What kinds of questions can Kofa answer in 72 hours?

Questions about how a specific market, segment, or practice actually behaves on the ground — the kind best answered by a small number of practitioners who know it from the inside.

What questions can't be answered in 72 hours?

Questions that require a representative statistical sample, longitudinal tracking, or consensus across an entire industry — those need broader, slower methods.

How do you write a brief that stays on the right side of that line?

Define the decision, the specific unknown, and the people who would know it — tightly enough that a short, structured conversation can resolve it.

Dispatch · Seeker Guide · Vol. I

Five Questions You Can Answer in 72 Hours With a Practitioner Session — And Five You Can't

Kofa is not a substitute for every kind of research. Here is what it is genuinely built for — and what it is not the right tool for.

Aminu RabiuFounder, Kofa Insights Limited

August 2026

6 min read

Kofa is a software platform, not a research firm, a survey tool, or a way of producing statistically representative findings about a market. It is a way of reaching the person closest to a specific problem, the person whose judgment, experience, and operational knowledge can answer questions that data and desk research cannot.

That is a precise and limited claim — and precision matters here. Organisations that use Kofa well know exactly what kind of question to bring to a practitioner session, and what kind to answer some other way. Those that try to use it for everything find it serves them worse than those who use it for exactly what it's built for.

So before describing what Kofa can do in 72 hours, it's worth being equally clear about what it cannot.

Five questions you can answer in 72 hours

Each of these has something in common: the answer exists inside the experience of a specific kind of person, and their account of it is worth more than anything a survey, a dataset, or a desk report could produce.

Questions a practitioner session answers — with what 72 hours produces

Decision type · Market entry

"Why are businesses in this geography avoiding a credit product they formally qualify for?"

This is not a data question. Uptake data tells you how many applied; it can't tell you why those who qualified chose not to. The answer is usually operational — branch officer relationships, informal signals that tell experienced borrowers which lenders are genuinely accessible, seasonal dynamics that formal analysis does not capture. A loan officer who has interviewed hundreds of borrowers in that geography holds this answer in a way no survey could produce.

72 hrs producesA specific, conditional account of the actual barriers — named by someone who has watched them operate in practice. Enough to redesign the approach before it is deployed.

Decision type · Programme design

"What will actually prevent community participation in this programme — and what won't appear in the consultation records?"

Community consultations record agreement. They were not designed to surface the dynamics that determine whether that agreement is genuine. A cooperative leader or field officer who knows the community — its informal authorities, its historical disputes, its difference between public and private consent — can tell you what the consultation process was never able to ask.

72 hrs producesThe specific community dynamics the programme will encounter — named before implementation begins, when they can still change the design.

Decision type · Portfolio risk

"Where is the concentrated informal dependency in this supply chain that doesn't appear in the logistics assessment?"

Logistics assessments map routes, costs, and formal contracts. They rarely surface the unofficial arrangement at a specific point in the chain — the transporter relationship, the checkpoint dynamic, the seasonal dependency — that is invisible in any document and essential to whether the chain holds under pressure. A logistics manager who has worked that route knows.

72 hrs producesThe specific informal dependencies named and located — the basis for a risk assessment that reflects how the supply chain actually operates, not how it is documented to operate.

Decision type · Investment decision

"Is the assumption at the centre of our investment thesis accurate — and under what conditions does it stop being true?"

Investment theses rest on assumptions about how a market behaves. Those assumptions are usually derived from desk research and sector analysis. They are rarely tested against the operational reality of someone who has been inside the market long enough to see the assumptions break. A practitioner session is a low-cost way to pressure-test a thesis before it becomes expensive to be wrong about it.

72 hrs producesA practitioner's account of whether the assumption holds in their experience — and the conditions under which it does not. Evidence worth having before the decision is made — not proof.

Decision type · Rapid intelligence

"What is actually happening on the ground in this market right now — not six months ago when the last report was written?"

Reports take time to produce and describe a market as it was when the research was conducted. Policy shifts, seasonal dynamics, informal market changes, and community-level developments can change the operating reality significantly within that gap. A practitioner who is currently working in the market knows what a published report cannot.

72 hrs producesCurrent operating reality — not historical analysis. What has changed, what remains true, and what the most recent report missed because it happened after the fieldwork was completed.

Five questions you can't.

