Sara & Nik’s Podcast:
Samy Ben Sadok

About this conversation

Today’s episode of Sara & Nik’s Podcast brings in Samy Ben Sadok, an SEO expert with 10 years of experience and a solo SaaS developer in his spare time. The topic of discussion: SEO, AI, and how humans are still the strongest link between them.

Samy lays it all out, from how he brought himself into the SEO field to learning about AI and how it operates, and, finally, to his GEO Toolbox platform, which took over his life over the past year.

The Road Into SEO and Why Testing Beats Pure Knowledge

Samy’s road into SEO was anything but straight. He started with computer science, then took a 2-year hiatus to gain his online poker expertise, after which he ended up in digital marketing and data analytics.

It only took a step to go from that to SEO and it came in the form of a question: “How do we get traffic?”, posed by his, then, web integration clients. Naturally, this emboldened Samy to start his research journey, which brought him to a clear-cut conclusion:

“I absorbed everything that was available online. But at some point that's not enough. You need to test it yourself.”

So, he started testing. It took him 4-5 years before he could replicate repeatable frameworks, which gave him the hands-on experience he needed to land a position at typing.com. It’s here where he learned how to grow traffic organically under the mentorship of someone who wasn’t even an SEO expert.

This first-hand experience with SEO and website growth taught him two valuable lessons. First is that testing and experimenting with your own project and ideas is critical for putting your knowledge to the test. Second is that the current AI landscape is too polarized.

Companies either fire entire teams to replace them with AI or deem AI useless and never touch it again. Samy doesn’t sit at either side of the spectrum. Instead, his position is more moderate:

“I think it's a great tool that can be used in specific cases but we should remember to keep our critical thinking.”

Creating AI Content That Actually Ranks

Once Samy had acquired sufficient SEO knowledge and tested it in the trenches, the next step came naturally: implementing it in the AI sector. In other words – write and rank AI content as effectively as human-written one.

The process was by no means easy. It involved studying the top-ranked pages, building a fact sheet to link every claim to its verified source, and working with several LLM models to figure out the right structure and approach. This took the time to write one article with AI to 1-1.5 hours + the time necessary to cross-check the article for any AI hallucinations, especially regarding raw data.

The fix he found came in the form of a cross-model QA: double check the article with multiple AI models to identify any discrepancies.

Samy then shares exactly what contributed to his English articles ranking so well with Google: his boots-on-the-ground research tactic. He went to forums and Reddit to find real pain points discussed by real people to include them in his article. This allowed his articles to find their target audience immediately.

Nik backed this talking point by explaining how he built a tool designed specifically for the same purpose: scrape forums and social media for people discussing industry-specific problems that he could then use in his own content creation.

Samy’s GEO Toolbox and What It Cost

Samy’s site started publishing in June, with the early traffic coming almost exclusively from LLMs. He’s now at 10,000 unique monthly visitors and almost 100,000 impressions, with an average position of around seven.

GEO Toolbox, which is the crown of Samy’s efforts, initially operated as a basic monitoring tool, tracking whether brands appear in AI answers. The platform then evolved to a more complex system, including a PR placement tracker for clients who wanted proof that they’re being cited by AI.

The problem, as Samy puts it, is that the built-in content studio isn’t as effective or optimal as his Claude Code skill is. On the other hand, the AI can’t handle all of the tasks on its own. The middle ground is the human using AI as a tool, which Nik immediately agrees with, explaining that that’s exactly how his business operates:

“I wanted to send a strong message. We are operators in the AI era, where we’re using AI to amplify quality and output for our clients. We are not using AI to replace the output.”

The closing has Samy admit that working on the project burned him out after only making $800 as compensation for 6 months of 14-hour work days and that a vacation is well-deserved. At this point, Sara’s advice hits the spot:

“You should start to monetize as soon as you can”, she said, in the idea that paying customers validate the project and almost always give feedback that matters.