Don’t Run AI in Silos: Democratized Data and AI Can Connect a Retailer End to End
by Vijay Raghavendra
For all the potential that AI and machine learning can offer a retailer, its value is extremely curtailed if it operates in a silo. It’s an interesting dilemma that hearkens back to the age-old problem of retailers not working across the organization.
Aligning teams — with or without AI — is always going to be a challenge, but one interesting element of AI is the democratization of the technology and the data it consumes. In short, virtually anybody in a retailer organization can use AI to improve performance in their roles — that is, if the data streams and machine learning algorithms are fully connected throughout the organization.
Democratizing Data and AI
The democratization of data can be powerful when it is transformed into actionable recommendations by role-focused AI solutions. Across the organization, AI has the power to transform category managers, store operations teams, or logistics planners into an army of functional data scientists, enabling any employee to use connected data to their benefit.
A merchandising team can use machine learning to automate pricing and promotions strategies, and if that data is connected, the team can see real-time adjustments occurring from the supply chain side. Similarly, store associates can use their handhelds equipped with computer vision algorithms while walking the store aisles and leverage streamlined data for up-to-the-second accuracy on stockouts, pricing, or planogram deviations so they can take immediate corrective action. A category manager reviewing shopper behaviors through orders and sales performance can leverage data from a fully connected AI program to better forecast demand for products at a regional or store-by-store level and adjust assortments to meet the needs of each store’s shoppers.
In most retailers, these data sets work in silos, limiting their users’ visibility and providing equally limited value. By fully connecting data and AI, category managers, demand planners and more can use the tool as effective copilots.
A PwC report on AI market size found that in the next seven years, nearly 85% of the retail marketplace will be using AI and it could contribute up to $15.7 trillion to the global economy by 2030. AI is a fast-moving train, but if retailers can take a pause and make sure they deploy it correctly, they will see greater business results and more efficiency.
Foundational strategies to consider when getting teams aligned around AI include:
Start by Connecting Disparate Data Sources. Throughout the organization, retailers are operating data in silos. For example, one department has been using AI to monitor inventory and predict routes from the warehouse to the store. Pleased with the results, the retailer inputs AI in a separate unit, completely disconnected. Retailers can start by identifying the silos and get that data connected.
Focus on Explainability and Human Feedback to Build Trust. Don’t develop AI/ML algorithms as a black box. Prioritize explainability of the predictions and recommendations from the algorithms and then surface this context and data to the business users. Build in mechanisms from the get-go to incorporate feedback from the business users to train the algorithms on an ongoing basis to remain relevant and accurate. Transparency in AI’s recommendations and what sources drove its conclusions accelerate trust and adoption.
Embrace a Role-Based Point of View. Before rolling out AI, know how each persona or user will leverage the recommendations. Also, develop clear use cases on how each persona can use the data and collaborate more effectively. Upfront planning will make the AI smarter and more productive.
Connecting a Retail Organization
AI can serve as a powerful asset for all in retail. Where teams previously spent hours or days analyzing silos of data, with a fully connected operation, they can get holistic results in minutes, act on pinpointed recommendations targeted to their specific roles, and deliver measurable business impact.
There’s tremendous power in leveraging AI throughout the organization to build a more data-driven, customer-focused culture with a sustainable competitive advantage.
Vijay Raghavendra is the chief technology officer at SymphonyAI.