Speaking to Clients About AI Optimization Efforts + Evolving Landscape

[hannahmoen]

Hello everyone!

I’m curious how other agencies are navigating the impact of AI on SEO — both organic and local.

• How in-depth are you going with your AI strategies when discussing them with clients and potential clients?
• Are you leaning into specific tools you're using, measurable impacts on rankings and visibility, or more focused on the challenges of tracking things like Google’s AI Overviews?

I’d love to hear how others are approaching these conversations — what you prioritize communicating, how you simplify the technical side for non-marketing/data folks, and how you’re addressing the current limitations in tracking and reporting.

Internally, I feel like I can talk about AI endlessly, but externally it's much harder to distill the strategic adaptation into something digestible for clients who aren’t as deep in the weeds.

Curious how others are approaching this!

 

[Colan Nielsen]

This is a great question.

Here's a few thoughts on this:

1. Prioritize educating clients on how AI is impacting their business specifically (via data), as well as their industry as a whole
2. Always be running your own tests and experiments and make your clients aware.
3. Focus on actual business goals when having conversations with clients. Figure out what's important, such as sales/revenue and focus on that

 

[Ray Litvak]

Hello everyone!

I’m curious how other agencies are navigating the impact of AI on SEO — both organic and local.

• How in-depth are you going with your AI strategies when discussing them with clients and potential clients?
• Are you leaning into specific tools you're using, measurable impacts on rankings and visibility, or more focused on the challenges of tracking things like Google’s AI Overviews?

I’d love to hear how others are approaching these conversations — what you prioritize communicating, how you simplify the technical side for non-marketing/data folks, and how you’re addressing the current limitations in tracking and reporting.

Internally, I feel like I can talk about AI endlessly, but externally it's much harder to distill the strategic adaptation into something digestible for clients who aren’t as deep in the weeds.

Curious how others are approaching this!

I gently explain that moving forward, a client's site must be optimized for classic search engines (Google, Bing) and Emerging AI Search platforms, such as (ChatGPT search, Claude, Perplexity) and the like. If they disagree, show them stats that demonstrate the growth of AI Search, and perhaps one or 2 competitors that rank or outrank them on these new platforms. I break it down into SEO Serices and AIO Services, which can be a value added/additional serice offering.

 

[hannahmoen]

Thank you all for the feedback here, very helpful! Going to put together more internal insights to pair with what we know about AI Search more broadly and hope that bridges the gap!

 

[chadkimball]

at this point, one of the big traffic opportunities is to become a SOURCE for the LLMs to reference when giving their answers. If you start thinking like a SOURCE not as a search result page, it makes more sense. Here are some things Thomas Magee posted the other day related to this:

– Build long-form, clearly structured content that directly answers intent-driven queries, inject citations and references into external posts so AI connects your content across the web
– Get mentioned in forums like Reddit, Quora, and community threads where search starts
– Create simple knowledge docs and link them via popular AI indexers or tools

 

[ubertiger]

It's not public how the ranking in legacy search engines (Google, Bing, etc.) vs. LLMs (ChatGPT) exactly works. So, my opinion and experience is that when you rank well in Google, you have a good chance to be mentioned as source in ChatGPT. Often you see the following in ChatGPT where it is probably querying external sources or a copy of them:

Often, LLMs hallucinate, i.e. show false positives. I think that this is one of the main points LLM developers want to reduce to a minimum because the answers are given by the LLM itself and it hurts their reputation vs. Google which just show third-party sources for which it is not responsible.

Therefore, in my opinion trustable link sources (Wikipedia, Forbes, Wallstreet Journal, National Geographic, etc.) are way more important for being cited in LLMs compared to getting ranked in Google.

I do not have much experience with clients to be AI educated. I just think that relabeling SEO services as AI Search / GEO packages and trying to sell them something which is basically the same is misleading.