A word like “date” can have a number of meanings – a romantic night out; a sweet, chewy fruit; a specific day of the year.
Without background and context, there’s no way to know what a word means. And context is an equally crucial part of engaging with AI tools and training them to run smoothly. If marketers’ tools aren’t trained with an understanding of previous performance and strategy, they won’t get very far in optimizing for future outcomes.
In order to fully “unlock” the benefits of AI, it needs to be trained on the context of a particular brand and the knowledge held by each individual member of the marketing team, according to Sandeep Menon, CEO and co-founder of marketing orchestration platform Auxia.
Auxia was developed to solve for that “context and intelligence layer” that AI needs to thrive, he said.
You got a problem?
Auxia set out to address three key marketing challenges. The first, said Menon, is that marketing is “inherently tribal,” meaning that every company has its own internal approach that can’t easily be transferred across brands or industries. Life-cycle marketing at Google, for instance, looks nothing like Uber’s strategy, he said.
Another issue – one that many companies are aiming to solve, particularly with the rise of the MCP – is the breadth of different tools marketers work with across platforms.
“They all do wonderful things, Menon said, “but they don’t talk to each other.” No individual employee has the context of each tool and task, either, so connecting the entire campaign workflow is another goal of Auxia’s.
And the third challenge is that marketing is a “multiplayer” experience, so any AI tool built for the industry needs to be accessible across teams.
One-stop shop
Auxia believes it’s on its way to solving these challenges with its end-to-end agentic marketing platform, which includes two products: Auxia Decisioning (a cross-channel tool that builds a unique profile for each customer to determine what ads they’re likely to respond to) and Auxia Agent Studio (its newer offering, launched in August, composed of a variety of agents that can measure campaign performance and execute changes).
Marketers aren’t satisfied with a tool that stops at the analysis step, said Menon, because then they’re left asking, “What do I do next?” Auxia connects to a client’s repository of existing content and data from external platforms like Figma and Salesforce. Connecting into these platforms enables Auxia’s agents to suggest next steps, like which assets might perform better for which audiences or what creative needs to be refreshed.
The agents are trained on a company’s own first- and third-party data and no company’s data is used to inform another’s decisioning. This approach addresses the tribalism concern off the bat, Menon said, and means brand context is fully accounted for and data privacy is upheld. If an employee switches to another team or leaves the company, their knowledge has already been built into the agent suite, so there isn’t a gap in the workflow, Menon added.
The agent studio, which includes specialized agents with roles ranging from real-time campaign optimization to bid tracking, also unifies a marketer’s toolkit in one place. Rather than having to keep track of dozens of different agents or tools and determine which one is the best for a given task, the agent studio serves as a central “control plane,” said Menon, which then determines which sub-agent is best suited for a task.
Auxia’s agents test and analyze performance on the user level, including their past behavior and engagement. Most of Auxia’s optimization is on its clients’ owned and operated channels (rather than large, third-party ad platforms like Google or Meta), so Auxia has a wealth of first-party data directly from clients.
Access to this real-time, first-party data allows Auxia to match the right ads to the right users. For instance, Auxia can target readers of a publisher client with ads related to the articles they’ve recently read. But it can also track other site behavior, like if the reader came onto the site via another ad, and use that to determine what sort of content resonates with them.
AI-ndustrial revolution?
Still, the industry is early along in the AI journey. Menon compared it to the rise of electricity in the late 19th and early 20th centuries: When electric motors were introduced into factories, he said, they initially just replaced steam engines with the new motors, rather than actually reformatting the factory structure.
When new technology emerges, productivity won’t get truly better until the process itself has changed, he said, rather than just making a one-to-one swap of old tech for new. But right now, he said, “a lot of what we’re trying to do is drive that transformation.”
For Auxia, that “transformation” means partnering with CMOs and helping them change the way their teams are structured, said Menon. The humans should be focused on strategic decisions, whereas the machines should be doing more manual, repetitive work.
Making the most of AI is “not a technology problem,” he added. “It’s an organizational problem.”
