A business can look busy while remaining difficult to grow. Orders may arrive, staff may work long hours and customers may recognize the brand, yet the owner still struggles to decide which problem deserves attention first. Is the constraint access to resources, an unreliable process, weak information or uncertainty outside the firm’s control? A broad growth plan is less useful when it treats these different problems as one.
Recent evidence on Ethiopia’s formal private sector offers a reason to ask more precise questions. It can help an owner avoid assuming that a challenge faced by one type of firm describes every enterprise. This article uses that evidence as context for an internal diagnosis. It is not a lending recommendation or a forecast of individual business performance. The worked examples are hypothetical and are intended to show how a small team can organize a decision before committing to a project.
What the new firm evidence actually covers
The World Bank’s July 2026 analysis draws on the 2025 Ethiopia Enterprise Survey, with interviews involving 1,011 formal private firms in manufacturing and services. It reports a pronounced size divide in credit constraints, including 75.9 percent of small firms compared with 28.3 percent of large firms. The coverage matters: these findings do not represent every informal enterprise, household business or sector in the country. [World Bank: What new Enterprise Survey data reveal about Ethiopia’s firm-size divide]
The same analysis cautions against treating finance as a complete explanation of investment behaviour. That is a useful boundary for business planning. National evidence can identify important pressures and questions; it cannot establish which specific action will improve a particular firm. Use the survey to sharpen an investigation, then use the firm’s own records to understand the decision in front of it. A good diagnosis joins the two levels without pretending that either one is sufficient alone.
Separate symptoms from constraints
An owner might describe the problem as slow growth, but that phrase does not identify a process that can be changed. Ask what the team observes. Are orders being declined, deliveries delayed, repeat purchases falling or quotations taking too long? Define the symptom in terms that staff can check.
Imagine a hypothetical maintenance business with enough enquiries but few completed jobs. Several explanations are possible. Customers may not accept quotations, spare parts may arrive late or staff schedules may leave appointments unconfirmed. Each explanation points toward a different intervention. A marketing campaign could increase enquiries while making the backlog worse. A new scheduling tool could help only if someone maintains accurate availability. Before choosing a solution, follow a small sample of jobs from enquiry to completion and identify where the process actually loses time or customers.
Build a basic operating picture
Start with records the business already uses. A short review can connect enquiries, quotations, confirmed orders, delivery records and customer issues. The objective is not to create a sophisticated dashboard immediately. It is to establish whether the team can describe the same operating period consistently.
Agree a few definitions. What counts as an enquiry? When does an order become confirmed? Is a cancelled order recorded separately from an unsuccessful quotation? A change in definitions can look like a change in performance even when the underlying work is similar. Record missing information explicitly and avoid manufacturing precision from incomplete files. If only part of the business has dependable records, state that limitation. A modest but traceable picture provides a stronger starting point for advice than a large spreadsheet whose totals cannot be explained by the people responsible for the work.
Distinguish an information gap from a resource gap
Some decisions are difficult because the business lacks resources. Others are difficult because it cannot see how existing resources are being used. These situations can occur together, but the response should not be identical. Ask what evidence would distinguish them.
For the maintenance business, uncompleted jobs may be waiting for a part that genuinely cannot be obtained within the promised period. That is different from a part already in stock but not linked to the correct job. A simple stock check cannot solve every supply constraint, yet it can prevent the team from misidentifying the problem. Similarly, a request for additional staff should be considered alongside work allocation and avoidable rework. The purpose is not to dismiss external constraints. It is to ensure that an improvement project addresses a problem the proposed activity can plausibly influence.
Make uncertainty visible in the plan
A growth plan often presents one expected future as though all assumptions are equally dependable. Instead, identify the assumptions that matter most to the next decision. These might include a supplier’s delivery reliability, a customer’s order timing or the availability of a required service.
Describe what the team would observe if an assumption stopped holding. Then define an action the business can realistically take. This is different from claiming to predict the external environment. The 2026 World Bank partnership framework places attention on business conditions, infrastructure, skills and job creation, which reinforces the importance of looking beyond a single firm-level metric. It does not provide a firm-specific outcome guarantee. [World Bank: Ethiopia Country Partnership Framework announcement, June 2026] A practical plan should show which actions belong to the business, which depend on another party and which require a decision to wait, narrow the scope or seek additional information.
Test the bottleneck with a small change
Once the team has a plausible explanation, design a limited test. Suppose quotations are slow because staff repeatedly ask the owner for standard information. The business could prepare an approved service information sheet and a quotation checklist for a defined set of jobs. Keep unusual cases on the existing approval route.
Measure what happens to accepted quotations, correction time and customer questions during the test. Do not use the number of templates produced as proof of improvement. If the change helps, document the conditions under which it worked. If it fails, examine whether the explanation was wrong, the process was not followed or the measure was unsuitable. A small experiment does not eliminate the need for judgement. It creates an observable link between a proposed explanation and an operating result, which makes the next decision easier to discuss.
Give a business adviser a decision to support
“Help us grow” is a broad invitation that can produce an equally broad report. A more useful brief identifies the operating symptom, the records available and the decision the owner needs to make. Explain which constraints the business already understands and which remain uncertain.
For example, the maintenance business could ask an adviser to examine the enquiry-to-booking process and propose a tested way to reduce unresolved quotations. The assignment could include a small record review, staff interviews and a revised handover procedure. Those are suggested methods, not a description of a real engagement. Agree how the adviser will distinguish evidence from assumptions. Also ask what the team will be able to do independently at completion. A report has more value when it changes a usable process than when it restates national challenges without connecting them to the client’s work.
Do not confuse a useful indicator with a growth promise
A shorter quotation cycle, clearer stock record or more consistent handover can be a meaningful improvement. None automatically proves that overall growth will follow. Demand, competition, supplier conditions and other factors may affect the outcome. Keep the language of evaluation proportional to what was measured.
Use a small set of indicators linked to the intervention. Record the starting point, the observation period and any important changes in the type of work being handled. If the owner changed prices during the same period, a change in customer conversion cannot be attributed to the new process alone. State such complications rather than hiding them. This discipline protects the business from buying a success story that cannot be checked. It also helps staff understand which part of the result they can influence and which part needs further investigation.
Turn the diagnosis into a working review rhythm
A diagnosis loses value if the resulting actions are not revisited. Choose a regular, short review that fits the business’s pace. Discuss outstanding exceptions, recent observations and the next action, with a named owner for each item. Keep the record simple enough that it is maintained during busy periods.
The review should be a place to resolve questions rather than a ceremony for presenting numbers. Ask whether a repeated exception points to a process problem, a missing decision or an external constraint. Close actions only when the agreed evidence is available. If the team keeps carrying the same action forward, reconsider whether it is specific, achievable and assigned to someone with the authority to complete it. The habit of reviewing a few meaningful observations can be more useful than repeatedly replacing one large growth plan with another.
Use national evidence without losing the individual business
Ethiopia’s firm-level evidence highlights real differences in the pressures faced by businesses. Its most useful contribution to an owner’s next move is not a universal formula but a warning against universal explanations. A small firm’s problem should be investigated at the scale of the decision it needs to make.
Start with a clear symptom, connect the records, separate the possible causes and test a manageable response. Use national research to understand context and limitations, while keeping the firm’s operating evidence visible. That approach will not remove every obstacle in the business environment. It can, however, help the owner choose an assignment that addresses a real question, judge whether the work helped and avoid spending scarce management attention on a solution that belongs to a different problem.
