Driving sales with intelligent customer recommendations

A successful packaged goods company was interested to see how data and machine learning could drive bottom-line growth for its business.

Servian recognised that there was an opportunity to help the company’s sales team to have more targeted conversations with its retail channels to ensure that the right products were promoted, at the right time and in the right location. The intent was to ensure that the retail channel and its retail partners were maximising revenue.

the approach

Servian was engaged and proposed a recommendation model that could suggest which products each store should stock — taking into account season, location and product-specific inputs. The recommendation model would have an interface for salespeople to use in conversation with retail stores, and it could suggest what products would increase revenue for that store location and why.

the framework

The importance of being in control of its own intellectual property was a neat fit for Servian’s six-step ‘ProductionML’ process, which helps to uplift the internal capacity for machine learning capability.

ProductionML is a six-step process which seeks to work collaboratively with customers to build internal machine learning capability while using a first use case as a demonstration of the technical possibilities of both machine learning and cloud technologies.

The six steps are:

  • Step 1 – Business Impact Workshop
  • Step 2 – Model Development
  • Step 3 – Solution Architecture
  • Step 4 – Model: Evolving Dev to Prod
  • Step 5 – Governance
  • Step 6 – Model: Operate and Innovate

the process

Utilising Google Cloud Compute, Storage and Data Studio, Servian was able to take the data sets provided by the packaged goods company and construct a recommender model using machine learning. The interface of the recommender is a dashboard that customises for each retail store. Recommendation models were explored with the company, and the team iterated quickly through an agile process. The team arrived at a minimum viable product that could be taken on the road by their salespeople.

To ensure the application drove real business value into the future, Servian advised the packaged goods company on how to release it as a controlled experiment within the company’s sales team. This allowed for the impact to be measured and for feedback to be incorporated to tune the recommender model’s decisions.

the architecture

platform used

To achieve the outcomes required, the following GCP products were used:

Google Cloud provides a set of well-integrated tools which aid in the development of machine learning insights and the ability to deploy them into a production setting. A further advantage of using Google Cloud on top of the short development time is the clear path to a production grade machine learning system that is resilient, transparent and adaptable.

the results

The recommendation model that was developed provides the frontline sales team with data-driven insight into customer behaviour and store sales. This has enabled the company to improve its allocation of stock and sharpen its buying decisions, resulting in an increase in sales revenue.

By harnessing the power of machine learning, the model continues to develop and improve its accuracy, supporting the company’s long-term projection of growth.