AI Brews Green: Pioneering Sustainable Futures at Starbucks
By: Simar Badwal, Elizabeth Baker, Christina Fleischer, Samantha Wentworth
Introduction
There are many different definitions of Artificial Intelligence (AI) and types of AI. As far as the ways we explored how AI could impact and transform Starbucks, we define it as: a technology which uses machine learning and data analytics to provide recommendations (and/or takes actions) to improve and optimize processes, deliver personalized experiences, and find opportunities for automation.
Emerging Applications of AI for Improving Starbucks Sustainability
We believe that AI can improve Starbuck’s overall sustainability in four keyways: in-store management of energy, supply chain optimization, predictive analytics for ordering, and personalized recommendations and recommending healthier ordering habits.
First, let’s discuss how Starbucks can utilize AI to help with in-store energy management, which relates directly to SDG 7, affordable and clean energy. In the short-term, AI can utilize data such as foot traffic, weather, and average order numbers to adjust heating and lighting systems for maximum efficiency, this could lead to immediate reductions in energy waste and costs. The company could also invest in AI powered systems to predict equipment failures and preventative maintenance needs, reducing energy consumption associated with inefficient operations. AI could also be leveraged to work together with employees to suggest ways to reduce energy waste, like switching off one machine, or only using certain blenders, fostering a culture of energy conservation among baristas. In the long-term AI could be used to address in-store energy conservation by facilitating an integration of renewable energy. AI could be used to analyze energy production and consumption and maximize the use of renewable energy accordingly. Also, as stores continue to rely on AI for shorter-term energy impacts, they can build habits of dimming lights, turning off machines, and minimizing temperature adjustments that would greatly reduce energy costs and waste.
Next, AI could greatly impact Starbuck’s supply chain optimization, reducing costs and over consumption, in line with Sustainable Development Goal 12, responsible consumption and production. In the short-term, AI can help stores with demand forecasting, route optimization, and supplier collaboration. AI can analyze historical sales data, weather patterns, and local demographics to predict demand for different products at each individual franchise location. This ensures more accurate inventory management, minimizing the probability of overstocking or understocking any products. AI could also optimize delivery routes for Starbucks suppliers. This in turn could reduce fuel consumption and carbon emissions. AI can identify the most efficient route using road conditions, traffic patterns, and weather leading to immediate environmental benefits. Also, with supplier collaboration, AI platforms can facilitate communication and ordering with suppliers, helping with real-time tracking of inventory, schedules, and delivery status. This transparency between stores and suppliers allows for better coordination and responsiveness and would help minimize waste. In the long-term utilizing AI for supply chain initiatives could help with sustainable sourcing and circular economy initiatives. AI can help Starbucks identify sustainable options for suppliers, by utilizing data from production methods, labor practices, and environmental certifications. For circular economy initiatives, AI analytics can support Starbucks’ effort to implement more recycling programs and reusable container options. This has been a huge priority for Starbucks recently, and by tackling materials through supply chain with AI, they could be one step closer to achieving this.
Additionally, we believe there is an opportunity for Starbucks to leverage AI by using predictive analytics for the ingredient supply forecasts, orders, and related processes. Ultimately, by reducing the amount of unused ingredients in individual stores, this would reduce waste and lead to reduced costs for the business. This application of AI would address Sustainable Development Goal 12, Responsible Consumption and Production, both in the short-term, but more notably in the long-term. By leveraging AI in the ingredient ordering and forecasting processes, the orders can be better optimized (in the short-term) for the location demographics, seasonality, trends, and historical data. For example, an influencer known for their Starbucks orders starts trending – they just started promoting an oat milk and caramel syrup beverage. With AI, Starbucks’ models can track this trend and automatically place orders for more oat milk and caramel syrup in areas popular with Gen Z for two weeks, while reducing their orders for other milks and syrups. On the other hand, if Starbucks’ AI model analyzes data (historical or predictive) to understand that there is a dip in coconut milk orders during a 2-week period for a certain location, they can reduce the orders for coconut milk, and reduce waste and costs in the process. This process can be much more accurate, more precise, and much faster by using AI. Those are both examples of short-term environmental performance in the value chain. As far as long-term, Starbucks can more closely analyze and predict consumer trends and seasonality over time – such as when deciding to release a new drink or pastry, particularly with their seasonal menus. By using AI, Starbucks can predict what types of offerings will be popular or less popular within a few years, particularly by analyzing data trends and looking externally and in the marketplace for information. By using AI, Starbucks can then revamp their ingredient list to understand what is wasted in certain locations, during certain seasons or time periods, or universally across locations. Over time, these more accurate forecasts and precise order scheduling will lead to less waste, lesser environmental impact, and greater sustainability.
