Skip to content
Microsoft Industry Blogs - APAC

Have you ever imagined how different your day would be if you had a dedicated personal assistant? Someone who could arrive early and stay late, doing the administrative legwork so that you could make the most of your time. Look no further than Machine Learning – an emerging technology that enables computers to learn from and make predictions about data without explicit instructions. For Consumer Goods Companies (CPG), Machine Learning can, for example, ensure that field reps are routed to shop visits more efficiently and empowered to manage daily tasks quickly. Imagine, for example, if you could deploy a robot to check the aisles for merchandising compliance, freeing up your human capital to focus on more value-added tasks. Saving time in each shop gives reps the opportunity to complete more visits, resulting in potentially significant cost savings. But Machine Learning does more than streamline shop visits – it is an effective way of applying historical data to a problem by creating a model and using it to predict future behaviour. Over time, Machine Learning identifies patterns and trends – like when a promotion works and with what parameters – that stakeholders can leverage to improve company strategy. Read on to learn more about how Machine Learning could help CPG companies plan where to go and what to do, get more done with less data entry and deepen data analysis.

Automate and Optimise Route Planning

A lot of work goes into identifying which points of sale must be visited, with what frequency and what has to be done there. Imagine being empowered by software that considers multiple factors beyond geography to automatically predict the best time to visit a particular shop and improve overall routing efficiency. If a shop usually has a slump every July, recently hired a new manager or has a new promotion coming up the software will adjust routing accordingly to optimise employee time. Machine Learning technology also generates tailored to-do lists for shop visits that are based on what a particular shop needs. So, before the account rep gets to the shop they will be made aware of issues like faulty equipment and they won’t have to spend time determining which tasks are required. Once reps arrive on site, the software can also help them to streamline their audit activities.

Enable the ‘Perfect’ Shop Visit

When the account reps are conducting a shop audit, imagine having technology that helps them to be more efficient with their time. Not only would they be able to avoid pen and paper – which 64% of retail execution professionals still use[1] – but they could skip digital data entry by using digital image recognition and speech-to-text functionality. Digital image recognition allows reps to take pictures of product displays in the shop instead of recording inspection results manually. From an image, a model can evaluate out-of-stocks, facings, prices, share of shelf and planogram compliance. Where a human operator would have to visually assess each detail to find errant product placement, the software finds errors and inconsistencies in seconds. Machine Learning also enables reps to verbally dictate notes, commands and order placements to a wearable device such as a smart watch or headset. The system isolates key words from the dictation, which will trigger actions in the Retail Execution software. Digitally-captured data saves retail execution professionals time and avoids the mistakes inherent to manual data collection. Data from the visits is disseminated in real time, so that managers receive audit results immediately instead of months after completion.

Deepen data analysis

Once the data has been collected, the final benefit of applying Machine Learning in retail execution is to find patterns in data that can help predict the best step to take next. CPG companies are dealing with enormous volumes of data on sales, shop inventories, deliveries and promotions at thousands of retail outlets. Using spreadsheets for tracking and analysis is time-consuming and spreadsheets can only do what you tell them to do. But Machine Learning automatically identifies common patterns and trends that would normally be difficult to uncover. For example, a ML solution can analyse data to predict the exact impact of a promotion in a major shop chain or determine the ROI of a loyalty programme at a certain shop. Understanding data at a granular level makes it easier to measure product performance, recognise issues and scale best practices across the board.

Microsoft Machine Learning

Machine Learning can significantly help route field employees more efficiently, automate repetitive manual processes and improve data analysis and insights across the organisation. Ultimately, these benefits help you keep up with the growth of your product market and make better decisions about promotions, campaigns and investments. Microsoft and its partners will continue to drive ongoing investments in Machine Learning and retail execution to help better position CPG companies for an increasingly digital age. To learn more about current in-market solutions for retail execution, take a look at AFS POP Retail Execution and AFS Retail Execution on AppSource today. [1]EKN Outlook, 2016