This section matters as much as the one above. Kofa is precise about what it does well partly because it is equally precise about what it doesn't. A tool that oversells its scope becomes less useful to the people who understand what they actually need.

The five question types below are the ones where a single practitioner session is the wrong tool on its own, not the ones where Kofa is simply less effective. The answer requires something one conversation cannot produce, however good the practitioner is. Kofa's network moves the edge: run the same brief across many practitioners and a few of these become answerable in a directional way. What stays true is that none of them become a substitute for a method built to produce defensible statistics.

Questions a practitioner session cannot reliably answer — and why

Wrong tool for · Statistical representation

"What percentage of SMEs in Tanzania face this barrier?"

One practitioner is one vantage point. Their account is deep, specific, and credible — but a single session is not a sample from which a defensible percentage can be derived. Run the same brief across a dozen practitioners in Kofa's network, though, and their accounts start to triangulate: a directional read on how prevalent the barrier is, which contexts it concentrates in, and where it breaks down. That is far more than a guess, though not a statistically representative figure. When the decision needs population-level statistics, commission the survey. It tells you what to measure before you spend on one.

Use insteadA properly designed survey with a representative sample — for the defensible number. Run the same brief across Kofa's network first: a dozen practitioner accounts triangulate a directional read on prevalence that tells you exactly what to measure, and where, before you spend on the survey.

Wrong tool for · Longitudinal trends

"How has credit access in this market changed over the last ten years?"

A practitioner's account of change over time is valuable context, not longitudinal data. Memory is selective, and the most salient changes are not necessarily the most statistically significant ones. A question that requires a trend line built from systematically collected data over time cannot be answered by recollection — however experienced the practitioner.

Use insteadPanel data, administrative records, or longitudinal research. The practitioner session can help you interpret what the trend data means — not replace it.

Wrong tool for · Causal inference

"Did this intervention cause the improvement in uptake — or was it something else?"

Practitioners observe. They do not run controlled experiments. A practitioner who saw uptake improve after an intervention has a view on why — and that view is worth taking seriously. But it cannot establish causation. If the answer requires separating the effect of the intervention from confounding factors, the methodology required is experimental or quasi-experimental, not conversational.

Use insteadRandomised evaluation, difference-in-differences, or other impact evaluation methodologies. The practitioner session can help design the evaluation — not replace it.

Wrong tool for · Baseline measurement

"What is the current level of financial inclusion in this district — as a baseline for our programme?"

A defensible baseline needs a methodology built for it — defined sampling and replicable measurement. One practitioner's estimate of current conditions is useful context, not a baseline. But a network engagement — the same brief run across the practitioners actually operating in the district — produces a directional current-state picture a single estimate cannot: where conditions concentrate, what the outliers look like, which assumptions the district quietly violates. Use that to shape the programme and to design the formal baseline survey that measures it properly.

Use insteadA structured baseline survey with defined sampling and measurement protocols — for the defensible figure. Run the brief across the network first to get a directional current-state picture (where conditions concentrate, which assumptions the district quietly violates) and to set the right indicators before the formal survey.

Wrong tool for · Regulatory compliance mapping

"What are the regulatory requirements for operating in this market — and are we compliant?"

A practitioner's working knowledge of the regulatory environment is valuable orientation, not legal advice or a compliance audit. The answer to a regulatory question may change depending on your specific structure, jurisdiction, and the most recent regulatory update. A question where the answer has legal consequences requires authoritative legal sources, not practitioner accounts.

Use insteadQualified legal counsel in the relevant jurisdiction. The practitioner session can give you the informal understanding of how regulation operates in practice — a valuable supplement, not a replacement.

Kofa answers the questions that have never been in any database. It does not answer the questions that databases were built to answer. Knowing the difference is the entire skill.

Why the line matters

The organisations that use practitioner sessions most effectively are usually the ones with the clearest sense of what kind of question they're bringing — not because they're more sophisticated, but because they've stopped expecting any single tool to answer every kind of question.

A practitioner session used for the wrong kind of question produces two problems at once. The conversation is less productive, because the question was never answerable in this format. And the answer carries more confidence than it should — a credible, articulate practitioner sounds authoritative even when the question they were asked requires a different kind of evidence entirely.

