So we had a sure shot winner in 2019. Some of our customers in Cosmetics, Personal Care and Alcobrew (Liquor) industry have made real progress in constantly measuring the quality of visual merchandising execution at retail stores.
Before we dive into details, just to cover the bases –Visual merchandising is the practice in the retail industry of developing floor plans, planograms, and three-dimensional displays in order to maximize sales. The purpose of such visual merchandising is to attract, engage, and motivate the customer towards making a purchase.
Why no one was a winner in 2018:
Well, brands have done quite a lot traditionally and very successfully at that. No one is a winner because their effort has largely been restricted to a fewer key store whereas great visual merchandising can help across all categories of stores. On the other hand at key stores method of external audit on a periodic basis has left a lot desired in terms of day to day consistency and affordability of the exercise.
New method will throw many winners:
The new way is to make use of repeat visits of own sales reps and onsite product promoters at modern trade (large format) stores to receive visual feedback using mobile app technology. This allows for highly reliable data at affordability levels that makes it possible to scale it across all categories of stores.
What’s required of technology to make the new method work
Here are the key elements that are making our customers achieve success –
- Zero training mobile app interface to communicate to feet on the street about what picture evidence you require from them on every store visit
- Simple navigation and flow that does not take more than 2 minutes to provide the feedback
- Backend framework that allows a high degree of configurability to accommodate ever-changing campaigns across the category of stores. Feel like taking our 60-min-challenge? Define your campaign elements and see it in our app within 60 mins. Let’s do that.
- The ability for middle managers to assess the quality with simple browse for the close loop action
- Technology robustness to work offline or in interrupted network even if store activity involves multiple high-resolution pictures
There is a game-changer that’s playing itself out
Machine learning-based image processing is now making it possible to identify face counts from real shelf pictures of own and competing brands even with all the visual noise that exists across different stores. For certain product categories, our technology has reached a level of identifying 70% of objects in a picture with a reasonable level of training data. Drop-in your request here to see this in action.