eCommerce Virtual Assistant Research

We are interested in collaborating with eCommerce retailers who are exploring how virtual assistant technology could support day-to-day commercial decision-making within their online operations.

Running an eCommerce channel often means operating a “mini business” inside a larger organisation. eCommerce managers are required to make frequent decisions across performance analysis, prioritisation, planning, merchandising, marketing activity, and operational follow-up — often without dedicated analytical or operational support. This can lead to high cognitive load, reactive workflows, and missed opportunities.

Our research focuses on understanding whether an AI-based virtual assistant can help reduce this burden by supporting managers in their everyday work. Rather than replacing human judgement, the aim is to explore how an assistant can help with:

• highlighting what matters most on a given day

• providing context around performance and trends

• supporting prioritisation of actions

• explaining not just what is happening, but why it matters and when action may be required

Participating retailers would contribute real-world operational context and feedback, helping to assess how such technology fits into actual eCommerce workflows and decision patterns.

Initial research is grounded in eCommerce because online platforms provide rich, observable analytics and decision signals, allowing meaningful evaluation in live environments. However, insights from this work are expected to inform broader applications across other forms of commerce over time.

We are particularly interested in working with retailers who are open to:

• experimenting with new ways of interacting with analytics and operational information

• reflecting on how decisions are currently made day-to-day

• providing qualitative feedback on whether AI-assisted guidance could reduce workload or improve clarity

All collaboration is conducted at a high level, with detailed technical approaches shared selectively and under appropriate confidentiality arrangements.