Understanding store activity doesn’t (only!) require consumer goods brands to channel the Selfridge’s spirit. They need data-driven insights.
Whether it’s developed markets or beyond, one top-of-the-agenda item for consumer packaged goods (CPGs) companies, is store level insight: the holy-grail. After all, manufacturers are a step away from the consumer (the retail store being closest to the consumer, with the exception of direct e-commerce), and knowing what happens at the store level, in a way, takes them closer to the consumers.
Historically, store level insight has meant snapshots of data made available by the sales force or select data sharing from retail customers. The other conventional sources of retail insight are syndicated point-of-sale scanner data and consumer panel data, which are not usually store level.
Traditional data sources miss the full picture on retail insights
A closer look at traditional sources of insights such as sales force audit, syndicated point-of-sale data and consumer panel data will tell you what their vulnerabilities are.
Salesforce audits take quite a bit of effort, lacks the desired precision in general, and costs more to scale up or improve accuracy. Despite these, it still doesn’t deliver comprehensive insight by itself.
Syndicated Point-of-Sale or Consumer Panel data sources rely on sampling a few thousand outlets across formats (or few hundred thousand households at best, for consumer panel) to project market level trends. While these have proven over decades to be enormously useful, they have had their challenges as well. There have been instances of sample size related concerns raised by CPG brands. Also, higher the level of granularity needed, higher the cost: and still it is at a market-area level, and not outlet level.
“Store-level insights”: in Hi-Definition
What if, a CPG could get critical insights into:
• What is selling where (exactly which store) and how much of it
• The interplay between pricing, assortments, store level displays and special promotions, in turn influencing market share (store level share of their brands vs category)
• Integrated and localized insights by being able to blend store level data with vicinity demographics
If your answer is, “we do it, already”, let me ask: is it just for your Top “10” (or N) customers or your broader market base?
Granular store-level insights will help brands move the right inventory to the right place and drive double-digit improvements to demand service levels.
There are technologies that enable store level data acquisition & insight that will ensure retail inventory meets demand while guarding brands against overstock and stock outs. The low hanging fruit is case fill/sell-through improvements (in double digits, as has been experienced by adopters).
Blended with other traditional sources, it has potential to deliver exponential value through improved efficiency of trade spends, and support consumer-centric merchandising. For these reasons, it has been seen that retail customers of CPG also willingly collaborate (an oft-cited constraint is the willingness of retailers in sharing data, but that is rapidly changing, if not already).
Global brands that make the most of the technologies enabling store-level insights can emerge as category captains.
Getting started on scalable store-level insights
Scalable ‘store level insight’ (vs just for a few top customers) is a hard nut to crack, and decision-makers at global brands know this for a fact. The challenges are many – different formats across retailers, the coordination effort, harmonizing data for different frequencies, to name a few. However, the benefits are too big to ignore: more so, as tomorrow’s leading CPG enterprises will derive competitive advantage through this key aspect.
A possible solution to balance the challenges vs benefits lies in leveraging cloud-based retail data acquisition platforms that come with adequate operational & data management support.
Combined with the right analytic themes, this help achieve a broad range of goals – improved alignment of inventory with sell-out trends to immediately impact case-fills, and better-localized consumption visibility leading to better assortments. This sets in a virtuous cycle that delivers better value to all stakeholders.
If brands could combine data from such a source with demand signals from other sources such as social media, they could sense localized demand signals, take ‘stock’ at a store level, and suggest incremental orders where necessary – resulting in improved responsiveness
Brands that will win at each store, would have also found a way for their retail execution to leverage store-level insights, balancing execution with insight. The picture is clear & bright, with Hi-Definition Store Level Insight!
Try deep learning using MATLAB