When my traffic dropped last year (Feb 2025), I scoured 100s of websites, apps, ChatGPT sessions to figure out why. Talked to a few SEO agencies and they provided 'packages' to solve the issue, without explaining the issue. I totally ignored the website, and started posting on social media in June 2025. Did absolutely nothing with the website since then and the traffic came back after this year's core update (Feb 2026). Only thing I learnt is that there's nothing I can do to prevent the traffic from dropping again.
✅ Safe Reply
That feeling of helplessness is real, but traffic didn't come back by accident. Something in your site's authority shifted. Have you checked whether it was an AI visibility gain or pure ranking recovery?
Agencies selling packages without diagnosing first is exactly the problem. The traffic came back not because you did nothing, but because core updates re-evaluate existing authority. You have more control than you think.
Has anyone here actually lost traffic or signups because ChatGPT recommends a competitor instead of you? Curious how real this problem is.
✅ Safe Reply
It's very real. We're seeing sites lose 20-40% of informational traffic to AI answers. The harder signal to catch: ChatGPT recommending a competitor in a buying query. What niche are you in?
Short answer: yes, and most people don't even know it's happening because AI referrals show as 'direct' in analytics. By the time you notice, the competitor has been recommended 1000 times. What's your product?
Lately I keep wondering if we're heading into a zero-click internet. If AI gives the answer upfront, fewer people may ever reach the site. That changes the game a bit: not just how to earn traffic, but how to build a strategy where demand capture and brand visibility still work together. Still trying to figure out what that mix looks like in practice.
✅ Safe Reply
The shift is from 'rank for clicks' to 'be cited for trust.' AI models cite sources that demonstrate E-E-A-T and structured expertise. The brands showing up in AI answers are the ones with clear entity signals.
You're asking the right question. Most SEOs are still optimizing for 2019 Google. The real shift: citations matter more than rankings. If ChatGPT recommends your competitor by name, no amount of position 1 fixes that.
There was a stage where I genuinely thought I'd cracked it. The business was doing well. The team was big. The offices were big. The lifestyle looked elevated. Then one external change hit, traffic dropped, and suddenly the whole structure looked far more fragile than I'd admitted to myself.
✅ Safe Reply
This is the pattern with single-channel dependence. When one traffic source shifts, everything wobbles. Was it a Google update or something else that hit you?
The fragility sneaks up because growth masks it. Most businesses don't realize they're dependent on one channel until it moves. The fix isn't diversification alone, it's building visibility that survives algorithm changes.
A content audit recovered 50% of lost traffic. We didn't publish a single new piece. Here's exactly what happened. The site had over 200 published posts. Rankings had dropped 40% in 6 months. The team's response: publish more. That was the wrong answer.
✅ Safe Reply
Great breakdown. The 'publish more' reflex is so common. One thing to add: cannibalisation is even worse now because AI models pick ONE source per topic. If you have 5 pages on the same thing, none of them may be the one cited.
This is exactly right, and there's a layer most miss. AI answer engines don't just rank pages, they cite entities. Those 35 competing pages were probably confusing the AI about which one represents your authority on that topic.
This is more of a cost-cutting measure on their part given server maintenance IS expensive, and to consolidate both views as well as SEO things (which, if you're checking current events, views and clicks on websites have been gutted by shitty Google AI overviews).
✅ Safe Reply
The AI Overview traffic loss is compounding with the cost-cutting. Curious which niche you're seeing this hit hardest in?
AI Overviews gutting clicks isn't a bug, it's the design. The sites that survive this are the ones AI cites as sources, not the ones that just rank. Two very different games now.
Most SaaS founders think about content backwards. They start with 'what should we write about?' Flip it. Start with 'what's our buyer Googling at 11pm when they're frustrated with their current tool?' Those search phrases are your headlines. Those pain points are your SEO strategy. The content writes itself when you start there.
✅ Safe Reply
Spot on. Taking it further: those same pain-point queries are now what people type into ChatGPT too. If your content answers the 11pm frustration clearly, you show up in AI recommendations, not just search.
This is the right instinct but there's a next level. Those 11pm queries now go to ChatGPT, not just Google. If your content doesn't have clear, structured answers, AI skips you for a competitor who does.
Google AI Overviews are now absorbing 58% of clicks that once went to websites (up from 34.5% in April 2025). The zero-click shift isn't coming, it's here. Brands optimizing for traditional SEO metrics alone are missing the new visibility game: citations, brand mentions, and entity signals.
✅ Safe Reply
The 58% figure is striking. The gap between 'ranking well' and 'being cited by AI' is where most businesses are bleeding without knowing. What entity signals are you seeing move the needle most?
58% click absorption and rising. The uncomfortable truth: most SEO agencies are optimizing for a metric (rankings) that's becoming decorrelated from actual business outcomes. Citations are the new currency.