Is SEO dead?
Apparently a lot of people are wondering these days. Between AI Overviews sitting on top of results and more people asking ChatGPT their questions instead of scrolling through ten blue links, it’s a fair question.
No. I don’t think it’s dead. It’s just splitting into two categories, and mixing them up is causing a bit of anxiety.
I’ve spent months figuring out what works and what doesn’t in AI search, and I wanted to share my findings with you regarding LLM SEO!
Why Everyone’s Asking If SEO Is Dead
The anxiety is understandable. Google’s AI Overviews now show up above traditional results for a huge number of searches. ChatGPT has become a default research tool for millions of people. Perplexity built its entire product around answering questions directly instead of sending you off to read ten different websites. Zero-click search has entered the mainstream.
But let’s not scream doom and gloom when it isn’t warranted.
If you’re a business owner or marketing manager watching your organic traffic reports, you’ve likely already noticed a session count that looks flatter than it should. That creates uncertainty and confusion, which can lead to panic.
So when someone types is SEO dead into Google (yes, that’s a real, frequently searched phrase), I get why. Things are shifting into unfamiliar territory. Business owners have spent countless hours learning SEO, only for the rules to change.
AI and LLM search are growing fast, but they’re still a small slice of total search volume compared to traditional search engines. Google still handles the overwhelming majority of queries worldwide. AI search is where things are heading, not where most of your traffic lives today. In short, you have time to get ahead of the curve!
So is SEO dead? No. It’s splitting into two buckets: what’s stayed the same and what’s new. Confusing the two is what leads business owners to either panic and throw money at the wrong things, or ignore AI search entirely and fall behind.
What Hasn’t Changed: The Fundamentals Are Still Doing the Work
Technical health should still be important to you. AI crawlers need a clean, fast, accessible site just as much as traditional search engines always have. If Googlebot has trouble crawling your site, OAI-SearchBot and PerplexityBot run into the same trouble. Site speed, mobile experience, and basic crawlability aren’t relics of old SEO. Just as a slow, cluttered site was never going to rank well in traditional search, it isn’t going to get cited in an AI answer.
Search intent matching hasn’t gone anywhere either. Content still has to answer the question a reader, or a model reading on that reader’s behalf, is asking. Large language models are, in a sense, just another reader. If your content rambles for 500 words before answering the question, that hurts you with a model the same way it hurts you with a person skimming for an answer. The businesses that win here are the ones that already write clear, useful content.
And then there’s authority. Backlinks, expertise, and brand recognition remain some of the strongest signals. A locally known contractor with real reviews and press mentions still has an edge over a brand-new site with none of that history, no matter which engine or model is doing the evaluating.
LLMs lean on the same trust signals search engines have used for years. They just fold those signals into citation decisions instead of rankings. The signal is the same, the output just looks different.
If you’ve spent years building technical health, matching content to intent, and earning authority the hard way, your work hasn’t been wasted. You’re not starting over. You do need to add to the foundation, though.
What Changed: The New Rules of LLM SEO
Now for what’s new. This is the heart of LLM SEO, and it’s where I spend most of my testing time these days. Some corners of the industry call this generative engine optimization, or GEO. Different name, same idea.
Three things make up this new layer of work (this is about to get technical, but try to bear with me):
- Crawler access and llms.txt. Explicitly allowing AI bots onto your site, OAI-SearchBot, PerplexityBot, Google-Extended, and others, is a new, concrete step that traditional SEO never asked for. Most sites haven’t checked their robots.txt file against this list, which means they’re invisible to the bots that feed ChatGPT, Perplexity, and Google’s AI Overviews. An llms.txt file, a plain-language guide for AI crawlers sitting at the root of your site, is quickly becoming a companion piece to your standard sitemap. It won’t take the place of good technical SEO, but skipping it means opting out of AI visibility.
- Content structure built for extraction. Traditional SEO wanted you to write for humans scrolling top to bottom. LLM SEO wants self-contained paragraphs, front-loaded answers, and a clean heading hierarchy that a model can lift a sentence or two from without losing the meaning. Write each section so it could stand on its own as the answer.
- Citation-worthiness over rankings. The goal isn’t “rank #1” anymore. It’s “get mentioned.” Original data, clear sourcing, and structured data like FAQ or How-To schema all raise the odds that a model cites your page instead of a competitor’s. The result is citation-worthy content that models want to reference by name. A single original data point, even something as small as a survey of your own customers, can do more for this than another rewritten listicle ever will.
This is what makes LLM SEO unique.
How to Actually Get Cited (or “Rank”) in ChatGPT
If you’re trying to figure out how to rank in ChatGPT, think about it differently. “Ranking” isn’t quite the right word, but the tactics are learnable.
Start with ChatGPT’s search feature. It draws partly from Bing’s index. If your team has never touched Bing Webmaster Tools they’re relatively easy to learn! Submit your sitemap, check for crawl errors, and verify your site the same way you would in Google Search Console.
PR placements, forum discussions, comparison sites, review platforms. All of these are inputs the models weigh when deciding who to cite. The more places your brand shows up in a credible context, the more likely a model treats you as a trustworthy source worth naming.
Freshness plays a role, too! Keep publish and update dates current and visible on your content. A model has to make a judgment call about which source is most current.
ChatGPT, Perplexity, and Google’s AI Overviews aren’t identical systems pulling from identical sources. Treating them as one target isn’t the best way to go about it. That said, this isn’t the time for a platform-by-platform breakdown. The tactics above give you a solid starting point across all three.
So if your team is stuck on how to rank in ChatGPT specifically, start with Bing Webmaster Tools, build branded mentions, and keep your content visibly fresh. Those three moves cover most of the ground.
Measuring What’s Working (Beyond Rankings and Clicks)
When a reader gets their answer from an AI summary and never clicks through to your site, your analytics don’t see a session. They don’t see a bounce. They don’t see anything at all. That reader existed, and your content may have influenced their decision, but your dashboard has no record of it. There’s no getting around that.
Here are some starting points to consider:
- Track referral segments from chat.openai.com, perplexity.ai, and gemini.google.com in your analytics. The volume is often small right now, but it’s trackable.
- Watch branded search volume over time. If people are searching your company name more often after an AI tool mentioned you, that’s an indicator of AI-driven awareness.
- Keep an eye on unexplained upticks in direct traffic. It’s likely coming from someone who read about you in an AI answer and typed your URL in directly, or searched your brand name afterward.
Set your expectations accordingly. Full attribution for AI-driven traffic isn’t solved industry-wide yet. But there are still ways to measure. Keeping an eye on certain signals, as already discussed, can be incredibly helpful.
The Bottom Line
SEO is far from dead. It’s just no longer one thing.
Before you decide what to prioritize, run a quick self-check:
- Is your site technically healthy: fast, mobile-friendly, and easy for any crawler to access?
- Does your content answer the questions your audience is asking?
- Have you built authority through backlinks, reviews, and brand mentions over time?
If you answered yes to those three, you’re already in a good position. Just add the new practices on top of that foundation. Open your site to AI crawlers, structure your content so it’s easy to lift and cite, and start tracking the new signals (even if they’re imperfect for now). That’s LLM SEO done right.
If you answered no to one or more, that’s fine too. It just means the fundamentals are where your budget and attention belong first. AI-visibility tactics built on a shaky foundation will crumble.
If you want a second set of eyes on where your site stands, that’s what we do at ArachnidWorks. I’m always happy to get a business’s ranking and visibility to a good spot. Reach out and we’ll walk through it with you.
