Lily Ray on why shortcuts rarely win in SEO or AI search

Early in Lily Ray’s SEO career, her CEO offered her a deal: Get the company to No. 1 in Google, and he’d buy her a drum set.

This was around 2011. Ray had discovered that buying links worked. Really well. So she kept buying them.

Then came Penguin.

• “I think it was August 2012. Went to work that day and all of our rankings disappeared,” Ray said.

The company had gone from ranking second, third, or fourth for nearly every keyword it cared about to being gone from the search results.

• “That was my first experience with algorithm updates, and that probably informed a lot of how I approach SEO to this day.”

More than a decade later, Ray is watching people experiment with generative AI, automated content, and GEO tactics.

• “Everyone’s looking for shortcuts, and I get it. There’s a lot of pressure right now, a lot of excitement with AI, and a lot of people want to see really fast results.
• “I actually think trying to drive results quickly is one of the worst things you can do unless you’re only focused on results for the next year or two.”

From Penguin to E-E-A-T

Ray stumbled into SEO around 2010 while studying politics and Spanish at New York University. She found her first SEO job on Craigslist.

Growing up near Silicon Valley with a software engineer father, Ray had learned about computers and HTML as a kid. SEO quickly showed her she could produce measurable results.

She added meta keywords, picked up links from sites such as Reddit, StumbleUpon, and Digg, and watched traffic rise.

• “I can actually have an impact,” she remembered thinking.

Ray spent her first several years largely outside the SEO conference community. When she eventually started attending events, she saw stages dominated by older men and didn’t initially consider speaking herself.

It took until around 2017 or 2018 for that to change.

• “I think it took until probably 2017 or 2018 to even consider that I had anything important to say.”

She was terrified of public speaking. Years later, she would keynote MozCon, a conference she had once attended while looking up to speakers such as Rand Fishkin and Cindy Krum.

• “You’re more valuable and have more valuable ideas and thoughts than you think you do. And other people want to hear them.”

Her approach to SEO was changing, too. After Penguin, Ray moved toward what she considered safer fundamentals: title tags, headings, image optimization, internal linking, and other on-site work.

Then the 2018 Medic update hit one of her agency’s clients hard.

Ray spent months trying to understand what Google was doing, digging into its Search Quality Rater Guidelines and E-A-T – expertise, authoritativeness, and trustworthiness, which Google later expanded to E-E-A-T with the addition of experience.

E-E-A-T, Ray said, doesn’t work like optimizing a title tag or Core Web Vitals, where you have something more defined to work with.

• “You don’t have like, ‘I have good E-E-A-T.’ It’s not just what you say on your website, it’s how other companies and brands and everybody evaluates – and reviewers and users evaluate – your brand.”

That includes how well-known and trusted a brand is and what customers say about it – factors Ray is now examining in AI search.

The helpful content update changed her approach

The helpful content update presented Ray with a different problem.

She spent hundreds of hours studying affected websites and, at the time, was vocal that she thought Google had gotten parts of the update wrong.

Many sites that were taking advantage of Google’s algorithms deserved to lose visibility, she said. They had generic content, stock photography, and little evidence that real people were behind the work.

But Ray also saw recipe bloggers, travel bloggers, and other site owners who appeared to be following much of Google’s guidance get hit.

• “I would say like maybe 40% of the sites that they went after were really trying to do things right.
• “I talked to so many dozens of business owners who lost literally their entire livelihoods,” Ray said.

That experience continues to influence how she approaches SEO and GEO.

• “I don’t want anybody to get caught up in that again this time. If we see like the GEO atomic bomb update and anybody that’s been experimenting with GEO stuff has their site removed from Google Search – like, it sounds extreme, but there’s precedent for this.”

Less is more

Ray is particularly wary of using generative AI to scale content production.

Automated workflows can make it tempting to publish enormous amounts of content, she said, especially when you believe your particular AI workflow produces something better than everyone else’s.

But much of that content ends up “just saying what everybody else has already said.”

Her approach today is more conservative.

• “Less is more. I don’t think that the best way to achieve SEO success right now is to create as much content as possible.”

Instead, she favors creating a smaller number of valuable assets, updating them over time, and building a presence beyond a company’s own website.

• “The sites that try to create as much content as possible eventually get burned, including very big brands.”

AI search has its own spam problem

AI platforms are already confronting low-quality content.

Ray said there is plenty of “spammy low-quality stuff” being cited by ChatGPT, while generative AI has made producing SEO spam easier than ever.

She also saw early signs of AI platforms trying to improve their results. ChatGPT, for example, uses fan-out queries that restrict searches to specific domains.

• “I think this is going to be a really, really big challenge for them,” Ray said.

She also expected the tactics that work today to change.

• “Every single time we’re in this situation where there’s a lot of spam and it’s working really, really well, they come up with some solution where they crack down on it in a really big way,” Ray said of Google.

Good SEO still matters in AI search

Ray strongly rejects another argument that has emerged with generative AI: that you can abandon SEO and focus exclusively on AI search.

She called that idea “completely blatantly false.”

• “If you understood how AI search works, you would understand that SEO is more important than ever. The single best and most sustainable and important thing you can do for AI search visibility is good SEO.”

What Ray considers success, however, extends beyond traffic.

For AI search, she said, winning means being recommended as the best or most relevant brand for the topics, products, and services a company wants to be known for.

And she doesn’t mean publishing listicles that declare your company the best.

• “I think it’s earning that placement or that recommendation naturally, organically,” Ray said.

Over time, she expects LLMs and search engines to get better at demoting attempts to manipulate those recommendations.

• “I think that the ones that really truly are the best and that have been recommended as the best by their customers and everything over the years are the ones that are going to win.”

The next 20 years

Ray doesn’t pretend to know what search will look like 20 years from now. AI is changing too quickly for that.

But she expects companies will still need to put information into whatever systems exist, and brands will still look for ways to optimize how they appear within them.

• “There’s always going to be manipulation and spam and everybody trying to hack the system.”

Ray ultimately brought the conversation back to quality:

• “What if you were to build a system that was trying to truly evaluate what the best service is? How do you make your service align with that?”