Every time I hear that SEO doesn't convert because paid is more efficient.
Eli Schwartz
These are the best posts from Eli Schwartz.
9 viral posts with 4,135 likes, 366 comments, and 306 shares.
4 image posts, 0 carousel posts, 0 video posts, 5 text posts.
👉 Go deeper on Eli Schwartz's LinkedIn with the ContentIn Chrome extension 👈
Best Posts by Eli Schwartz on LinkedIn
Prediction: More companies will hire "SEO Engineers" but not update the salary band.
Companies are happy to slap "engineer" on an SEO job title when it helps with recruiting. It sounds technical, and it signals to the world how AI-first they are.
But the paycheck will still look like a coordinator role.
- SEO Specialist average: $55k-$86k/year
- Mid-level Software Engineer: $89k-$110k/year
- Senior Software Engineer: $100k-$260k+/year
- Entry-level gap between the two: $30,000+
You can't have it both ways. Either SEO is a technical discipline that requires engineering-level thinking, or it's not. Pick one.
I think the SEO role should be up-leveled and paid appropriately.
Companies are happy to slap "engineer" on an SEO job title when it helps with recruiting. It sounds technical, and it signals to the world how AI-first they are.
But the paycheck will still look like a coordinator role.
- SEO Specialist average: $55k-$86k/year
- Mid-level Software Engineer: $89k-$110k/year
- Senior Software Engineer: $100k-$260k+/year
- Entry-level gap between the two: $30,000+
You can't have it both ways. Either SEO is a technical discipline that requires engineering-level thinking, or it's not. Pick one.
I think the SEO role should be up-leveled and paid appropriately.
There are tactical steps to improve a site or brand's visibility in LLMs, but most of what I see companies doing in practice fits this meme.
If a company already has high search visibility due to a strong brand and effective SEO, it will likely have high visibility in LLMs with little additional effort.
Based on my experience with many brands, the best way to improve LLM visibility is through brand marketing, NOT the AEO checklist sold by AEO agencies, AI-generated content, or Reddit astroturfing citation-building campaigns that many companies are investing in.
If a company already has high search visibility due to a strong brand and effective SEO, it will likely have high visibility in LLMs with little additional effort.
Based on my experience with many brands, the best way to improve LLM visibility is through brand marketing, NOT the AEO checklist sold by AEO agencies, AI-generated content, or Reddit astroturfing citation-building campaigns that many companies are investing in.
LLM visibility and AEO tool promoters have taken the spam crown from backlink sellers.
Same playbook, new product.
Cold DMs warning me that my brand is "invisible to AI" unless I sign up for their vibe-coded tool have taken the place of the Google Sheet of high-DA backlinks I must buy.
LinkedIn posts with screenshots of screenshots of graphs with no X & Y access.
Comments on my LinkedIn posts telling me how wrong I am with the job title of the commenter being something along the lines of "I help brands show up on ChatGPT" or "Founder of AEOGEOLLMO tool for LLM visibility."
It's the 2012 backlink seller era, just wearing a different hat.
The fear-based pitch is identical:
- Your competitors are already doing this
- Google (now: ChatGPT) will bury you if you don't act
- Buy our tool before it's too late
Back then, it was domain authority scores and private blog networks. Now it's "LLM citation tracking" and "AI search share of voice."
Some of these tools are useful if they have proprietary data or a methodology, but most are just vibe-coded prompt scraping with a UI. The space is real, and the problem is real, but much of what's hitting inboxes right now is manufactured urgency wrapped around FOMO.
The best SEOs in 2012 ignored the backlink spam and focused on building things worth linking to.
Same answer applies now.
Same playbook, new product.
Cold DMs warning me that my brand is "invisible to AI" unless I sign up for their vibe-coded tool have taken the place of the Google Sheet of high-DA backlinks I must buy.
LinkedIn posts with screenshots of screenshots of graphs with no X & Y access.
Comments on my LinkedIn posts telling me how wrong I am with the job title of the commenter being something along the lines of "I help brands show up on ChatGPT" or "Founder of AEOGEOLLMO tool for LLM visibility."
It's the 2012 backlink seller era, just wearing a different hat.
The fear-based pitch is identical:
- Your competitors are already doing this
- Google (now: ChatGPT) will bury you if you don't act
- Buy our tool before it's too late
Back then, it was domain authority scores and private blog networks. Now it's "LLM citation tracking" and "AI search share of voice."
Some of these tools are useful if they have proprietary data or a methodology, but most are just vibe-coded prompt scraping with a UI. The space is real, and the problem is real, but much of what's hitting inboxes right now is manufactured urgency wrapped around FOMO.
The best SEOs in 2012 ignored the backlink spam and focused on building things worth linking to.
Same answer applies now.
No, Google search is not suddenly going away. Everything Google announced at I/O has already been happening, so there won't be a sudden decrease in traffic.
However, the new features Google shared, which may eventually launch, do show that Google is developing new ways to keep people in search results rather than clicking off to websites.
The hot takes on social media will create strong demand for websites that will still need creative marketers to drive what will be a more limited pool of clicks from both traditional search and LLMs.
