What it does
This KPI evaluates the visual consistency of packaging within a brand or product range on retail shelves. Specifically, it measures the visual consistency between a designated "critical SKU" and other SKUs within the same brand or range, providing a quantitative basis for assessing brand coherence and packaging alignment.
Case Example: let's say we test a new pack design in our potato range, from brown to red.
What is the consistency to the current range (all the other potato SKUs) and to the brand in total (across both ranges the brand offers)?
Left: The current design has high consistency to the potato range (score: 87) and the brand in total (84)
Right: the new red design has much lower consistency to both the potato range (-17% to 70) as well as the brand (-15% to 69).
Why it matters
Consistent packaging helps build a strong, recognizable brand block on the shelf, which is critical for effective brand activation. However, too much similarity can make it difficult to differentiate products and make navigation within a brand block a tedious task for shoppers. This KPI helps strike the right balance, ensuring a high level of visual consistency that supports brand recognition without compromising the shopper's ability to easily differentiate between products.
How it works
Advanced deep learning models are used to identify all SKUs on a shelf and then measure the visual consistency between the critical SKU and others in the same group. This analysis focuses on ensuring consistency within brand or assortment groups. Adjustments are made to account for bias if the critical SKU is part of the group being evaluated. The output is presented in a tabular format, providing insight into the visual coherence of packaging within each group.
How to achieve great results
- Design packaging elements such as colors, fonts, and imagery to enhance visual harmony across brand or range groups.
- Ensure that key visual cues remain consistent while allowing for differentiation to avoid visual clutter on the shelf.
- Introduce subtle variations in packaging design to maintain brand identity while allowing for easy differentiation between products within the same group.
- Experiment with elements such as graphics, logos, or product placement to strike the right balance between consistency and variety.
AI models used
This KPI uses state-of-the-art deep learning models, which have been trained on a dataset of millions of images. These models transform SKU images into numerical vectors that enable accurate consistency measurements between products. It also includes a specialized deep learning model trained on millions of annotated SKUs from retail shelves. This model excels at identifying and cataloging each SKU from an uploaded shelf image, ensuring detailed and accurate analysis.