Given the SEO trends for 2024 and the likely introduction of AI (SGE) features to Google we will probably witness way less website traffic.
Searchers will stay on Google and consume your content summaries there.
People may even buy your products or services without visiting your "Web site".
So how do we measure online success then? I have a few ideas here:
I also welcome additional suggestions from others I may include in the post itself.
You can also get quoted in the article when you confirm my existing choices.
Industry icon Wil Reynolds has written a thorough article on why legacy SEO metrics like traffic and rankings don't work in the AI SEO era: https://www.seerinteractive.com/insights/why-2020s-seo-kpis-wont-work-in-2024-in-a-genai-data-scarce-world
I hope you believe me now haha.
I haven't wrapped my head around a solution yet, but this sounds a lot like the dawn of streaming and DVR-type of replays. TV metrics were often deflated because streaming devices either weren't being counted, or couldn't be properly early on. Comscore was God back then (and often still is), but was waaaay behind in including streaming numbers in performance. I've been out of the TV world for a while, so I haven't kept up with its improvement, but when we see things like "Super Bowl 20 (back in the 80s) was the most watched Super Bowl of all time," it's probably not true, and more recent numbers over the last 15 years have failed to properly account for streaming. In turn, when Google starts not forcing people to even visit your website to complete a purchase, or they can get a summary of results right on the SERP (ditto for, say, a search result showing up in ChatGPT that doesn't require you to go directly to a website), numbers will understandably plummet. If we in the internet age are as slow on the uptick as television was to account for these numbers, we're going to be massively in the dark and missing TONS of insights.
Hopefully, people smarter than me are ahead of the curve on this already.
If (when) we get to a point where most search engines don't even send traffic to websites anymore, then it falls on those search engines to provide better analytics on what users are doing while on their search pages.
We currently receive Search Console data about Google usage, but that's limited, anonymized, and lacks the granularity to allow for proper insights.
Then you have Google Lens integrated into the search but, again, there's little data around that if users don't make it to a website. If you depend on Image Search quite a bit, you need separate Lens data as it's being used a lot from there.
Bing doesn't even split their traffic into search facets, so it's even worse data to make decisions off of.
If search engines decide to change the flow of users in a way that can affect so many businesses and livelihoods, then they should also assume the responsibility of offering better analytics to the websites that feed their AI.
Hello Tadeusz,
As far as metrics go, specifically for AI, Bing is currently providing Chat visibility metrics but it's still combined with web. There's no specific information about queries or pages. Bing stated that "Keywords and Pages data is only available for Web traffic and not applicable to Chat and other verticals." I'm sure they will eventually separate them and be able to see clicks and a visibility measure.
I am curious what Bing and Google SGE will report on. I'd like to see the following:
Ranking: If there's a way to order the content they pulled from such as a primary reference, supporting reference 1, supporting reference 2, etc.
Type of content, integrated media: Whether it's textual, image, video, file, etc.
Queries: This might be more of a challenge if people are using more conversational type queries.
Follow-up queries (some type of engagement): If additional queries in a conversation thread with the LLM are focused on your content, or if the LLM continues to pull from your content, could there be some type of engagement metric
These are just a few I've been thinking bout. What else would you like to see?
Hola Tadeusz! I have written and talked a lot about this topic since when Rand Fishkin started popularizing the definition of 0 click SERP at The Inbounder quite a few years ago.
Sincerely, I share the opinion of @Aleyda Solis here, and very convinced that what us Italian SEOs call "Liquid Google" (and more and more "Liquid Search") is something that already existed before SGE, and that responds to a natural evolution that f how people search solutions online.
Sure, we cannot deny the necessity Google has to make people stay on its properties, as it was an average news site and so increasing the possibility to show them ads and the clicks on them.
But that's not the real reason (in fact, emails showing the existence of the insatisfaction of the Ads Team because of this behavior of the SERPs started to be shared).
The fact is that more and more people naturally tend to not conform themselves with just a query followed by a click, and they dig dig dig until they find what they do really are looking for.
This is made inside the Google ecosystem (Search, YouTube, merchant listings, Maps...) and outside Google (start in TikTok, stop by in YouTube, passing from Etsy, having a concersation with ChatGPT or Perplexity, and then going to Google with a very long tail search).
The Search Journey is much more diluted in many steps and the Search Session became longer.
So, the 0 click SERPs in reality are just a symptom of this behavior, not the real problem.
Concluding... IMHO SEO is not anymore just about "Search" meant as Universal Search of Google, but about visibility along all the search journey independently of where the search journey is taking place.
That's why, even if some facets of SEO will remain or even will have a stronger relevance (i.e.: technical SEO), others as semantic search (aka optimizing graphs as the shopping graph) and entity search (i.e. Entity Salience is becoming more important than pure keyword targeting) are going to be our main commitment.
Hi Tadeusz, I don't think that the SGE will end up "stealing" traffic from sites offering products/services directly, other than for highly informational/factual queries as I see the SGE as a featured snippets in stereoids. For commercial/transactional/navigational queries it will refer them to the site where users will end up converting. In some cases, the traffic will shift from one page type to another, but in order to fulfill the user need they need to refer it. I share more about this in this article here: https://www.aleydasolis.com/en/search-engine-optimization/the-3-main-types-of-google-sge-snapshots-and-their-level-of-ranked-pages-traffic-risk/
On the other hand, if eventually we end up in the scenario in which they surface your products/services within their interface directly without referring the user to your site in a way that is satisfying, that "organic /LLM transaction buying your product/service" happening through their interface should be the way to measure success, I guess. I don't see this happening easily though, without alienating entire industries and killing the incentive to create / publish content which is also a key part of their business model as a search engine too.
I hope this helps :)