Retail Doesn't Need More AI Tools. It Needs the Right AI Roles.

The brief usually starts with "we need more AI and data capability". That's a fair instinct. It's also too vague to hire against.

Retail is one of the best places to use AI well. You have huge volumes of customer, product and stock data, tight margins, and decisions made thousands of times a day.

But a tool doesn't make a decision. A person does.

The ten roles I'd put at the top of the list

These are the AI roles I see making the biggest difference in retail right now. Not every retailer needs all ten. Most need more than one.

Head of Data & AI. Owns the AI and data roadmap across ecommerce, marketing, supply chain and stores, and ties it back to commercial outcomes. Without this person, everything else becomes a collection of pilots.

Retail or ecommerce data scientist. Works on customer analytics, experimentation and personalisation across web and app, with conversion and basket size as the scoreboard.

Recommendation or personalisation engineer. Builds and tunes the systems that decide which products and content each customer sees online.

CRM and loyalty data scientist. Uses AI for segmentation, churn prediction, offer targeting and loyalty program performance.

Pricing and promotion data scientist. Optimises prices, promotions and markdowns with predictive models and testing.

Demand forecasting data scientist. Forecasts demand at SKU, store and channel level, so replenishment and buying run on evidence rather than gut feel.

Supply chain optimisation or inventory analytics lead. Focuses on allocation, replenishment, network design and stock efficiency.

Retail marketing AI specialist. Connects AI tools with campaigns, channels and the martech stack, so marketing gets more targeted and less manual.

Operations and workforce analytics lead. Uses data to improve store labour planning, productivity and in-store execution. As a former CPO, I'd ask People and Culture to watch this one closely.

AI adoption or enablement lead. Drives the training and change that gets marketers, merchandisers, planners and store teams actually using AI day to day. I've unpacked what an AI enablement manager actually does separately.

The title is not the job

Look at how many of those roles say "data scientist". On paper they look interchangeable. In practice, a pricing specialist and a loyalty specialist solve very different problems, and hiring one when you needed the other quietly costs you six to twelve months.

That's not a talent shortage. It's a scoping problem.

My guide to every role in an AI team shows how they fit together.

If you're hiring into this

Start with the decision, not the title. Pick the commercial decision you most want to improve, whether that's markdowns, stock or loyalty, and hire the role that owns it.

Don't skip the lead or the adoption role. Specialists without a leader drift. Tools without adoption sit on the shelf.

Be deliberate about how you fill each role. Some of these are long-term capability. Others are a defined project. The Bot, Build, Borrow, Buy framework I use with clients helps you decide what to hire, what to grow from within and what to borrow.

Where I'd start

If I were a retailer starting from scratch, I'd pick the one commercial decision costing you the most, whether that's markdowns, stock or loyalty, and hire the person who owns it. Get that first hire right and the next ones are much easier to scope.

If you're not sure which role that is for your business, my AI Skills & Roles Map can help. It looks at the skills you already have across ecommerce, marketing, data and supply chain and turns the gaps into a practical hiring and upskilling plan. Get in touch if you'd like to talk it through.