When a bank engages a consultant on agribusiness lending, the conversation often opens on the wrong subject. The consultant talks about training, frameworks, and capability. The chief executive is thinking about the loan book, the cost of funds, and where next year's growth comes from. The two are not talking about the same thing, and the gap matters.

A chief executive does not lie awake worrying that the team lacks an agribusiness framework. They worry that the portfolio is concentrated, that growth has flattened, that a competitor is taking the deposits, and that the regulator is asking harder questions about sector exposure. Agribusiness lending, done well, speaks to every one of those concerns. The work only earns attention when it is framed against them.

The outcome is the product

The deliverable a bank values is not a report on value chains. It is a larger, better-diversified portfolio that performs. It is fee income from transaction flows the bank was not capturing. It is the deposits that sit alongside a lending relationship. It is a defensible position in a sector a competitor has not yet understood. The analysis, the frameworks, and the capability building are the means. The growth is the product.

Why the framing changes the work

This is not a presentation trick. Framing the work against the outcome changes what the work has to include. If the objective is a performing portfolio rather than a trained team, the engagement cannot stop at knowledge transfer. It has to run through product design, pilot, and the first cohort of live lending, because that is where the outcome is actually produced. An engagement scoped around the outcome looks different, costs differently, and is judged differently.

What this means for how we work

We start every agribusiness engagement from the bank's own numbers and the growth it is trying to reach, not from a generic curriculum. The sector knowledge, the value chain mapping, and the capability building all still happen. They are simply pointed at a commercial result the chief executive already cares about, and the engagement is not finished until that result is in sight.

Agriculture is the largest employer across much of East and Sub-Saharan Africa and a major share of output, yet it remains chronically underserved by formal lending. The usual explanation is that agriculture is simply too risky. That is too easy. Risk is real, but it is priced and managed in far harder sectors. The more honest explanation is that most lenders are held back by three specific constraints, and each is fixable.

Limited sector knowledge

Most credit teams are trained to read balance sheets, not value chains. They can assess a borrower but not the system the borrower sits in: who supplies the inputs, who aggregates the output, where the cash actually moves, and which points in the chain carry the real risk. Without that view, every agricultural exposure looks like an undifferentiated bet on a farmer and the weather, and gets priced accordingly. The knowledge gap, not the risk, is what makes the sector look unbankable.

Poor product fit

Agricultural cash flows are seasonal, lumpy, and tied to harvest and off-take cycles. A standard working-capital facility or a monthly-repayment term loan fits none of that. When the product does not match the cash flow, default follows, and the default confirms the bank's prior belief that the sector is too risky. The product, not the borrower, was the problem.

Weak market connectivity

Agricultural lending works best when the lender is connected to the off-takers, aggregators, and input suppliers that anchor the chain. These relationships de-risk lending, because they create visibility of contracts, prices, and volumes. Most banks approach agriculture as a series of isolated borrowers rather than as participants in a connected system, and forgo the risk reduction that connectivity provides.

Naming them is the first step

None of these three constraints is intractable. Sector knowledge can be built. Products can be designed around real cash-flow patterns. Market connections can be established deliberately. What stops most banks is that the three are rarely named precisely, so the response stays vague. A diagnostic that isolates which constraint is binding, and by how much, turns a general sense that agriculture is hard into a specific, addressable plan.

A great deal of technical assistance ends at the workshop. The team is trained, the framework is handed over, the final report is filed, and the engagement closes. Everyone has done what they agreed to do. And often, nothing changes in the loan book. The knowledge was transferred and then sat unused, because knowledge is not where the value is.

Income comes from deployment

A bank does not earn a return on a trained team or a well-designed framework. It earns a return on products that are in market and portfolios that perform. Everything between the training and the performing portfolio, product design, pricing, pilot, the first cohort of borrowers, the early monitoring, is where the value is actually created. Stopping at knowledge transfer stops one step short of the point.

Why support has to run to the portfolio

The hardest part of agribusiness lending is not understanding the sector. It is the translation from understanding into a live, priced, risk-managed product that real borrowers take up and repay. That translation is where most initiatives stall, and it is exactly the part that ends-at-the-workshop support leaves undone. Support that runs all the way to the portfolio carries the bank through the translation, not just up to it.

What running to the portfolio looks like

It means designing the product, not just describing the opportunity. It means pricing and structuring against real cash-flow patterns. It means standing up a pilot with success measures, watching the first cohort, and adjusting before scale. And it means staying engaged through the early monitoring, when the portfolio is small enough to fix and the lessons are cheapest to learn. The objective is not a team that knows about agribusiness lending. It is a bank that is doing it, and doing it well.

There is a difference between understanding finance and understanding a sector, and it is larger than it looks. A banker can read any company's accounts. Reading a sector is a different discipline: working out, across the whole chain of participants, who actually makes money, who carries the real risk, and how cash and goods move from one end to the other. That is an operator's view, not an analyst's, and it is rarer than it should be.

Who makes money

In any value chain, margin is not evenly distributed. It pools at particular points, where there is scarcity, concentration, or a hard-to-replicate capability, and it is thin everywhere else. A lender that knows where the margin pools knows where the bankable, repayable cash flows sit. A lender that does not will keep lending against the thin parts of the chain and wondering why the portfolio underperforms.

Who carries the risk

Risk, like margin, sits unevenly along a chain. The participant who holds inventory through a price swing, or who has paid for inputs before the off-taker has committed, is carrying a risk that the contract structure may or may not compensate. Reading a sector means seeing where the risk actually lands, which is often not where the formal documents say it does, and pricing and structuring accordingly.

