How Scientific approach with AI/ML modeling improves Planogram Assortment in Retail and Consumer Products Industry

By Chida Sadayappan & Dinesh Kumar

A planogram is a model that specifies exactly how and how much quantity of products should be displayed on the shelves to maximize the sales and enhance customer experience. Traditionally Retail & Consumer Products industries relied heavily on their past statics for forecasting and heuristic methods / human judgement to perform Planogram product assortment resulting in not only lost sales due to out of stock but also higher wastage due to product stale and increased service levels in stocking up the merchandize. Retail and Consumer products industries have begun to realize that the traditional approach has its limitation and is not working well in a rapidly evolving and highly competitive market, Having said that AI / ML has started playing a key and big role in the planogram assortment by helping rank and recommend the products for maximizing sales. There are three key aspects mainly that help build a successful ML based Planogram assortment:

·      Do we have historical sales data available in an accessible state to adopt AI / ML?

·      Should we build or buy the AI/ML based Planogram?

·      How do we ensure successful adoption of the ML based Planogram?

Do we have historical sales data available in an accessible state to adopt AI / ML?

Before taking the journey towards implementing AI/ML on a larger scale for planogram assortment recommendation, Retail & Consumer Products industries need to consider two things: 

Deriving insights from data: Historical sales performance data is a key element in building an AI / ML based planogram recommender. The most important consideration therefore is to focus first on building a centralized data repository to make data available centrally in a public cloud platform along with other historical sales signal data such as promotions, discounts, demographics etc. Historical sales data combined with external sources such as weather, holiday etc. will help predict & recommend the ideal planogram assortment by uncovering the past sales pattern

Revamping legacy systems / human judgement with data-focused strategies: Consumer Products & Retail industries have been depending too long on statistics based forecasting techniques to predict demand and relied on heuristic methods / human judgement to perform planogram assortment leading to lost sales/revenue and coupled with that the legacy systems are tightly coupled making it difficult to adopt AI/ML without rethinking and strategizing the Tech spend. This has resulted in most of the initial excitement of AI/ML experiments and POC’s going into the shelf/backburner. It's unrealistic to replace these legacy systems in one swoop; instead, the industry should implement a phased approach, starting with better data management and governance; rationalizing IT architecture; adopting cloud for gaining better computing power to enable AI/ML; and adopting MLOPs / DevSecOps to automate development process for continuous integration, continuous delivery, innovation, and monitoring.

Should we build or buy?

After overcoming the initial hurdles and obstacles and deciding to move ahead with the AI/ML based Planogram solution, it's time to answer the million-dollar question: build or buy?

Build: Although there are commercial off the shelf AI/ML based Planogram solutions available in the market, but it is important to evaluate the solution and make right assessment before deciding. Key drivers to help decide may include:

·      There could be very unique factors specific to the Retail / Consumer Products industry such as the mix of existing & new products for assortment change very frequently, new/innovation products are introduced in the stores almost every month, and lastly integrating the AI/ML product with the custom built visual assortment mobile application / other prescriber applications will become complicated and expensive.

·      Gaining full ownership of the code and model is of utmost importance to avoid compromising the secrets and retaining competitive edge.

·      Industry is going for a long-term commitment and perceives deployment of AI/ML technology as a differentiation, and benefits of custom solution far outweigh in terms of buying a solution and customizing it to meet the needs vs building and operationalizing the solution using MLOps considering the frequency of retraining.

·      Who is looking for the ML based planogram assortment solution? If it is the Consumer Products company or the Manufacturer, then custom-built solutions may work better because of the constraints / limitations they will have to deal with at physical store locations

Buy: In case of small and mid-sized retailers, it makes a very good case to buy a commercial Planogram solution instead of building as these applications will deliver quick value and as well they can take advantage of the pre-trained models built using real-world data by the product vendor resulting in higher accuracy, than the time invested in hiring and training the right talent and also the investment in supporting infrastructure would be huge compared with buying an off-the-shelf solution. Even if small and mid-sized retailers choose to build their own AI with open-source tools, it can cost millions of dollars and can take months to train ML algorithms to do what most vendors have already achieved. 

Commercial off-the-shelf AI Planogram solutions provide huge advantages by helping understand localized customer behavior patterns and operational process thus enabling data driven decision making process. It also helps in cutting time spent in cleaning data, smoothing production issues, and avoiding reinventing the wheel when deploying models daily or building in the documentation and best practices for enabling reproducibility.

How do we ensure successful adoption of the ML based Planogram?

All the excitement of implementing an AI/ML based Planogram will be short lived if not for establishing a proper evaluation criteria, signing up business & SME’s for their expertise, training the field staff to adopt and use the ML recommendations and most importantly integrating the Planogram solution output with the consumption system / application. We can ensure the readiness for adoption of AI/ML based Planogram with: 

Defining appropriate metrics for evaluation: AI/ML Planogram stakeholders play a very important role in ensuring the success of AI/ML based Planogram. Throughout the journey from initial conceptualization to POC building and to rollout of the solution into production, their expertise is needed to help define the objectives, key performance indicators and in evaluation of the AI/ML based recommendations against the defined objectives and KPI’s. To measure the success of the AI/ML based recommendations, it is essential to measure the lift in sales growth in the stores using AI/ML recommendations v/s stores not using, comparison of sales lift from the prior week and same week last year as well as measuring reduction in logistics related expenditures is a key indicator.

Test and Learn: It is imperative for the ML Planogram initiative stakeholders to communicate and explain how the ML based recommendation is key for growth of business and is a more scientific way of executing Planogram assortment to maximize sales, optimize logistics but also how it aligns with overall organizational strategy. Test and Learn activity can be done in bite size steps to measure and evaluate against the KPI's defined by picking a handful of test stores. ML product recommendations for the selected stores should be made available to the integrated mobile delivery / schematic app for consumption by the field employees in the store for executing Planogram assortment. The Test and Learn activity can be carried out over a period of few weeks to a few months depending on the defined objectives and goals and evaluating how they are being met. After completing the successful Test & Learn activity milestone, decision should be made on rolling out the solution across all the stores.

Consider leveraging ethics frameworks like Deloitte’s Trustworthy AI ? which consists of six dimensions for organizations to consider when designing, developing, deploying, and operating AI systems. The framework helps manage common risks and challenges related to AI ethics and governance,

To summarize, AI/ML has come a long way in providing data driven insights for Retail and Consumer Product industries and helping them improve sales, stay ahead of competitors and more importantly helping them understand customer behavior better leading to better customer satisfaction.