Here's what you should not do right now:
➡️ Do not suddenly run out and buy an LLM tracking tool. Most of the tools have inaccurate data and will certainly not help you get more traffic.
➡️ Do not give up on SEO, there will always be traffic from the ten blue links, it's less than it used to be, but giving up means you will now have zero.
➡️ Do not fire your SEO team because they don't have the answers you think they should have on AEO. Empower them to get the answers. Starting over with a new team (and a very limited talent pool) guarantees that you will be behind your competition.
➡️ Do not start chasing a tactic for LLM visibility that sounds too good to be true. Spam is spam, and it's not worth the complete loss of traffic when it gets nailed.
➡️ Do not assume that your users are only going to use LLMs to find your vertical. Who uses LLMs and how they use them are very category-dependent.
✅ Do read my newsletter from this week on what I learned attending I/O https://lnkd.in/efX9-eag
However, the new features Google shared, which may eventually launch, do show that Google is developing new ways to keep people in search results rather than clicking off to websites.
The hot takes on social media will create strong demand for websites that will still need creative marketers to drive what will be a more limited pool of clicks from both traditional search and LLMs.
Here's what you should not do right now:
➡️ Do not suddenly run out and buy an LLM tracking tool. Most of the tools have inaccurate data and will certainly not help you get more traffic.
➡️ Do not give up on SEO, there will always be traffic from the ten blue links, it's less than it used to be, but giving up means you will now have zero.
➡️ Do not fire your SEO team because they don't have the answers you think they should have on AEO. Empower them to get the answers. Starting over with a new team (and a very limited talent pool) guarantees that you will be behind your competition.
➡️ Do not start chasing a tactic for LLM visibility that sounds too good to be true. Spam is spam, and it's not worth the complete loss of traffic when it gets nailed.
➡️ Do not assume that your users are only going to use LLMs to find your vertical. Who uses LLMs and how they use them are very category-dependent.
✅ Do read my newsletter from this week on what I learned attending I/O https://lnkd.in/efX9-eag
Some SEO teams genuinely thought they'd cracked the code.
AI-generated content at scale pumping out thousands of pages.
Programmatic everything.
They called it a strategy, but it was closer to machine translation with a keyword list.
Glenn Gabe calls this Mt AI, and I love this term.
Google's updates didn't just slow them down. It wiped out years of work in a few algorithm cycles. Traffic gone.
The smug case studies quietly deleted.
The question they are asking is: how do we recover?
Wrong question.
You can't recover by doing more of the same thing that got you penalized. Publishing "better" AI content isn't a pivot. It's just slower self-destruction.
The actual pivot looks like this:
- Stop asking what keywords you can rank for
- Start asking what problems your users actually have
- Build content around intent, not search volume
- Write things people would share even if Google didn't exist
Don't try to outsmart the algorithm. Focus on being genuinely useful to real people, and the traffic will follow.
Nobody wanted to hear this when the content farm was printing traffic.
Finally, they're hearing it now.
AI-generated content at scale pumping out thousands of pages.
Programmatic everything.
They called it a strategy, but it was closer to machine translation with a keyword list.
Glenn Gabe calls this Mt AI, and I love this term.
Google's updates didn't just slow them down. It wiped out years of work in a few algorithm cycles. Traffic gone.
The smug case studies quietly deleted.
The question they are asking is: how do we recover?
Wrong question.
You can't recover by doing more of the same thing that got you penalized. Publishing "better" AI content isn't a pivot. It's just slower self-destruction.
The actual pivot looks like this:
- Stop asking what keywords you can rank for
- Start asking what problems your users actually have
- Build content around intent, not search volume
- Write things people would share even if Google didn't exist
Don't try to outsmart the algorithm. Focus on being genuinely useful to real people, and the traffic will follow.
Nobody wanted to hear this when the content farm was printing traffic.
Finally, they're hearing it now.
Investors are pouring money into "SEO replacement" tools right now.
And almost none of them have talked to an actual SEO manager.
I keep hearing the same pattern from VC diligence: a founder pitches "SEO is dead, AI killed it, here's what comes next," and VCs nod along because they've read the same three LinkedIn posts everyone else has.
But when you ask if they've done any diligence with people who actually run SEO channels within the companies that would be the supposed customers, the answer is almost always no.
Here's what SEO managers, the people who will use these tools, will actually tell you:
- Organic traffic is down in some categories, not all
- Brand search is holding up fine
- Bottom-funnel content still converts well
- The channel is changing, not disappearing
The "SEO is dead" narrative is being built by people who are trying to sell something new, whether it's their SaaS or an agency offering.
Believing that is not diligence, it's vibes.
If you're investing in a business whose entire thesis depends on SEO being dead, please spend 30 minutes with someone whose job is to keep it alive. I am happy to make introductions, and I do whenever I join one of these diligence calls.
And almost none of them have talked to an actual SEO manager.
I keep hearing the same pattern from VC diligence: a founder pitches "SEO is dead, AI killed it, here's what comes next," and VCs nod along because they've read the same three LinkedIn posts everyone else has.
