I keep hearing about the “death” of SEO, and its metrics along with it. Rankings are dead. Traffic is vanity. Domain authority is worthless. Keywords are useless. Yes all these things have been touted across socials as the AI hype machine ramps up
The reality is far more complex.
Traditional SEO metrics haven’t died. But they’ve become insufficient as a standalone.
The traditional SEO measurement framework that worked when Google showed ten blue links. But now that struggles to capture total performance in a world where 60% of searches end without clicks.
The change happened gradually. No dramatic moment of realisation. Just a steady increase in search features that made our old dashboards incomplete. Until AI entered the fold.
The change has now accelerated at a rapid pace.
“GEO” has become the market’s favourite acronym. Some hate it. Some love it. Some just want to know what it is, how to influence, and better yet, how to measure it.
So here’s where I am at.
Table of Contents
The Impression Coverage Revolution
Impressions used to be a vanity metric. Now they’re becoming one of the most important leading indicators I track.
Here’s why:
- Impressions are the precursor to everything else.
- You don’t get clicks and revenue without impression coverage.
- In a zero-click world, impressions might be the only metric that matters.
- AI results lend themselves to more brand coverage.
- Building a brand = more eyeballs (impressions)
- Link building work = more eyeballs (impressions with a sprinkle of links)
When AI Overviews jumped from 6.49% to 13.14% of all queries in just two months, impression coverage became the first way to show value. If your impressions increase on targeted terms, that’s movement in the right direction.
My measurement framework starts simple:
- Google Search Console tracking query counts and impression coverage.
- As ranks increase, impression coverage increases.
- Those two metrics combined show the first sign of progress.
- Baseline AI ovevriew rankings
- Baseline ChatGPT mentions (although is is harder to track properly.
Impressions progress into clicks. Clicks progress into leads or revenue. That’s the chain that matters.
Understanding keyword research fundamentals becomes crucial when building this framework. People will talk about semantics, and entities, and passage retrieval etc. But all those things are themselves shaped by keyword research.
Why Domain Authority Became a Ghost Ship Metric
Domain authority exemplifies the problem with legacy metrics.
Links have been gamed for so long that authority scores no longer reflect actual value. Yet the indystry still bases so much of it’s link pricing off “DA”.
John Mueller made this crystal clear: “Google doesn’t use Domain Authority at all” for crawling, indexing, or ranking. Yet SEOs keep chasing these third-party scores.
Now, DA isn’t useless. It’s still valuable. But the point is DA as a standalone metric is not a true representation of a domain’s value. Much like clicks/sessions alone. See the trend?
I see plenty of high-authority websites that are essentially ghost ships.
They have impressive domain authority scores but have been penalised or abandoned and now rank for nothing and get no traffic.
On the surface, they look powerful, on paper but drive zero real value.
Instead, I also look at relevance, freshness, current rankings, and ranking velocity. These factors reveal whether a domain is actually valuable and active.
The Multi-Platform Reality
Traditional ranking position still matters. Organic position one remains arguably the most powerful marketing position available in my opinion, as it typically drives the most traffic and conversions for enterprise businesses.
But ranking happens across platforms now. You can rank in Google through YouTube videos, Reddit posts, LLM results or forum discussions. So SEO extends far beyond just websites and content.
This puts emphasis on having a multi-modal strategy.
Sometimes the answer is creating a video. Sometimes it’s being active on forums. The goal becomes total brand coverage through organic discovery.
This is what “GEO is, and this shift towards generative engine optimisation reflects how search is evolving, not dying.
I track these as “branded assets”.
YouTube videos, Reddit posts, forum mentions, ChatGPT citations. They’re all measurable through impressions, ranking positions, and extrapolated traffic.
When your Reddit thread ranks alongside your blog post, that’s expanded coverage.
When you’re citied in ChatGPT and that leads to a click, that’s expanded coverage.
When you’re surfaced in an AI overview but get no click, that’s still expanded coverage.
What rings true here is the old adage: “You have to be in it to win it”
Measuring What Actually Moves Business
The conversation with executives changes completely when you shift from traditional metrics to coverage metrics. In 2025, many executives have their eyes firmly set on AI search too, even if they don’t quite get what it entails.
Instead of “we rank for 500 keywords and got 10,000 sessions,”
also say,
“We improved our coverage in AI Overviews from 80 mentions to 100 mentions last month.”
Or, “Our total organic impression coverage increased even when we turned off brand ad spend.”
Or, “Our organic views on Reddit and YouTube grew by 10% MoM”
These metrics matter to businesses regardless of clicks.
The click becomes a bonus. You might get a click if you rank in an AI Overview, but you definitely won’t get one if you’re not there.
Coverage metrics show presence where your audience searches. That’s valuable even without direct attribution.
The Tracking Challenge
Measuring this expanded framework requires new tools. I use Ahrefs to track keyword sets where AI Overviews appear as search features. It shows where my brand or website appears in AI Overviews at scale.
ChatGPT tracking remains more challenging. The platform is still small compared to Google, and proper tracking doesn’t exist yet. ChatGPT also has memory, creating brand biases from previous conversations. That makes measurement incredibly complex.
Ahrefs also offers LLM tracking, which is a great starting point. It’s not perfect, no, but it’s something.
Google’s AI mode represents a similar challenge for measurement frameworks.
But at the end of the day, it’s too early to tell how valuable ChatGPT coverage actually is. Way too early. Google’s AI Overviews remain the priority because they’re measurable and significant in scale in an engine that still commands 90+% of search market share.
The Personalisation Problem
This measurement challenge will only intensify.
AI systems with memory and conversation history create personalised results that fragment the concept of universal rankings.
You can test an keyword in incognito, but LLM results don’t allow a similar thing.
How do you define “ranking” when results depend on individual user history? How do you benchmark performance when every search experience becomes unique?
I don’t have answers yet. Nobody does.
What I know is this: the businesses winning at SEO today track impression coverage, query count growth, multi-platform presence, and AI Overview mentions alongside traditional metrics to get a complete view of their brand.
They measure total brand coverage instead of just website performance. They understand that rankings matter across YouTube, Reddit, AI, and forums, not just their domain. If you’re looking to adapt your strategy, working with SEO specialists who understand these evolving metrics becomes essential.
The metrics aren’t dying. They’re multiplying. The question is whether your measurement framework can keep up.
