User behaviour is changing, so our strategies must change with them.
Before we dive in, we want to acknowledge that we’re simply sharing observations from working with 100+ eCommerce brands heavily invested in search.
There is no concrete evidence that suggests LLM (Large Language Model) tools like ChatGPT are driving anywhere near as much revenue as organic search.
While LLMs are on the rise, we want to make it clear that we are still entirely focused on the things that we see are making our clients the most revenue, and that is organic search from Google.
The purpose of this article is to help you and your team prepare for when they are make it into the mainstream.
Here we’ll dive into three core ideas:
- The difference between an LLM search and a Google Search.
- How the buyer’s journey differs from an LLM search vs Google Search.
- How Google SEO and LLM optimisation go hand in hand.
At the end of this article, you’ll also learn about an easy way to track your brand’s visibility across major LLMs.
What Is LLM Search and How Does It Differ from Google Search?
Tools like ChatGPT, Perplexity, and other LLMs are changing how people find and consume information.
A recent study shows that 34% of adults in the US who were surveyed now use LLM tools, with this number only looking to increase. We know there is an adoption lag from the US through to AU/NZ, so we can expect stats to make their way across the Pacific in the next 8-16 months.
The conventional search engine system was originally built for “transparency and trust,” but somewhere along the way, SEOs found ways to game the system through keyword stuffing, spammy backlinks, and other dodgy tactics that put a damper on the entire industry.
With LLMs, instead of returning a page with a bunch of blue links, users get information on brands and products not only based on what they “ask” in the moment, but also what they’ve asked previously, and in some cases, what the LLM predicts they might ask in the future.
Google search is binary: enter the search, get the result in the form of a link the user clicks on. This user journey means that Google is still king when it comes to click -> convert, since it takes people directly to a category / product page based on their intent.
LLMs are dynamic: ask a question, get a response in plain English (with maybe a few sources cited), and then have the ability to ask hyper-contextual follow-up questions to refine the answer. This user journey means that LLMs are starting to play a bigger role in how people discover and find out about products and brands, but we’re yet to see it significantly affect last click conversion.
This shift could fundamentally change how people find and consume information about brands and products. While it has not been adopted into the mainstream yet, we believe it is coming, it’s just hard to determine when.
Less Clicking, More Consuming
We’re assuming you’ve used an LLM before (and if you haven’t, go to ChatGPT or Perplexity and give it a try!)
You know how you ask a question, and it spits back information that feels natural and flows?
That’s by design! We’re at a point where humans just want the answer with the least effort possible.
Buyers want personalised responses and results based on what they’ve asked, without having to sift through websites.
Because of this, conventional ‘organic search’ where you search, click and then scan through a website could slowly lose its grip on the market (but it still isn’t here yet).
Buyers want information on demand, hyper-contextualised, from sources they trust.
With an LLM, a product or brand recommendation is nested within an answer to a user’s question or prompt, making it feel like a true recommendation from a trusted advisor… Not an ad or sponsored placement.
Since the playbook for how LLMs rank information is still a black box, nobody can reverse engineer the system to get their brand ‘to the top.’
Great for the user, they get un-manipulated results that match their intent. Much harder for brands trying to be mentioned in an LLM because there’s no clear-cut formula for being featured, only speculations and anecdotes.
Here’s Our Anecdote – The Fundamentals of Ranking on LLMs are Similar to SEO
In SEO, there’s a term called E-E-A-T:
- E = Experience
- E = Expertise
- A = Authoritativeness
- T = Trust
These have been the guiding principles in SEO for years.
Essentially, they come down to knowing what you’re talking about in a unique way, with adequate references to prove you’re not just making things up.
From our research and observations, brands that have fundamentally good E-E-A-T are showing up in LLMs.
They’re scanning for brand and product mentions across the entire internet, social media, forums like Reddit, and other corners where people gather.
They recommend brands and products with solid on-page setups (schema, internal linking, page context)…
But they’re also prioritising brands and products that best match user intent. This is based on how they ask the question and who they are.
Essentially, LLMs are far smarter at personalisation and understanding E-E-A-T because they understand human language (hence the name, Large Language Model).
How to Track Your LLM Visibility Using Ahrefs
While there’s no step-by-step, guaranteed way to consistently show up in LLMs (yet), there is a way to track how many times your brand or products are being mentioned in answers.
Ahrefs has introduced a new feature that allows you to see LLM mentions. Their new AI Citations feature is a game-changer.
The workflow is simple. Here’s how to access it:
- Log into your Ahrefs dashboard
- Enter the web domain you want to investigate
- Navigate to the AI Overview section to look for AI Citations metrics (on the left)
- Review which AI tools are linking to your domain
- Analyse the context and keywords associated with each citation
You can find out more on how to track your brand’s AI Search results with Ahrefs here.