But when you ask if they've done any diligence with people who actually run SEO channels within the companies that would be the supposed customers, the answer is almost always no.
Here's what SEO managers, the people who will use these tools, will actually tell you:
- Organic traffic is down in some categories, not all
- Brand search is holding up fine
- Bottom-funnel content still converts well
- The channel is changing, not disappearing
The "SEO is dead" narrative is being built by people who are trying to sell something new, whether it's their SaaS or an agency offering.
Believing that is not diligence, it's vibes.
If you're investing in a business whose entire thesis depends on SEO being dead, please spend 30 minutes with someone whose job is to keep it alive. I am happy to make introductions, and I do whenever I join one of these diligence calls.
Many think AI content is the new silver bullet for SEO.
It's not. We've been here before.
Almost two decades ago, there was a WordPress plugin called Caffeinated Content. It scraped RSS feeds, swapped words with random synonyms, and published the result as "original" content. This process, called "spinning content," was completely unreadable. Genuinely useless, but it worked. It was a sort of manual, find-and-replace version of today's AI content.
I used it and, of course, layered on the gray hat backlinking that was popular at the time. I made real AdSense money from sites designed just for this.
In 2009, I got a cease and desist from Sarah Palin's team because one of my junk content sites was outranking her official pages in search results. That's how broken the system was.
Then Google launched Panda in 2011 and wiped it all out overnight. Useless content was certainly not rendered obsolete, but the programmatic versions were greatly reduced until the rise of AI-generated content.
The pattern we are seeing now is a recurrent theme in gaming SEO, and its end will be the same.
- Someone finds a shortcut that games the algorithm
- It works, so everyone piles in
- The web fills up with garbage
- Google (or whoever) nukes it
- Everyone acts surprised
- The media writes an article about how bad Google is and then points to their own sites as victims
AI content is just the current version of this cycle. It's faster, cheaper, and more convincing than spinning content and $5 content on Upwork ever was. Nevertheless, the underlying logic is identical: produce volume, capture rankings, collect clicks.
And yes, it's working right now for a lot of people.
But Google is already adjusting, and when the correction comes, it won't be gradual.
The only content that survives these resets is content that actually helps someone.
Not content written for a crawler or LLM.
Not content planted on Reddit.
Not content with an automated agent that creates to match your competitors.
Not content with the perfect schema.
Not content broken up to match prompts in FAQ's.
Content that a real person, or increasingly their AI assistant, finds genuinely useful.
This advice was true in 2011, and it remains true in 2026.
It's not. We've been here before.
Almost two decades ago, there was a WordPress plugin called Caffeinated Content. It scraped RSS feeds, swapped words with random synonyms, and published the result as "original" content. This process, called "spinning content," was completely unreadable. Genuinely useless, but it worked. It was a sort of manual, find-and-replace version of today's AI content.
I used it and, of course, layered on the gray hat backlinking that was popular at the time. I made real AdSense money from sites designed just for this.
In 2009, I got a cease and desist from Sarah Palin's team because one of my junk content sites was outranking her official pages in search results. That's how broken the system was.
Then Google launched Panda in 2011 and wiped it all out overnight. Useless content was certainly not rendered obsolete, but the programmatic versions were greatly reduced until the rise of AI-generated content.
The pattern we are seeing now is a recurrent theme in gaming SEO, and its end will be the same.
- Someone finds a shortcut that games the algorithm
- It works, so everyone piles in
- The web fills up with garbage
- Google (or whoever) nukes it
- Everyone acts surprised
- The media writes an article about how bad Google is and then points to their own sites as victims
AI content is just the current version of this cycle. It's faster, cheaper, and more convincing than spinning content and $5 content on Upwork ever was. Nevertheless, the underlying logic is identical: produce volume, capture rankings, collect clicks.
And yes, it's working right now for a lot of people.
But Google is already adjusting, and when the correction comes, it won't be gradual.
The only content that survives these resets is content that actually helps someone.
Not content written for a crawler or LLM.
Not content planted on Reddit.
Not content with an automated agent that creates to match your competitors.
Not content with the perfect schema.
Not content broken up to match prompts in FAQ's.
Content that a real person, or increasingly their AI assistant, finds genuinely useful.
This advice was true in 2011, and it remains true in 2026.
The AI mode tsunami is coming.
Google just announced monetization for AI Mode.
They're rolling out three revenue streams:
1. Ads integrated directly into AI responses
2. Direct product offers from retailers
3. Universal Commerce Protocol for seamless transactions
Translation: AI Mode is becoming a shopping mall.
This might make AI mode even more valuable to Google than traditional search, so the chances of AI mode becoming the default just got more likely.
See the link to the post in the comments
Google just announced monetization for AI Mode.
They're rolling out three revenue streams:
1. Ads integrated directly into AI responses
2. Direct product offers from retailers
3. Universal Commerce Protocol for seamless transactions
Translation: AI Mode is becoming a shopping mall.
This might make AI mode even more valuable to Google than traditional search, so the chances of AI mode becoming the default just got more likely.
See the link to the post in the comments