SEO & Content Strategy
Google is quietly becoming more "YouTube-like" — rewarding personality, lived experience, and proof over polished, generic listicles. Here's what's changing in titles, media, and the technical side of AI-built websites, and exactly what to do about each one.
In this guide
For years, ranking well meant writing the most comprehensive, neutral, encyclopedia-style version of a topic. That formula is losing ground. Google has been steadily shifting its weight toward content that reads like it came from a real person who actually did the thing — not a summary assembled from other summaries.
This mirrors exactly what happened on YouTube years ago: a polished corporate explainer video routinely loses watch time and engagement to a creator filming themselves actually using the product, making mistakes, and giving an honest opinion. Google's search results are increasingly favoring that same energy — personality, proof, and lived experience over textbook-perfect descriptions.
Two forces are driving this. First, generic AI-generated content has flooded the web, making truly original, first-person content rarer and more valuable by comparison. Second, AI Overviews already summarize the generic, factual version of most answers — so the only reason left for someone to click through to your page is if you're offering something an AI summary can't: a real opinion, a personal test, a genuine story.
A title like "10 Best Marketing Tools to Buy" promises a generic, possibly AI-assembled list — the kind of content a reader has already seen a hundred times, and the kind an AI Overview can summarize without sending them anywhere. A title like "I Tested 10 Marketing Tools for 6 Months: My Honest Pros & Cons" promises something an AI summary cannot replace: a specific person's lived experience over a specific period of time.
That shift in framing is doing real work on click-through rate. The generic title sells information. The personal title sells proof. Readers have learned to be skeptical of "Best X" listicles that all say the same thing — but a specific timeframe, a named test, and the promise of honest downsides signal that the content might actually be different from the fifty other pages targeting the same keyword.
E-E-A-T — Experience, Expertise, Authoritativeness, and Trust — has existed in Google's quality guidelines for a while, but the "Experience" part is now carrying far more weight than it used to, and Google is increasingly able to verify it through media, not just claims of it.
Concretely, this means original images you actually took, screenshots of your own results, embedded video of you demonstrating something, and custom data or case studies from your own work are being treated as stronger trust signals than stock photography or purely descriptive text. Sites that embed relevant, self-produced video directly into blog posts — not just linking out to YouTube, but actually placing the video in context within the article — are seeing meaningful visibility lifts as a result.
The logic makes sense once you think about who this benefits: a stock photo of "a person typing on a laptop" proves nothing. A screenshot of your own analytics dashboard, a photo from your own client meeting, or a 90-second video of you actually using the tool you're reviewing proves you did the thing you're writing about. That's a much harder signal to fake at scale, which is exactly why Google is leaning on it more.
"Vibe coding" is the practice of building a website or app by describing what you want in plain language to an AI coding tool — like Claude Code, Cursor, Lovable, v0, or Replit — and letting the AI generate the structure, styling, and logic, often with little to no manual coding involved. It's become extremely popular for spinning up modern, visually polished sites fast, and Collins Dictionary even named it a word of the year in 2025.
The problem isn't the visual output — vibe-coded sites can look genuinely great. The problem is what happens under the hood. These tools are very good at making something that looks right in a browser, but they don't automatically set up the technical SEO foundation a search engine needs: clean canonical tags, a proper sitemap, a correctly configured robots.txt, and content that's actually present in the page's initial HTML rather than locked inside a JavaScript bundle that only renders after the page loads.
This isn't speculation — Google's own Search Relations team has addressed it directly. On the official Search Off The Record podcast, John Mueller and Martin Splitt both described testing AI coding tools for personal projects and running into the same gap every time: the tools produced functional, good-looking sites quickly, but getting the SEO fundamentals right still required giving the AI specific technical instructions from the start.
Mueller has also previously reviewed a vibe-coded project shared on Reddit and pointed out concrete, common problems: crawlability issues, outdated meta tags, and content stored entirely inside JavaScript files that search engines simply couldn't read. His broader conclusion is reassuring but specific: vibe-coded sites are not penalized just for being vibe-coded — Google has said they're essentially normal websites and can rank perfectly well — but they inherit every technical SEO risk of any heavily JavaScript-reliant site, because the AI tool won't fix that for you unless you explicitly ask it to.
Putting all three trends together, here's what to actually check before publishing or launching:
The common thread across all three trends is the same: Google is getting better at telling the difference between content that was actually lived and content that was merely assembled. That applies to your writing, your media, and now even the website itself. A modern, beautifully designed, AI-built site is genuinely fine — Google has said so directly — but it doesn't get a pass on the fundamentals, and neither does generic content just because it's well-organized.
The practical takeaway is almost reassuring: the advantage now goes to people and businesses willing to show real proof of their work, rather than the ones who can produce the most polished-sounding generic page. That's a harder bar to fake, but an easier one to clear honestly if you're actually doing the work.
I help businesses in Kerala build personal-first content strategies and technically sound websites that actually rank — whether vibe-coded or built from scratch.
Book a Free ConsultationSource note: Details on Google's vibe-coding guidance are based on statements from Google Search Relations team members John Mueller and Martin Splitt on the official Search Off The Record podcast, and prior public reviews by Mueller, as reported by Search Engine Journal and other SEO trade publications as of mid-2026. Google's guidance and ranking systems continue to evolve.
