Farmer Value Creation vs Last Mile Delivery

Key Messages

The business models of companies that fully rely on their own staff appear to create significantly more value for farmers than those that fully or partially rely on intermediaries. There is a lot of nuance behind these findings. Our data suggests that there are challenges related to the quality of service delivery when intermediaries do not have the same capabilities as a company’s own staff. At the same time, our data also suggests that intermediaries can help create stronger and larger smallholder-inclusive business models.

Our full analysis explores the following aspects in more detail:

  • The Quality of Service Provision
  • The Relationship between Farmers and Companies
  • The Reach of Service Provision and Market Access

Keep reading to find out more.

Understanding the role of last mile delivery in value creation at farm-level

Does using your own staff to deliver services to farmers lead to better quality outcomes at farm-level? That is the key question for this analysis, and the answer is somewhat complex. Using your own staff for last mile delivery logically improves the level of control you have over service delivery. However, control doesn’t necessarily equate to quality services and in turn doesn’t guarantee more farmer value created.

At a first glance, our results seem to confirm the original hypothesis, which states that companies using their own staff are associated with creating 43% and 96% more value at farm-level when compared to companies that rely on intermediaries or a combination of both respectively. Furthermore, our modelling results, in which we control for the estimated relationships between farmer value created and all other drivers that we have analyzed, confirm this relationship. They do, however, highlight further nuances. Read more about the results from our machine learning analysis.

Read more: How do we distinguish between different last mile delivery methods?


Understanding the relationship between how services are delivered and how value is created at farm-level is particularly important in the context of a company’s investment in farmers, and the amount of revenue that it recoups. We discovered that:

  1. For Service Delivery Cost per Farmer, using your own staff is (unsurprisingly) associated with a higher cost. Click here to read more about the analysis of last mile delivery versus investment per farmer.
  2. For Direct Cost Recovery, businesses that involve their own staff in last mile delivery, either as a standalone option or in combination with intermediaries, recoup more of their service costs through service payments compared with those business models that rely only on intermediaries. Click here to read more about the analysis of last mile delivery versus direct cost recovery.

Diving deeper: what do we think explains these results?

We see the Last Mile Delivery influencing Value Creation at farm-level in different ways, including:

  1. The Quality of Service Provision – The efficacy of services can be lower when delivered by intermediaries who may not have the same capabilities as a company’s staff.
  2. The Relationship Between Farmer and a Company– Intermediaries can either support or hinder trust between a farmer and a company.
  3. The Reach of Service Provision and Market Access – Leveraging intermediaries to support last mile delivery allows companies to better ensure that services reach remote and underserved groups of people.

Of the three factors above, the first highlights how using intermediaries could lead to lower value creation, while the second and third factors cautiously indicate the positive impact that intermediaries could have. However, the way in which companies use intermediaries can help ensure getting the best of both worlds. Well-trained, properly-resourced and effectively-incentivized intermediaries can reduce risks around service quality, fraud, and theft. We are therefore confident that the key priority here is to continue finding ways of empowering community-based intermediaries, whether they be agents, lead farmers or farmer organizations, or even Small and medium-sized enterprises (SMEs), and micro-entrepreneurs playing similar roles, to be able to provide the same or better quality of goods and services as company-employed staff.

Given our current research, we have determined what can make a more effective agent model (several of these also apply to lead farmers and farmer groups):

For more insight on best practices for last mile delivery and agent models, click here to read an innovation guide on Commission-based agent models.


Implications: So what does this mean for you?

Based on our current findings, we see the following implications for different audiences:


Reflections on data limitations and further research

The Insights Hub is a living document which we are constantly updating with new data, new analysis, validation by our partners, etc. For the results on this page, we would like to emphasize the following:

Major caveats and limitations of our current approach

Although we believe our analyses and insights offer a solid set of insights that can already be used to inform decision-making, there are a number of caveats that we wish to be open about.

  • 1. Focusing on $ value creation per farmer alone can give a narrow view on impact
  • 2. Our data looks at value creation for the average farmer
  • 3. Our quantitative analyses are only capturing the presence of last mile delivery options

How we plan to update these findings in the near future

  • 1. Incorporation of other indicators to look at value creation
  • 2. Investigation into how equal income improvements are within a model
  • 3. Consideration of the means by which value has been created

Suggestions for additional research by our peers and partners

  1. Incentive structures for agents and other intermediaries
  2. Comparative assessment of value created for intermediaries
  3. Research on the enabling environment conditions that are needed for agents to flourish

Strength of Relationship 3/5

  • Moderate relationship between driver and outcome variables
  • Results are consistent across analytical models used
  • Several limitations regarding sample or indicator