Lastly, we considered how Starbucks could leverage AI to provide customers with personalized recommendations and even promote healthier alternatives – enhancing their social impact. The first aspect of this also uses AI for predictive analytics. As Starbucks captures more data and orders from individuals, they can better optimize recommendations of beverages – particularly in their mobile app. Depending on the time of day, month, and recent trends (both for the customer and in the marketplace), Starbucks can recommend the exact right personalized drink for each individual. This performance improvement can start in the short-term but would become even more optimized and precise in the long-term as more data is captured and the AI model continues to learn. As a result, this AI optimization will make it easier/quicker for the customer to place an order, they’ll have a positive service operations experience, and they’ll be more likely to remain loyal to Starbucks in the long-term. In fact, for business and revenue growth, Starbucks could leverage AI to automatically place orders for customers based on their recent history, location, and AI insights, with the ability for customers to opt-in to this feature. Similar to the example above, this aspect could address Sustainable Development Goal 12, Responsible Consumption and Production, as Starbucks gains more insights into (and even start guiding) which ingredients should be stocked and offered to customers.
Furthermore, by leveraging this data to use AI to make personalized recommendations, Starbucks can also address Sustainable Development Goal 3, Good Health and Well-Being. For example, instead of strictly keeping drink recommendations to the customer’s taste profile, Starbucks can start recommending healthier options – such as less syrup pumps, sugar-free syrup, nonfat instead of whole milk, non-dairy milk, etc. By improving the AI recommendations for healthier options in the short-term, the health and well-being of society could improve in the long-term. Particularly with the sugary syrups and high calorie pastries, Starbucks could start phasing these options out, promote heathier orders, and even start addressing obesity and diabetes rates for the public – improving social impact performance. In addition, AI can be leveraged to flag unhealthy options on the menu so that Starbucks can address them, accurately determine and predict what menu items or ingredients could be removed without public outcry, and work toward developing a healthier menu for its customers in the long-term. Over time, the impact of AI can greatly improve personalized recommendations, customer satisfaction, and in the long-term, be leveraged to improve the health of its customers.
Effects on Stakeholders in the Value Chain
As mentioned above, AI could have effects on various members of the value chain including employees, suppliers, customers, and investors.
For Employees and Suppliers, there is the benefit of optimizing processes and opportunities for automation via in-store management of energy and supply chain optimization. Starting with employees, the process optimization guided by AI driven tools will help streamline in-store operations, making it a less stressful work environment as routine tasks get eliminated from their responsibilities. This implementation of AI requires upskilling the employees which helps them gain new skills in AI oversight, and technology management. And lastly the energy conservation efforts can help foster a culture of sustainability which employees can apply to their personal lives. This will serve as an extremely beneficial tool to suppliers through supply chain optimization, as the more forecasted the orders are, the better a company can plan their production activities. This will also lead to better communication and collaboration between Starbucks and their suppliers as information becomes more transparent. Lastly, it encourages suppliers to take their own initiatives when it comes to sustainability goals, aligning with the circular economy efforts.
For Customers and Investors, there is the benefit of delivering personalized experiences with predicative analytics and ordering while promoting healthier ordering habits. Improving efficiency and ordering will make customers feel more aligned with Starbucks values, leading to increased sales due to a more satisfied customer experience which creates value for investors. Predictive analytics for ordering via AI would mean optimization of trends and historical data to predict new trends offering moments of surprise and delight to customers with AI predicted future limited time offers and new beverages in the long-term, anticipating customer’s needs. Increased satisfaction with AI predicted new beverages and pastries on menus drives customer traffic which increases sales which satisfies investors. Predictive AI could also leverage customer data to enhance the ability to recommend the perfect drink for a customer based on taste trends or desired health and wellness goals. Brand loyalty would increase from AI induced support of health and wellness goals for customers via predictive analysis which drives ROI for investors.
Conclusion
The integration of AI within Starbucks’ value chain would represent a pioneering step towards redefining the nexus between technology and sustainable business practices. It is important to note that a successful implementation of AI in Starbucks’ value chain relies on a balanced approach between technological innovation and ethical considerations, stakeholder engagement, and continuous learning. In addition, it demands a forward-thinking leadership that understands the ethical dimensions of AI, is committed to upskilling the workforce, and dedicated to transparently navigating the complexities of digital transformation in alignment with its core value of ‘delivering performance through the lens of humanity’.
