Last year on this blog, I announced the GAIN Business Model Research (BMR) Project: funded by the Netherlands Ministry of Foreign Affairs, this project aims to identify promising business models to reach consumers on lower incomes with nutritious foods. Such approaches could help to improve the quality of diets—which currently are often lacking in food diversity and quality—if they can provide food that meets customers’ needs at an affordable price. But they must do so in ways that are profitable and sustainable for the company. How can they do this?
To find out, the BMR project undertook a systematic review of existing research and evidence. We screened about 8,000 documents before selecting 74 documents, and 99 case-study companies, that were eligible for inclusion. We included companies producing highly nutritious foods as well as those producing less nutritious alternatives—with the understanding that they might offer insights that could be applied to nutritious foods. We then analysed these documents to extract and synthesise 13 specific approaches used by firms to reach lower-income consumers with food and beverage products. Here, we discuss one of these approaches.
Final distribution to the consumer is a key challenge in addressing affordability for lower-income consumers, particularly in rural and remote areas or crowded lower-income urban neighbourhoods. One way to address this, particularly in places where there are few existing retail outlets, is through creation of a bespoke last-mile distribution (LMD) network. Creating such a network involves recruiting and training distribution/sales agents; equipping them for the job; and providing them with appropriate pay, incentives, and supervision. The agents can be employees, contactors, or micro-franchisees and can sell on foot, bike, using a pushcart or adapted motorcycle, from their homes, or through stalls. The product portfolio that they sell can exclusively belong to the business owner or include other goods. Incentives and motivations for workers can vary by context. For example, one comparative study found that earning income was the main driver for distributors in Ethiopia, whereas for those in Bangladesh it was gaining social status in the community. Many LMD networks have a focus on women, aiming to both empower women as sellers and reach women as consumers; this approach may have particular benefits for nutritious foods, given women’s oft-central role in cooking.
One example is the Danone Kiteiras model in Brazil, which employs women from low-income neighbourhoods to promote and sell Danone dairy products door-to-door in their communities. As of 2018, it was reported to have a continuously expanding network of saleswomen as well as increasing sales, selling around 148 tons of dairy products per month and reaching an estimated 80,000 consumers. Unilever uses a network of over a million ‘Shokti Ammas’ and ‘Shoktimaans’ to sell their products in India, applying similar models in Egypt, Vietnam, Sri Lanka, Pakistan, Ethiopia, Colombia, Nigeria, and other markets. Throughout West Africa, dairy processor FanMilk sells frozen dairy desserts through about 25,000 mobile street vendors who reach lower-income urban and peri-urban areas, including customers who do not have freezers and thus cannot store such frozen foods. In Venezuela, Coca Cola adopted a home-based LMD network by installing branded refrigerators in 30,000 homes in low-income areas, enabling their residents to sell Coke products from their homes. The approach proved popular with consumers, and to help increase the profitability for the home retailers and thus expand the network, Coca Cola also actively sought out other products for them to sell (e.g., phone cards, cosmetics).
Considering small firms, in Ethiopia, GUTS Agro-industries, a processor of fortified porridge products, runs a network of women entrepreneurs who sell door-to-door to households and retailers in low-income areas; the sellers are given branded uniforms, custom tricycles, and training and may also sell other, non-competing products. A Nigerian baby food company, BabyGrubz, uses a network of women who are paid to be exclusive distributors of the company’s complementary foods while providing peer-to-peer mentoring on the importance of nutritious foods and exclusive breastfeeding.
Such networks have numerous advantages. They can reach consumers where they are, increasing the products’ convenience; reach areas where existing retail does not; fill logistics gaps in the value chain; provide marketing; gain consumer feedback; allow for targeting specific customer segments (e.g., parents of infants); and give owners considerable control over pricing and other aspects. Because they allow for active engagement and one-on-one marketing — which traditional retailers are often unable or unwilling to provide — they can be particularly helpful with a product that is new or requires explanation or consumer education (‘high-touch’ products). They can also be useful for products with unproven demand (‘high push’ products) or which require special equipment to sell (e.g., a cold chain or heating system), as traditional retailers are often risk-averse and thus unwilling to make such investments. The use of a known, trusted person as a promoter can help inspire trust in a new product among consumers. The convenience aspect can be particularly strong for ready-to-eat products and those that are bulky or heavy, as consumers may be willing to pay more for the direct vending service. Bespoke LMD networks can also complement existing retail channels to reach additional consumers (e.g., those who might not shop in supermarkets); this can help limit ‘cannibalisation’ of one sales channel with another.
At the same time, bespoke LMD networks also have numerous downsides. They are complex and require a significant investment of time and money to set up and run (particularly with regards to supervision and incentivisation), typically more so than traditional retail (due to higher management costs and the need to offer higher margins to retain staff who sell few products). Building a dedicated LMD network in rural areas, in particular, can be very costly due to the lack of residential density and small transaction sizes. These costs may be offset by reduced marketing costs or a large market size, but this is not guaranteed. It can take years before a network is up and running at scale. Sales agents often have both low volumes and small margins, estimated at an average of 6-8 USD per day in revenue and 1-2 USD in profits. In rural areas, sales agents (of food and other fast-moving consumer goods) tend to have low earnings, making half as much as a comparable urban agent. Women-specific networks can be effective and cost-efficient but may have limited reach, usually only work in settings where the position requires little or no travelling, and are most effective for products marketed mostly to women. For the women themselves, income earned varies widely from marginal to substantial, but does not necessarily translate into empowerment.
For example, when Grameen Danone introduced a fortified yogurt targeting lower-income consumers in Bangladesh, it used a network of door-to-door-selling ‘Shokti ladies’ in rural areas. However, it struggled with considerable turnover and difficulty creating sufficient incentives and eventually shifted focus to a fixed-retail channel that mostly reaches urban consumers (who have lower nutrition vulnerability, though still being poor and at risk of micronutrient deficiencies). Similarly, through work in Burkina Faso, Vietnam, and Madagascar, the NutriDev project of the NGO GRET (which aimed to develop commercially viable fortified complementary foods) concluded that strengthening existing distribution networks, rather than creating new ones, was the optimal approach in most settings – but that specific LMD networks could work in dense urban areas.
In some cases, shared distribution channels can be created, reducing costs and time; these can include both for-profit channels (e.g., agents of another company with a product that is not a direct competitor) and non-profit channels (e.g., community health workers who earn additional income by selling for-profit goods, self-help groups). However, this needs to be the right fit: the existing offering must be compatible with new product(s), and sales agents must have time and local credibility. In the case of using non-profit channels, careful attention is needed to ensure alignment between the non-profit mission and the for-profit good, and in some cases community members have been reluctant to purchase items from health and NGO workers whom they associate with providing free goods and services.
From a nutrition perspective, bespoke LMD networks are useful for directly targeting consumers with specific nutritional needs (e.g., young children) and educating consumers on nutrition topics or product attributes. But they may be challenging for certain nutritious foods that are heavy, bulky, or require careful handling or storage. But for the right products, and where the economic case works for the business, the distributors, and the customers, LMD networks can offer a new way to get products out to consumers who might not otherwise be able to access them—while supporting the livelihoods of sellers who might not have many other employment opportunities.