Matching product exposure to demand
Ensuring that products are not over or under exposed compared with demand is a key success factor in online selling. It’s tricky to achieve the ideal balance manually.
Take the example in the chart below. Products are often under exposed at the start of their life and over exposed towards their decline in popularity.
The X axis represents days. The purple columns stand for sales volumes; the orange represent the volume of product exposure done manually.
Reactions to changes in product demand
In a physical store, an experienced merchandiser with a nose for products in demand can adjust product exposure on the fly.
In the online world, the amount of product data processed go beyond human capacity to analyse. Furthermore, online retail is continually swept by micro trends, where products can go viral fast. In addition, sales are permanently affected by unpredictable phenomena, such as a summer music-hit, an abrupt change in weather, a sudden political event, etc.
If you want to capitalise on those trends, you have to respond immediately. Just a couple of hours later can be too late.
Balance long and short-term best sellers
An intelligent automated solution is needed, and this is where AI and machine learning comes in. The AI can process huge amounts of data in real-time and do it in milliseconds. It can distinguish relevant data and deliver results at the right time, in context.
By listening to data coming in real-time, the AI learns, react and trigger a response that is relevant to the moment, and to the individual, by using the visitor’s history and behaviour on the site as well as crowd wisdom - aggregated data from all visitors.
Finding the right balance between long- and short-term best sellers requires a recommender system that can balance the trade-offs between consistent good sellers over time, and best sellers over the short time span. That means exploiting smaller trends without hurting or sacrificing confirmed long-term sellers.