Overcoming Four Key Challenges of AI Integration in Retail

Overcoming Four Key Challenges of AI Integration in Retail

September 19, 2024

In recent years, the rapid development of Artificial Intelligence (AI) has sparked a wave of digital transformation in the retail industry. From intelligent recommendation systems to cashierless stores, AI is transforming every aspect of retail at an unprecedented pace.

However, this transformation comes with its share of challenges. As retailers adopt AI technologies, they face various obstacles, yet they are finding ways to overcome them and unlock new business models in the process.

Why Implement AI: Define Your Needs and Objectives

AI is merely a tool; its effectiveness depends on whether users clearly understand their goals and purposes for using it. If a business cannot answer questions like “Why do we need AI?” and “How will we use AI?” then the tool may end up being ineffective and even redundant.

This principle applies not just to AI but to any new technology, tool, or equipment. Before adopting or utilizing new innovations, it is crucial to identify the problems that need solving or the workflows that need improvement. Only by focusing on these needs can the tool truly deliver value by addressing the issues or streamlining workflow processes.

Data Management

The vast customer data held by retailers is the foundation for AI development and holds immense commercial value. However, how to securely manage and protect this data remains the primary challenge when adopting AI.

Moreover, ensuring the accuracy, completeness, and timeliness of data is critical, as it directly impacts the performance of AI models. For example, an e-commerce platform that incorrectly records a customer’s age or has incomplete purchase histories could lead to an AI-powered recommendation system suggesting unsuitable products to users.

Retailers can turn to third-party service providers for data management solutions. Many specialized providers not only offer data management services but also integrate AI capabilities, helping businesses manage their data more effectively while unlocking additional business value.

Privacy Issues

Privacy regulations, such as the EU’s GDPR and Taiwan’s Personal Data Protection Act, impose strict requirements on the collection, use, and protection of personal data, making privacy management a complex and challenging task.

Retailers can start by understanding the scope of services offered by third-party providers, reviewing contract terms, assessing how data access is controlled (such as using role-based access control, RBAC), and evaluating the methods for protecting sensitive data. Retailers should also verify if their internal data contains any sensitive information subject to regulations like GDPR or local data protection laws, and plan accordingly.

However, some AI applications do not raise any privacy concerns, as they handle only internal company data, such as product information, without involving any customer personal or sensitive data. These systems focus solely on operational data, ensuring there are no risks related to individual privacy or data protection regulations.

For example, Viscovery’s AI image recognition system, a powerful tool for speeding up checkout, only retains image data of the products. The key to managing this data is ensuring it is always up to date to maintain accuracy.

Integration Complexity and Technical Barriers

AI involves a wide array of complex tools and systems, from machine learning algorithms to automation systems. Integrating AI with stores’ existing systems often requires substantial resources and time. For small and medium-sized retailers, the high costs of technology investment pose a significant barrier, and the shortage of AI talent further limits AI’s widespread application in the industry.

Furthermore, retailers have diverse and complex technology needs, from customer behavior analysis and inventory management to personalized recommendations. Selecting the right AI solutions is no easy task. Therefore, close collaboration between technical and business teams is crucial to ensure AI meets business needs.

To overcome these challenges, retailers can:

Evaluate cloud-based AI services and partner with AI companies: Many AI providers offer pre-trained models that allow businesses to quickly integrate AI functions without building them from scratch, saving both time and money. For instance, Viscovery’s AI image recognition system, offered as a SaaS (Software as a Service) solution, can seamlessly integrate with retailers’ existing POS systems. This application can be implemented in just one to four weeks, breathing new life into the POS. It helps cashiers quickly identify items without barcodes, reducing checkout times by over 50%.
Leverage open-source tools: If in-house data scientists and engineers are available, businesses can use tools like TensorFlow and PyTorch to reduce development costs and create tailored AI solutions. There are abundant resources and communities online to help internal teams quickly learn and master these tools.

When selecting AI solutions, retailers should start small and gradually scale up. For example, begin with a chatbot for customer service, a product recommendation system, or an image recognition system for checkout, building experience along the way. It is also essential to strengthen communication between tech and business teams to define AI goals clearly and ensure solutions effectively address business challenges.

To address talent shortages, companies can train existing employees to acquire basic AI knowledge or collaborate with academic institutions to bring in AI talent. Participating in AI communities can also provide access to the latest technologies and experience sharing.

Adoption and Acceptance Among Employees and Customers

The adoption of AI in retail not only transforms business operations but also requires the acceptance of employees and customers.

Employees may resist AI due to fears of job displacement, while customers might feel unfamiliar or distrustful of automated services. These psychological barriers require time and resources for companies to address through education and training.

Retailers can take the following steps to improve acceptance:

Provide employee training: As Liao Ming-Chien, CEO of Taipei’s famous bakery chain I JY SHENG, shared, “The key to digital transformation is employees.” I JY SHENG has placed great emphasis on employee education and training to ensure they fully understand the purpose and benefits of AI. For example, the implementation of AI image recognition for checkout not only reduces customer wait times but also eases the workload on employees. In addition to technical training, it is important to emphasize the positive effect of AI on employees. ¹
Emphasize human-machine collaboration: Make it clear that AI is meant to assist, not replace employees, enabling them to focus on higher-value tasks.
Establish communication platforms: Hold regular employee meetings to provide transparent information and foster two-way communication.
Create a positive company culture: Encourage employees to participate in the AI integration process, making them feel part of the transformation.
Design user-friendly interfaces: Ensure that AI systems are simple and intuitive, so customers can easily adapt.
Maintain excellent customer service: While adopting AI, retain some level of human service to cater to specific customer needs.
Share success stories: Host success case-sharing sessions or write blog posts to showcase the benefits of AI to both employees and customers.

Navigating the Future of AI in Retail

Though the integration of AI in retail is full of challenges, with thoughtful and strategic planning, adaptable approaches, and ongoing learning, companies can overcome these obstacles and reap the benefits AI offers—improving efficiency, reducing operational costs, enhancing customer experiences, and even accelerating new product or service development.

In this AI-driven retail revolution, companies with innovative thinking and adaptability are more likely to stand out. Businesses should encourage employees to propose creative ideas, foster an open corporate culture, and collaborate with academia or AI companies to explore AI’s endless potential. At the same time, businesses must prioritize ethical AI development, ensuring its application aligns with societal values, paving the way for a smarter and more sustainable future.

[Reference]
¹ 陳君毅. “員工素質不高、創新能力不足都不是藉口！44歲的一之軒告訴你：數位轉型前應先了解問題！” foodNEXT, 27 June 2024, https://www.foodnext.net/news/industry/paper/51 11962911.