Testing SEO plays. Day job @wise. 🚀 Often political. https://nitter.cf/t.co/hiqmtLXKx6 / https://nitter.cf/t.co/GO4CKi8HWS
Leeds, UK
Joined March 2007
- Tweets4.4K
- Following236
- Followers302
- Likes311
It's not 100% luck.
I'm not saying Marc is saying/doing this, but to not mention that some luck is at play is dishonest.
If @levelsio lost everything today, I bet my net worth he’d make it again. And even faster.
Same for @jackfriks and a few other people I’ve had the pleasure to meet.
It’s anything but luck.
1
23
Shree retweeted
If @levelsio lost everything today, I bet my net worth he’d make it again. And even faster.
Same for @jackfriks and a few other people I’ve had the pleasure to meet.
It’s anything but luck.
This quoted post is unavailable.
What am I doing wrong @OfficialLoganK @_mohansolo — AI Pro users, still no Argon in @antigravity — I do have it in AGY CLI though 🤔.
I have restarted app, machine and checked for updates... not on a student plan but was on the 50% off monthly offer — surely that can't be it?
1
3
71
What is going on with @googlesearchc — reports are now over a week out of date. Indexing, sitemaps, all just useless, annoying when you're trying to jump on things quickly.
31
I'm moving from @Netlify to @Cloudflare workers. My domain is already setup in CF.
I would love when adding a new domain to a worker for it to 1-click replace the old records, rather than just warn me and make me do it manually like it's 2025.
1
1
46
Taxlight.co.uk was stuck on one indexed page for ages.
Dropping the sitemap from 6.4k pages to 1.4k pages seems to have helped.
Submitted 10 pages manually. It seems Google is crawling to add more and more.
This was 14 pages yesterday. 🐢 Slow and steady.
1
17
I built a programmatic SEO site.
In 3 weeks it went from 0 → 10,000 indexed pages and 6,000 impressions/day.
Then it fell off a cliff:
📉 -72% impressions
📉 -65% clicks
📉 2,300 pages wiped out overnight
Here’s how a simple sitemap bug almost killed the domain (and how we fixed it) 🧵👇
2
63
The cleanup in numbers:
• Origin URLs: 22,842 → 1,771 (-92% bloat)
• Faith URLs: 3,645 → 451 (-85% bloat)
• Empty URLs in sitemaps: 20,565 → 0
• Historical clicks preserved: 87%+
Total sitemap went from 67,000 → 41,500 URLs. Every single remaining page has real, dense content.
1
13
The fix was ruthless pruning:
1️⃣ Gated Category Hubs ($\ge 10$ names): Pruned 340 ghost categories. Only 223 high-volume hubs get indexable status.
2️⃣ Gated Letter Pages ($\ge 5$ names): A letter subpage only exists if that letter has $\ge 5$ names. Otherwise: clean 404 rewrite.
3️⃣ Killed Internal Crawl Loops: Updated the Alphabet bar UI to gray out / unlink empty letters so bots can’t crawl into doorway traps.
3
25
The lesson for programmatic SEO:
"Give Google everything and let them sort it out" works during the honeymoon crawl—until Google decides 80% of your site is thin noise.
Google doesn't index potential. It indexes density.
If you don't gate your programmatic matrices by actual data depth, you end up DDOSing your own crawl budget.
Fix shipped. Sitemaps submitted. Now tracking the recovery curve 📈
13
79 of those 81 Akkadian URLs had ZERO names.
When Googlebot crawled them, our template rendered an empty state with <meta name="robots" content="noindex">.
Across the site, we had submitted 20,565 completely empty pages directly to Google in our XML sitemaps.
Googlebot was burning its entire crawl budget hitting doorway dead-ends and being told "do not index me".
Meanwhile, 36,500 legitimate, data-rich name profile pages sat in Discovered – currently not indexed, completely starved of crawl capacity.
15
Digging into the raw GSC coverage export revealed the smoking gun:
19,935 pages were flagged as Excluded by ‘noindex’ tag.
To capture long-tail searches, we had set up routes for every origin + letter combination: /[gender]/origin/[origin]/[letter]
Our sitemap blindly looped through A → Z for all 282 origins and 3 genders.
Take "Akkadian" names: There were literally 2 names in our database. Yet our sitemap submitted 81 URLs for it.
1
3