my own site run past the limit.

A long title gets cut in the display. And the cut piece is precisely the end, where the part that tells one page from another usually sits.

I am leaving the defect live on purpose while this article circulates. That way you run the crawl and see the same number I saw. Fixing it later is easy, and having an example the reader can check is rare.

And there is a method reason for that. Auditing other people’s sites with screenshots of other people’s sites is cheap. Auditing your own, publishing the result, and leaving it open to be checked is the only version of the thing that costs you something.

What to do in the first ten minutes

Four questions, in order.

Do all your important pages appear in the Internal tab? If one is missing, no internal link is reaching it, and that problem comes before any optimization.

Is any page missing an H1, or carrying more than one? The H1 tab answers in two seconds.

How many pages have no meta description? The Meta Description tab gives the number.

How many titles run past 60 characters? That is where I came off badly, and it is probably where you will too.

What this crawl does not show

The tool reads the site like a visitor following links. It has no idea what Google did with any of that afterward.

How many requests Google spent, how many went to redirects, how long your server took to answer: none of that is here. It sits in another report, and that report is the subject of the piece on the Search Console crawl stats.

The two readings complete each other. One shows the site as it is. The other shows what Google did with it.

Frequently asked questions

Does the free version work for my site?

If it has fewer than 500 URLs, it covers everything this article shows. The free limit is 500 URLs per crawl. ¹

Why does my screen look different from the one in the article?

The test here was run on version 22.2, and the current version is 24.3. Tool interfaces change between versions. ³

Does the crawl show what Google did on my site?

No. It shows your site as a crawler walks it. What Google did sits in the Crawl stats report in Search Console.

Is a title over 60 characters an error?

In the code, no. The risk is truncation in the display, and the truncated piece is usually the one that tells the page apart.

Notes and sources

1. Free version limit: Screaming Frog, official site, screamingfrog.co.uk/seo-spider/, consulted on September 1, 2026. Crawl limit of 500 URLs.
2. Features restricted in the free version and license price: same source as note 1. Crawl configuration, saving crawls, JavaScript rendering, custom extraction, and API integrations. License at $279 or £199 a year.
3. Version tested and current version: the crawl described was run on Screaming Frog SEO Spider 22.2, on September 1, 2026. The version published on the official site on the same date is 24.3.
4. Crawl of seodeentidade.com, the author’s own site, published in Portuguese, run on September 1, 2026 at 7:58 a.m. Overview panel, Summary section.
5. Same crawl as note 4, H1 and Meta Description tabs.
6. Same crawl as note 4, Page Titles tab. Six of eleven URLs above 60 characters and above 561 pixels, or 54.55% on both criteria.