


makes and ranks things



Will I ever think of a joke that's for a smaller niche within a niche? Time will tell.

Good morning, Crawlers! Welcome to the eleventh floor of the Organic Search dungeon. This season of the crawl is proudly sponsored by the Alphabet Corporation, who also own the dungeon. Now, some quick patch notes. It has come to our attention that some crawlers were using sponsored listicles to take advantage of the System AI. Very clever. We love to see that kind of initiative. It has been patched in our latest update. Crawlers who relied on this exploit will find their rankings have been adjusted accordingly. A reminder about loot: reaching position one no longer guarantees a click. The System AI now answers the query for the user, using your content. You’re welcome. The rules for this floor have not changed. Make helpful content for users. The definition of “helpful” has not been published. That is all. Now get out there and rank, rank, rank!

TL;DR: I don't want a new seat at the table under a different name. I want recognition that SEO deserved that seat all along.

I think I finally figured out why the SEO vs GEO thing sticks in my craw so much. The things that would have laid the groundwork for good AI search performance today are the things SEOs have been talking about for 10 years. Next year is the 10-year anniversary of Traffic Think Tank. I know for a fact hundreds of people have been talking about the stuff that’s now making the rounds as “new approaches” for surfacing in ChatGPT, AI Overviews and Claude. (Not Perplexity. Nobody seems to care about Perplexity anymore.) This should be the victory lap. This is when you look at the CMO who ignored your requests and your advice for a decade and say “I fucking told you so.” Instead, loud voices in the SEO community are saying we should repackage it. Call it GEO. Pretend to the CMOs (who didn’t want to listen to their SEO teams) that this is all new. That they haven’t heard it for the last 10 years. That they haven’t been sitting in C-level seats knowing nothing about how the channel actually operates. So instead of telling them we knew better this entire time, the advice is to kowtow. Use the language they want to use. Make believe it’s all new and different, and they’re so smart for finally wanting to do it. “Oh gosh, who could have seen this coming? We better get on this right away.” Say whatever they want to hear, because they control the money. Maybe it’s because I’m lucky enough that I don’t need to close X deals in the next three months to pay my mortgage. I’m good. So maybe that’s my privilege showing. I don’t have to make believe, and other people do, because they decided to run an agency and have full-time salaries to pay. If you have to say the shibboleth to make your bills next month, okay. Fine. But that doesn’t define what’s true. It defines what’s commercially useful. And I’m entirely uninterested in saying things I don’t think are true just because they’re commercially useful. If that’s where you are, I feel bad for you, frankly. Every single time I hear the “new” GEO work explained, I can think of five blog posts from seven years ago that describe exactly the same thing. You’re playing a bullshit game to impress dumb CMOs so they’ll pay you. I get it. Money’s important. But don’t shit a shitter. Don’t pretend when you’re talking to me that this is new or different or revolutionary. It’s the same bullshit you were doing eight years ago. You’re just calling it something else and charging more. And if you can get more money for the same old shit with a new coat of paint, good for you. I’m as staunchly pro-capitalist as anyone. I respect and applaud your right to get money from dumb people. Just don’t expect me to pretend publicly that they’re not dumb. It's still "just" SEO. And if you’ve been saying this stuff for a decade, take your well-earned victory lap.

If you do good job selecting keywords, clicks can predate pages even showing as indexed in Search Console.


Ironically, I'm scraping Google's results way more now, because I have to do an initial pass to build the expected domains for validation, and then re-run low confidence SERPs over and over until I get a plausible result set.

Meanwhile, today, ChatGPT literally can not create a report from a template it was given without changing chart types etc over and over again despite being told not to.
Paul Christiano (OpenAI board member) yesterday: "There's material risk of AI takeover in the next 6-12 months. We have seen roughly the things we might expect to see shortly before something really bad were to happen. Even with [any] realistic policy responses, the risk remains exceptionally high. If we build this as quickly as we can in the easiest way we can the chance is [more like] a coin flip [within the next 3 years]. For most people in this room, the most likely reason they will die is because of an AI accident." Christiano: • Is on OpenAI's safety and security committee (which exists to block risky practices) • Co-invented RLHF at OpenAI back in 2017-2021 • From 2024-2026 was head of safety at the US gov's AI agency (now called CAISSI) (Video is from 'The Hinton Lectures' there's another on tonight and tomorrow.)

100% of children want ice cream for dinner
66% want Medicare for All 61% want a billionaire wealth tax 57% want a $20/hour minimum wage 63% want a data center moratorium 58% want to end military aid to Israel 73% want to ban super PACs 79% want a 10% credit card interest rate cap No more excuses. Democrats must deliver.

Hey Anon, Do you believe this? "Parasites will last forever, there's nothing Google could do against it!" We said the same while using Xrumer for mass spamming links "Links will last forever, there's nothing Google could do against it!" Yet Google DID something. (Search for 'Google Payday Loan Update' and 'Payday Loan Update 2.0' if you're too young) Parasites won't last forever. It's a fact. Nothing lasts forever. One day parasites won't be no more, one day you'll be dead. Whichever comes first. Until then your job is to push this to the absolute max, same as we did with Xrumer in the past Not next month, not next week, not tomorrow, but TODAY Assume it's the last day working and then act RalfyIndex helps you pushing serious volume RalfyIndex is the tool for pushing limits and achieving your goals Good luck anon 🍀

Just told Claude we need to make a new skill, and it replied that we made a skill to do that exact thing, three days ago. Might be a summer Friday over here.

Companies that offer integrated solutions for local inference will boom in the next few years. I'm not talking about another GPU in a case. Hardware + OS + an inference stack that makes open models easy to run. A plug-and-play system. Ease of use is still underrated.

Good day to fully fund my HSA for the year? We’ll find out.

Weird time for AI. A week ago I was laughing about the 512gb mac studio price. The breakeven vs API pricing is just insanity. Then the party that hates AI just won the house and senate. They have 2 years ahead of them of likely getting veto'ed left and right. So the bet seems to be... "Is RAM going to get more or less expensive in the next 24 months"