How cash actually moves

The single most useful thing to know about a sector is the real rhythm of its cash: when money comes in, when it goes out, how long the gaps are, and what bridges them today. Products that match that rhythm get repaid. Products that ignore it default. Almost every avoidable loss in sector lending traces back to a financing structure that was designed without knowing how the cash actually moves.

The discipline, not the sector

This way of reading a sector is not specific to agriculture. It is a commercial discipline that applies to any industry a lender wants to serve well. The reason it matters so much in agribusiness is that the sector is large, underserved, and badly misread, so the operator's view is both scarce and valuable. The discipline travels; the opportunity is simply largest where the misreading is greatest.

When retail growth flattens, the instinct is to add capacity: more branches, more relationship managers, more marketing spend. Sometimes that works. More often it adds cost without adding growth, because the constraint was never capacity. The constraint sits in the acquisition engine and the distribution mix, and adding headcount to a weak engine simply makes the weakness more expensive.

The acquisition engine

Most banks cannot say, with confidence, where their profitable customers actually come from: which channels, which segments, at what cost, with what retention. Without that, growth investment is a guess. The work of reviving stalled growth usually begins not with adding anything, but with tracing the existing customer value path end to end, and finding the points where good customers are being lost or never reached.

The distribution mix

The branch is an expensive way to reach a customer, and for many segments it is no longer the right one. Banks that grow profitably have deliberately built a mix, branch, agent, digital, partnership, matched to where their target customers are and what they cost to serve. Banks that have stalled are often over-weighted to the most expensive channel and under-built in the cheaper ones, and no amount of additional branch headcount fixes that.

Why headcount is the wrong lever

Adding staff to a weak acquisition engine raises the cost of every customer without raising the number of good ones. The growth that follows, if any, comes at a cost-to-serve that erodes the economics further. The lever that works is not more of the same capacity, but a sharper read on which segments and channels actually create value, and a deliberate shift of effort toward them.

Fixing the engine first

The sequence matters. Diagnose where growth leaks before deciding what to add. Rebuild the acquisition approach around the segments where the economics work. Get the channel mix right. Only then does additional capacity pay back, because it is being added to an engine that converts it into profitable growth rather than into cost.

Walk into most organisations that have missed their strategy and you will find that the strategy itself was sound. The market read was reasonable, the priorities were sensible, the plan was coherent. It failed anyway, in execution, and almost always for the same reason: no one owned the number, and there was no rhythm that made the absence of ownership visible until it was too late.

The plan is the easy part

Producing a strategy is a finite, well-understood exercise. A capable team, given time, will produce a credible plan. Executing it is a different kind of work: sustained, unglamorous, and dependent on dozens of people doing what they said they would, quarter after quarter, while the rest of the business competes for their attention. The difficulty is not designing the right thing. It is holding the organisation to it.

Owning the number

Execution discipline rests on a simple, hard practice: every initiative and every target has a named owner who is accountable for it, and that accountability is real. Not a working group, not a function, a person. When ownership is diffuse, slippage has no address, and slippage with no address compounds quietly until the year is lost. Naming the owner is the single most decisive execution move, and the one most often skipped.

The review rhythm

Ownership only works if it is exercised. A regular review rhythm, where owners report against their numbers and variances are surfaced early, is what turns accountability from a line in a document into a force that actually moves delivery. The rhythm does not need to be elaborate. It needs to be consistent, honest, and frequent enough that problems are caught while they are still cheap to fix.

Scorecards make it visible

Scorecards tie the initiatives and KPIs to the strategic objectives and put the state of delivery in front of the people accountable for it. They are not a reporting overhead. They are the instrument that makes execution legible, so that focus lands on the few things that are off track rather than being spread evenly across everything. Strategy rarely fails in design. It fails in the months after, and the months after are what we help organisations hold.

For a long time, retail banking growth and branch expansion were the same conversation. Reach a new market, open a branch. That equation has broken. The branch is now one channel among several, and for many customers and products it is the most expensive one. Growth has become a distribution question, how to reach more customers at a lower cost to serve, and the answer is rarely another branch.

Channel mix as a strategic choice

The banks that grow profitably treat channel mix as a deliberate decision, not an accident of history. Branch, agent network, digital, and partnership each have a cost to serve and a set of customers and needs they fit. The right mix puts each channel where its economics work: high-value, complex needs where a branch or relationship manager earns its cost; everyday transactions where an agent or a phone is far cheaper and just as effective.

The economics decide

Every channel has a cost to acquire and a cost to serve, and those numbers, not preference or habit, should decide where volume is steered. A transaction that costs a fraction to serve through an agent or an app does not belong in a branch. When a bank understands its channel economics clearly, the growth strategy almost writes itself: move reach and routine volume to the low-cost channels, and reserve the expensive channels for what genuinely needs them.

Agent and digital reach

For extending reach without extending cost, agent networks and digital channels are the decisive tools. A well-designed agent network reaches customers a branch never could, at a fraction of the fixed cost. Digital channels carry routine volume at lower marginal cost still. Neither replaces the branch entirely; both change what the branch is for, and free it to do the smaller amount of work that justifies its cost.

Distribution, not branches

The growth question is not how many branches to open. It is how to assemble a mix of channels that reaches the target customers profitably. Framed that way, the answer is usually a rebalancing, less weight on the most expensive channel, more on the reach and economics of agent and digital, rather than another round of branch investment that adds cost faster than it adds growth.