Retail: Transformation and the great Undoing
The digital age has brought in an era where the distribution of information and goods have become less important consideration for businesses, compared to well almost everything else. That’s because the internet made it free.
What has become more important though, is the need to understand how people interact with a given piece of information, be it goods, services, news, media or almost anything else.
Look at Airbnb, Netflix, Google, Facebook, Uber, Flipkart or Amazon. All control the demand for abundant goods to varying degrees. To that end, some collect passive user intent while some collect active user intent of users looking for a product or a service. All these businesses are massively successful, and all have one thing in common. Their business models for controlling the supply of websites, videos, cabs, goods or homes are based on a common leverage; these platforms have aggregated and understood their user base and their needs extremely well. This understanding gives them superior buying and negotiating power with their suppliers.
If we look at retail, mainly offline retail in developed markets like the United States, it is facing many challenges when viewed from a new age aggregator lens. Offline retailer historically tried to solve the supply side and got to a point where there was an oversupply of goods(all oversupply is not bad, in fact it is excellent for enabling convenience a la The Everything Store). What they evidently lack is the ability to understand user demand like The Everything Store.
Now, if we assume that there will always be some level of oversupply in offline retail, it needs to be rearranged in a manner, that reflects changing user demand ‘on a unit level.’ This rearrangement is difficult to achieve in offline retail, as there are many problems to solve on the supply side alone like complicated vendor relationships, SKU optimization, Staffing and Store Operations, etc.
We’ll address the first point in Part 1 of this blog here.
1) Steps to understand user demand on a unit and aggregate level
In online retail, it is entirely possible to understand a shopper’s intent. People search for things they need. Even if they don’t buy them then and there, it is an excellent input for their recommendation systems. It is possible for an online retailer to map correlations between site visits, visit durations, historical product review and come up with a recommendation system to target their existing shopper base. They can also show you products that ‘Other people also Bought’ ;)
A good grasp on aggregate user demand, helps online retailers have better relationship with their suppliers. Both parties win, and everyone enjoys business certainties.
In offline retail, there aren’t good enough technologies that enable understanding of unit level user demand, apart from a post-facto analysis on sales data. Nothing helps in understanding what people were searching for in stores. Consider this simple example; a typical Walmart store carries 75,000 SKUs on an average. The sheer number of product combinations to consider are 75000 factorial(10³³³⁰⁶⁰ product combinations). That number is so big that it is larger than the number of atoms in the entire observable universe(10⁸⁰ atoms). Now, even if you cluster them into classes, there are still too many product combinations to consider. The offline retailer needs to adopt technologies that help them in understanding something as simple as - “users looked at these products in their stores.”
On the lines of Amazon, offline Retail needs a significant bit of undoing. Integrating merchandising with user relationships would be a good starting point. By taking this approach retailers will be undoing the integration of store merchandise with demand and distribution models that are archaic and don’t have the complexity to model demand ‘on a unit level’.
2)Using this understanding to integrate user relationship with however their industry generates revenue
To be continued.