We’re extremely proud to have recently announced a contract with DFM Foods India for geodemographic customer profiling and strategic market expansion. In the spirit of transparency with our stakeholders, which includes shareholders, employees, and potential clients, we’ve decided to publish this post to shed some light on how we approach the problem of geospatially oriented customer profiling.
As a technology company focusing on developing artificial intelligence-powered solutions, data is the lifeblood of our operations and sets the tone for how we think about problems and their potential solutions. Our solution to understand DFM’s potential market is grounded in our ability to answer a central question in a precise, data-driven way – why are things working well?
While this may seem like a counterintuitive starting point, answering this question lies at the heart of expanding the market for any business in any category. Tolstoy begins Anna Karenina with “all happy families are alike, but every unhappy family is unhappy in its own way”. The opposite might be true in business, as every failed company is alike – in that they either run out of capital or fail to grow quickly, but each successful business is successful in its own way – doing something well that nobody else is.
Contrary to the unidimensional nature of failure, success requires the superposition of multidimensional factors working well simultaneously. Our goal at DeepSpatial is to identify these specific factors and the degree to which they are contributing to success. Once identified and isolated, their success can be replicated.
Our Approach to Geodemographic Customer Profiling
Consider the following example: a retail chain operating in a city is bringing in 3x the revenue of other outlets under the operation of the company. Given that branding and product offerings are largely consisting, the difference in performance must be attributed to non-obvious factors
For instance, this store might be operating in an area that is population dense, thereby receiving a much higher proportion of foot traffic compared to other outlets.
Or perhaps it’s the case that the outperformance of this store is grounded in the fact that it is in a relatively richer neighborhood, where greater disposable income in the neighborhood is translating to a premium basket of goods being consumed, with more expensive product purchases explaining higher revenue.
Maybe the real reason is entirely different – the store is in a location where households in the neighboring radius inhabit four or more people, so consumption baskets are by nature larger and spread across more product categories.
The list of possible explanations is potentially endless. At DeepSpatial, we identify these factors and the degree to which they are affecting current performance. This is a two-step process.
1) We begin by isolating the successful geospatial context(s) that are contributing to success,
2) We identify the degree to which these factors might be relevant
The first step of the process is a data-enrichment process, where we rely on a diversity of data sources ranging from proprietary information to publicly available information to client data such as specific store sales/revenue numbers.
The second step of the process is analysis-driven, where deployed AI algorithms rapidly test out and identify relevant factors contributing to specific success.
Once these insights have been generated, our algorithms then screen for other geographies where similar conditions for success are present, and these locations serve our clients when it comes to identifying the right geographies for market expansion.
The screenshot above showcases our algorithm identifying areas with high-performance potential within a certain geography, which can be scaled to any geography and demographic profile.
Our Goal with DFM Foods
When you’re a $250+ million company growing at 20%+ year over year, maintaining your current rate of growth rate is hard. Increasing it along with your market share might be even harder. This is where we step in.
In a country like India, where language, culture, cuisine, and in effect, consumption patterns change every 50 km, it is imperative to isolate the various socioeconomic variables that are contributing to the success of a product to identify the areas where they can be replicated.
In the FMCG product categories, margins are thin, competition is intense, and products are highly differentiated. With DFM, we will be identifying and isolating the demographic factors that are contributing to the success of a product, and precisely locate untapped markets for potential expansion.
Geodemographic customer profiling of their rather large and diverse customer base will serve DFM in three key respects:
1) Identifying untapped markets: by isolating the most relevant factors contributing to success, we will be able to locate alternate markets with similar profiles ripe for expansion
2) New product launches: with each distinct geography having a differentiated customer profile, the company will be able to leverage these insights to launch products targeted at specific categories of consumers. This will allow the company to test new products in segmented geographies in a specific, highly actionable way, thus saving the costs associated with launching new products
3) Increased operational efficiency: in analyzing the customer profiles, we will interpret and analyze data from DFM’s 3,000+ distributors and 1.2 million+ retailers to create predictive models of projecting consumer demand, and the most efficient to manage and scale inventory
Our solution offering with DFM foods will bring us one step closer to achieving product-market fit and will serve as a meaningful example to more intricately demonstrate our capabilities and our core product value.
If you’re in retail wondering where to expand next, schedule a demo with us today or reach out to email@example.com