Introduction: AI in your morning message
Do you ever wake up to the text, ‘Good morning, should I order a matcha tea latte at Starbucks near you?’ Most of us won’t be surprised by what we see on the screen, since we live in the AI era. But one thing that would surprise the reader is the drink, especially if they are not a matcha drinker. This small moment shows how AI decision-making is already part of daily life.
Something similar happened to one of our regulars who lives nearby and doesn’t use mobile order. He enjoys coming in early in the morning, grabbing his cup of coffee, and engaging with the team. One day, he woke up to the message. He was surprised—he’d never ordered matcha. He chatted about it with our team. He didn’t like the idea of AI deciding things for him.
For our Starbucks team, it’s not surprising, as we are aware that the company has been collecting data through its ‘e-CRM.’ And the drink recommendation was probably based on his previous order, maybe not for him or for someone else, or a mere recommendation, depending upon the popularity and demand of the drink. But the real question is, ‘Does everyone like such recommendations? Is it really the first thing they want to see from Starbucks popping onto their screen? Are people really enjoying AI making decisions on their behalf?’
AI: convenience or concern?
Maybe it’s a yes for some and a no for others. On the one hand, people loved integrating AI into daily operations, but now most find it daunting to share too much information. On the other hand, big corporations are getting used to it. AI helps them with data-driven decision-making, optimizing operations, predicting future growth, and tracking progress using KPIs and baseline performance.
This kind of data-driven decision-making now shapes menu changes, store closures, and staffing plans. These companies can access it only if people allow them to use it. This shift raises a key question about AI decision-making: who is really in control?
Most of us are in a quid pro quo relationship with the companies we associate with. We share our information in exchange for the service. Earlier, the retail industry focused on keeping consumers in the store as long as possible. It used tactics such as conspicuous endcaps and in-store promotions.
Now that everything is digital, there is no need for such a thing. If one can’t decide what they want, they can ask ChatGPT, which shares information. Most of us have developed a pattern of not dwelling on a single topic for too long, as it wastes time. But one must ask oneself, ‘Am I even allowing myself to think about myself?’
How Starbucks uses data and AI
Of the seven data collection methods, Starbucks uses first-party data from surveys (partners and consumers), transactional tracking, online tracking, and social media monitoring. The company relies on this first-party data to guide product and store decisions. In 2025, Starbucks CEO Brian Niccol announced plans to streamline the company’s menu. The plan cuts 30% of food and beverage offerings by the end of the year. Additionally, the company closed a few underperforming stores. These decisions were based on the data collected.
The company has been utilizing ‘Deep Brew,’ its AI machine learning platform. The company introduced it in 2019 as a digital backbone. It personalizes customer experience and optimizes behind-the-scenes store operations. Some of the key functions of the same are hyper-personalization, which powers the recommendation engine in the Starbucks app and suggests drinks and food items, as explained in the message our customers received; predictive inventory, which we will further discuss; labour and store efficiency; and location intelligence. Hyper-personalization drives the app’s drink suggestions and tailored offers. Predictive inventory helps stores stock the right items at the right time. Together, these tools shape everyday AI decision-making for both customers and store teams.
Conclusion: Who is really deciding?
To sum up, data helps a company gain insights into its future, and it would be injudicious for a company not to leverage digital trends. Ignoring these digital trends would put any brand at a disadvantage. But rather than blaming AI for making a decision, one must ask themselves if this tool is doing its job as it’s supposed to, then why am I blaming it? As most of us should know, AI was born out of curiosity, advancement in computing and math, and the desire to build a system that can learn and solve harder problems faster. Developers created AI to handle complex tasks more quickly than humans alone. So, deciding what one must have for breakfast should not be hard, unless there is a food crisis in today’s world. Yet we give AI tools the authority to make small decisions for us. Is it necessary or not? You decide. In the end, we must decide how much authority we give to AI decision-making in our own lives.
