Contentful has launched its new Answer Engine Optimization (AEO) tool, Palmata. Why? According to Contentful CMO Elisabeth Maxson one thing stands out in CMO discussions today - CMOs are moving past the tool discussions and focusing on how they show up in Large Language Models (LLMs):
What's interesting here is a CMO no longer controls that strategic narrative, and it's been exposed in a whole different way. Your CEO could be asking about your brand (in an LLM), a question, or a product, and get an instant result, and then instantly send that to a CMO. Same with the board, so we're feeling this immense pressure to solve for it, but there's still this element that is so new, and you know who is the expert on that, and what are the right insights that we're surfacing.
More than a visibility tracking tool
There are so many AEO tools available today. Some are part of another platform, some are standalone. All claim to help organizations understand where their brand sits in LLM recommendations. And they all work to some degree.
You could lump Palmata in with these tools, but if you look more closely, you see that it does a lot more than just track answer engine visibility, rankings, and mentions. It is an AI discovery engine that helps you understand the factors that influence where they sit in AI-generated answers and provides guidance on how to improve their standing. It's full circle guidance.Maxson says:
There are so many tools that are popping up, and they provide you with, how are you showing up, and where are you showing up, but it kind of ends there. It's really on the surface level, and so for us, we really wanted to go below the surface. What are these much deeper insights that we can be providing to brands to make sure that they really understand their reputation, because it's this reputation that's getting formed by these LLMs that is contributing to that, and in some cases, that is content you control.
Palmata is powered by Contentful's Sounder Discovery Agent, which runs on OpenAI's ChatGPT and Google Gemini (with Anthropic's Claude on the roadmap). The agent continuously observes, learns, and synthesizes public data to understand the signals that influence AI-generated answers.
There are four components to Palmata:
If your brand has enough publicly available content for an AI system to generate an opinion about it, then Palmata can help you understand that opinion and improve it.
Palamata is a separate product from the Contentful digital experience platform, but there are plans to integrate it into the platform in the future. It's available as a tiered subscription, starting with a free tier that anyone can use (so you don't have to be a Contentful customer).
Empowering teams where they work
In addition to direct access to Palmata for AEO work, there is a model context protocol (MCP) server that enables teams to access the product's capabilities within the AI assistants they already use. Maxson says:
We've built this product very quickly, just in under five months, and we have been hand in hand with our content team, giving consistent feedback with our Head of Product, Jason McGhee. Same with myself, I've been deep in the weeds, I'm in the product almost every day, experiencing it, of what it means from a CMO perspective, from my team's perspective, and it's really helped shape the way that marketers can be thinking very differently in this entire AI world we're all trying to figure out.
Unlike other AEO tools that require you to create prompts to test, Palmata comes with over 450 prompts. More interestingly, these prompts (and others you create yourself) tell you not only what content was considered, but what was not chosen. Maxson provided the example of everyone saying they should focus on their Reddit strategy because it's the number-one-cited source in LLMs, but they see that this user-generated content isn't being chosen. Maxson explains:
We're able to show you the prompts of the topics of what is actually showing up in the LLM. This content was actually considered, but it wasn't picked. So understanding that at a deeper level, you could have maybe the right content that you have created, but if it's not chosen, it's not actually getting surfaced in the right way.
Palmata includes a deep-dive feature called "What's Boosting Your Brand and What's Working Against You" that shows you which content is boosting the brand, such as partner content, public relations content, the brand website, and so on. It also tells you if that content is hurting your brand. Maxson goes on:
Why this is so tangible for Marketing teams is we then provide very clear actions, and within those actions, we tell you why we're suggesting the action, what is the recommendation, and we have a simulated outcome, and so it simulates all the content changes, it reads out things to get rid of, it adds in green of things to add, and it completely generates that whole new content for you. So, in the future, when this is integrated fully into Contentful, imagine publishing that straight to your website.
Helping brands adapt to the changing nature of content operations
The brand website is still very important to a company, but it now needs to serve not only actual people, but also LLMs. Understanding how to do this is a question Marketers are asking, says Maxson:
I've been in meetings with customers where they've asked, 'Do I need a separate website for bots? Will my website even exist in the future?' What I think is interesting is there's all this background and structure that you need to be thinking about the website, so LLMs can easily crawl your website for the information that they are pulling and crawling for, but also still solving for a buyer. As it stands today, you can't purchase with an LLM. I'm sure e-commerce will completely change that, but there is still a human element, and we, as humans, we have emotions, we have human judgment, we have lived experiences, we have reactions, we have feelings. There's still this element of humanity, and honestly, that's actually what's really important when we built our own website, our own brand.
It's also clear that many organizations are still trying to figure out who owns AEO and who should have access to the product, she adds:
One thing that I think will be interesting as this all evolves is who owns AEO in a marketing organization. That is not clear. We have customers that it's their growth team or the web team. It is an SEO person turned AEO person, or it's their product marketing team. It's their head of content. Some don't have owners at all. I have one customer who their paid media person owns this as a stretch project. So that is really interesting, it's not having this home, yet still having this huge priority to be able to figure it out.
Even product management and sales teams can leverage the insights a tool like Palmata provides.
My take
As a Marketing consultant, I've looked at a few AEO tools to support clients, and I've covered a few of them here on diginomica (see HubSpot's AEO tool and BridgeEdge's AI Hyper Cube). They don't all work the same, and the inner workings of them are, for the most part, a black box.
What I like most about Palmata is the deep research it does to truly understand where a brand sits in AI discovery, and how you can focus that research in a specific area depending on your needs.
I also like that it shows you not only what content is working but also what isn't, and what changes you can make. Knowing how a brand ranks or what its visibility in answer engines is (as an educated best guess) is fine, but what to do about it is so critical and something we still aren't getting from most AEO tools.
Yes, there are suggestions to create content that will help, but that's such a small piece of the puzzle, and it doesn't go far enough. Palmata appears to go to the next level, and for brands struggling with their AI reputation, it's a product worth checking out.
