AI and ML models are only as “intelligent” as the data that’s fed into them. Just over half (53%) of AI/ML models go from prototype to production, according to Gartner Inc. There are numerous factors that can make or break the success of your model, of course. For one thing, when you rely on models to grow your business, they need to be malleable to the myriad external factors that will affect your desired outcome. In other words, not all models will work in all situations, and it’s important to understand this.
Rather than using a blanket approach, more and more companies are starting to experiment with the concept of localized models.
Evolving AI/ML Models
Typically, when you’re using AI/ML models to drive your business, you’ll see a lot of value quickly with your first few versions of the model. If we’re looking at the journey of success with AI as a “zero to 100” scale, you can go from zero to 60 rather quickly by just making a few tweaks to your algorithms or models. But when you’re trying to make it the rest of the way to 100 — trying to realize even more value — that’s often the most difficult part of the journey.
Let’s say, for example, that you run a chain of stores and you have an AI model that’s used to predict how many staff members you need for a retail store to operate. In most situations, you’ll start with a base model (also known as a foundation model). And you’ll see some out-of-the-gate successes with that model right away. It can quickly take you to a certain level in your AI journey. But it grows exponentially harder to realize value and success from that point. It requires out-of-the-box thinking and a new approach to fully realize the model’s value. This is where the concept of localization can fit in.
Understanding Localized AI/ML Models
AI/ML models are trained with one set of data, but that dataset isn’t always applicable to a different location. For one thing, many AI/ML models are often trained with U.S.-based data. Artificial intelligence localization is an effort to create datasets to train models for many other markets in the world. For example, a U.S.-based company’s AI models might work fine for operations in the U.S., but they may fall short for markets abroad.
However, it isn’t just for global uses. Localization can also be used on a more micro level — for one side of the country versus another, for instance. A model might work fine for a restaurant chain’s stores on the East Coast, for instance, but not for the West Coast. Perhaps Californians are more likely to come to a pizzeria on weekends, whereas residents of Texas are more likely to get pizza on weekdays.
There are subtle differences between these locations that can mean the AI model you use for one isn’t going to work for another. Perhaps you’re using a model to determine staffing needs at each store — but that’s something that can change based on geographic location, and that needs to be factored in. Otherwise, your models won’t be useful. You can’t address the differences in behavior or traffic or other factors unless you have separate models for each location.
Localization can be done on an even more granular level, too. In a situation like the one mentioned above, you might find that rather than using the same AI model for all your U.S. stores, you have a model for each state or each city or even a model per brick-and-mortar location.
Getting Started With Localized Models
There are a lot of opportunities to experiment with localized models to help businesses gain a clearer understanding of their demographics and the unique needs and desires of different locations. It’s all too common for a company that’s getting started with AI models to get into this line of thinking that a model is “one and done.” That’s an incorrect notion; in fact, foundational to succeeding with AI is the recognition that it requires continuous iteration — and then running an iteration continuously until you find the optimal solution.
This can represent a lot of work from a technology standpoint, and that can prevent organizations from even considering the idea of localized models on top of what they’re already trying to tackle. But if AI is truly seen as a tool, a method for moving the needle on your business, then these are challenges you must tackle. Otherwise, you’ll never really get ahead. Your models won’t be successful in the way that you need them to be.
That said, keeping track of all these separate models can also be a major challenge. It requires a lot of experimentation. You need to be able to try new things regularly and continue to make tweaks, trying out different approaches for weekdays versus weekends, for instance. But this challenge isn’t insurmountable; there are tools available to help you with automating the management of all these different models.
What we’ve typically seen is that building models isn’t the problem; it’s organizing them and managing multiple models at scale that is. But you don’t have to go it alone, and this shouldn’t prevent you from experimenting with localized models. When it comes to the management aspect of your models, there are solutions that can assist with this, so don’t let that be a sticking point.
Room For Improvement
AI and ML models cannot fully succeed without the proper data. What organizations need to remember is that data isn’t homogeneous; because it can vary significantly based on location, localized AI/ML models provide a more accurate picture for companies to go by. The market offers tools to assist with localization and with management, so it’s a good time to create or improve on a model that will help you achieve your business goals.