Built for an AI-powered future
For over a decade, Ometria has been the leading customer data platform for enterprise retail brands including Sephora, Boden, Foot Locker, and two hundred more around the world. As a CDP (Customer Data Platform), we are the central infrastructure that connects hundreds of datasets to create detailed omnichannel identities for more than one billion people.
We help marketers understand what their customers are browsing, buying, returning, what they're interested in, when they're going to buy again, what channels they’re most likely to engage with, and how you should communicate with them.
When we built our CDP in 2013, no one knew what a CDP was. It wasn't a category, there was no one else doing it, and retailers were still trying to make the most of legacy ESPs. But we saw the AI era coming, and we knew that for marketers to be able to truly benefit from it, they would need more than just AI: They would need two very specific things.
Remember our engine example? The engine was the general purpose technology, but it wasn’t that alone that changed everything: It was the car. An engine running on its own achieves nothing—but put that engine in a car and fuel it up, it changes everything.
Now, cast your mind back to 2018, when OpenAI released GPT-1 and GPT-2. Chances are, you never used these early iterations. Flash forward to January 2023, and ChatGPT had already surpassed 100 million users within just a few months, making it the fastest-growing application in history. So, what made ChatGPT such a runaway success, while GPT-1 and GPT-2 never really took off?
The truth is, GPT-2 and earlier models were just an engine. They were powerful under the hood, but inaccessible to most people, requiring an API and technical skill to use. In other words, you had to build your own car just to take a drive; and who wants to do that? ChatGPT changed everything. It was the first fully packaged consumer product built on top of a language model. It was the car of the AI age, and anyone with an internet connection could just get in and drive.
The second missing thing that made ChatGPT so powerful? The data—or to continue our motoring metaphor, the fuel. GPT-1 and GPT-2 were trained on relatively limited datasets, but later models had the entirety of the data accessible on the internet and more to learn from, which meant the model gave answers which were transformationally better.
In summary, we knew that to leverage AI effectively, marketers would need all three of these elements. First: The engine, or the AI. Second, the fuel, or the data—and every last drop of it: Customer data, order data, behavioural data, channel interaction data, and everything in between. AI that is trained on or that operates with only a subset of that data, for example in an ESP or other point solution, would be just as ineffective as GPT-2 was.
Finally, they’d need the marketing equivalent of the car. The purpose-built product that gives them the ability to fully utilize AI. To use it to understand the customer, to launch campaigns, to optimize their experiences, to work out what will have the biggest impact on their strategy and then to implement it.
That’s why when we built Ometria, we built it to be AI-first from the start, and we built it as a unified end-to-end CDXP—a full Customer Data and Experience Platform. Not two systems duct-taped together or a platform where AI would be an add-on, but one tightly integrated product designed for a single purpose: To be the car that would transform the next generation of marketing.