AI ecommerce projects may fail without clean, centralised data: Report

Ecommerce businesses seeking to leverage artificial intelligence (AI) must prioritize the establishment of clean, structured, and centralized data systems, according to a recent report by global tech firm Nisum. The report emphasizes that the quality of data is a crucial factor in the successful implementation of AI projects, as AI systems heavily rely on the information they receive.

Data preparation is often underestimated, and businesses frequently encounter unforeseen issues with their data even after initiating an AI project.

Data spread across various systems, such as product information, inventory records, and customer data, hinders the effectiveness of AI tools in gaining a comprehensive understanding of a business. Moreover, poor data can exacerbate problems, as AI systems may reproduce and magnify inaccuracies, like inaccurate inventory counts or duplicated customer records, at a larger scale.

The most common failure point identified in the report is inadequate data infrastructure. For e-commerce companies, this issue becomes particularly significant when implementing AI for personalization, pricing, forecasting, or inventory management. A marketing system with advanced AI capabilities, for example, could promote unavailable products if it is not connected to the inventory system.

To overcome these challenges, businesses should adopt integrated architectures that enable generative AI, predictive analytics, and automation systems to function collaboratively. Centralizing data from different business areas is essential, as fragmented information across stores, e-commerce platforms, and apps can hinder the transition from broad customer segmentation to personalized experiences.

Nisum's report underscores the need for businesses to treat data preparation as an ongoing process, ensuring the provision of accurate, up-to-date, and consistent information to AI models. As AI adoption continues to grow, businesses must address data readiness to transform AI pilots into production-ready initiatives. Companies that focus on data infrastructure and quality are more likely to achieve financial returns from their AI investments, according to Anurag Chauhan, the newly appointed CEO of Nisum.

Despite the immense potential of AI in commerce, the gap between average results and top-performing outcomes remains substantial and continues to widen.