How to talk about GEO to clients
Right now, nearly every call I take involves an agency under pressure. Their clients are anxious about “falling behind” in the AI race, and they’re turning to their SEO teams for answers. That creates a lot of stress — but also an opportunity. This article is designed as a resource for how to respond to those client questions, and how Waikay can provide both short‑term clarity and long‑term strategy.
1. Reassure them — but stress that this matters
The first step is to calm the panic. Generative SEO (GEO) is a long game. No one has all the answers yet, and your clients haven’t already lost. What they need to hear is that you are more than capable of building a strategy once you have the right tools in place.
GEO is about building trust and brand authority in a way that machines can understand, then refining that strategy as real data emerges. To do this, you’ll need a tool that can turn AI’s messy, inconsistent outputs into actionable insights — not just pretty visualizations.
2. Explain that the strategy is different from before
Clients may push for “numbers” on how often their brand appears in AI answers. This comes from early tool vendors who marketed “AI search volume” without explaining what it really meant.
The truth: there are no reliable metrics for how many times a brand is mentioned in AI responses. What those tools provide is an approximation, based on thousands of prompts and statistical modeling. It’s expensive, imprecise, and not a meaningful measure of impact.
Instead, your strategy should focus on identifying patterns in AI outputs — entities, citations, competitors — and using those insights to shape a holistic approach.
3. Set expectations about progress
In traditional SEO, visibility often improves gradually over time. AI doesn’t work that way. With LLMs, you’re either visible and well‑understood, or you’re not.
Because the same question can produce five different answers, you’ll need to rely on month‑by‑month averages to measure presence and impact. The goal is to establish a strong, consistent presence in AI responses, not to chase a smooth upward curve.
4. Clarify the difference between training data and research history
This is critical. Many clients confuse the two.
- Training data: the information an LLM was trained on before it was released.
- Research history: the personalized results an LLM generates after repeated use, influenced by web access and user preferences.
If a client sees their brand appearing in AI responses, it doesn’t necessarily mean they’re in the training data. More often, it’s the model pulling from the web or tailoring results to the user’s history. The real goal is to shape what AI “knows” and “says” about your brand at the training‑data level — that’s where long‑term influence lies.
5. Reinforce that this is bigger than SEO
AI visibility isn’t just an SEO issue. It touches PR, brand reputation, finance, HR, and leadership. For example, when hiring, candidates may use AI to check your brand’s reputation. If that narrative isn’t monitored and managed, it can create real risks.
AI optimization is a holistic effort. Positive press, strong brand stories, and reputation management can influence AI outputs just as much — if not more — than traditional content strategies.
6. Find a tool that provides them with real insights on day 1
I honestly would not recommend Waikay at the end of this blog post if I didn’t think it was a great, well priced way for you to provide value to your customers from day 1. Here’s a link to a post I’ve written on how to do that.