What happens when the brief is on the wrong side of the line — and the right side

Brief on the wrong side — what follows

The practitioner answers sincerely — but the question required a different methodology, and the answer sounds more authoritative than it should

The organisation acts on an answer that is credible-sounding but not of the kind that the question required

The session produces a finding that later needs to be defended — and the evidence base turns out to be a single conversation rather than what the decision actually needed

The wrong tool used confidently produces an outcome that looks like research and functions like a guess

The practitioner's real knowledge — which was genuinely valuable — is applied to a question it was never equipped to answer, and the actual question it could have answered goes unasked

Brief on the right side — what follows

The practitioner is asked exactly the question their experience is closest to — and answers with the specificity that only proximity produces

The organisation receives something it genuinely could not have gotten any other way — an account of operating reality that data and desk research do not reach

The session produces a decision asset that is specific, conditional, and honest about what it is — practitioner testimony, not statistical evidence

The decision is better because it accounts for operating reality that previously went unaccounted for — not because it replaced the other evidence it needed

The practitioner's knowledge is applied to the exact question it was built to answer — and that answer changes the decision in the way that only ground truth can

How to write a brief that stays on the right side of it

A good brief is a focused document, not a list of everything you want to know about a market: it names one decision, identifies what kind of knowledge would change it, and describes the specific experience a practitioner needs to hold in order to have that knowledge.

The five principles below are what separate briefs that produce sharp, usable decision assets from briefs that produce interesting conversations with no clear answer.

What makes a brief work — five principles with examples of each

Name one decision

"We want to understand the agricultural finance landscape in Tanzania."

One decision, named. "We are deciding whether to pilot a receivables-based lending product in Dar es Salaam before Q4. We need to understand whether local agro-dealers have the kind of receivable cycles the product requires."

Describe the practitioner specifically

"Someone with experience in agricultural finance in Tanzania."

Specific experience required. "A loan officer or credit analyst who has worked directly with agro-dealers in Tanzania in the last two years — ideally someone who has seen both repayment cycles and default patterns in that specific context."

Ask what only they can answer

"What are the main barriers to credit access for SMEs in Tanzania?" (Answerable from a desk report.)

Their vantage point only. "In your direct experience working with agro-dealers — what determines whether a dealer who qualifies for receivables-based lending actually applies for it? And what have you seen go wrong when they do?"

Invite conditions and exceptions

"How does the credit market work for agro-dealers in the north?"

Conditions welcomed. "We are particularly interested in situations where the usual dynamics don't hold — specific districts, seasons, or dealer profiles where the pattern breaks. That's often where the most important information is."

Tell them what you already know

(No context provided. The practitioner spends time establishing what the seeker already understands.)

Context shared upfront. "We have reviewed IFC data on agricultural SME credit in the north and spoken with two sector analysts. We believe the key constraint is the receivables cycle mismatch. We want to know whether that's accurate — and what we're missing."

The brief determines what the session can be. A broad brief produces a general conversation; a focused brief produces a specific answer. The practitioner's knowledge is the same in both cases — what differs is whether they're asked the question their knowledge was built to answer.

Here is what the difference looks like at the briefing stage, before a single session has happened:

The same topic — a brief that drifts and a brief that works

Brief that drifts

"We are looking to understand the challenges facing smallholder farmers accessing agricultural inputs in Tanzania. We would like to speak with someone who has relevant experience in this space, to understand the key barriers and what might be done about them."

→ Could describe anyone. Produces a general conversation. The practitioner does not know what decision it will inform, which barriers matter most to the seeker, or what level of specificity they need. The session will be interesting and produce nothing citeable.

Brief that works

"We are evaluating whether to fund a last-mile input distribution programme in Zambia. Our model assumes that agro-dealers in rural areas can reliably receive and resell certified seed within a two-week window. We need to speak with someone who has run input distribution in that geography to understand whether that assumption is realistic — and what breaks it when it fails."

→ Names one decision. Identifies exactly what experience the practitioner needs. Asks a question only operational proximity can answer. The practitioner arrives knowing what they're there for and which parts of their experience are relevant.

The second brief takes ten more minutes to write. It produces a session that is three times as useful. The practitioner arrives prepared for a specific question rather than a general conversation. The answer is specific, conditional, and directly tied to the decision it was designed to inform. The difference between a conversation that ends and a decision asset that stays is usually made before anyone joins the call, in the brief. And once it stays, it compounds: every decision asset becomes searchable, retrievable through Ask Kofa, and the next brief picks up where the last one left off.

One question before your next decision

On the last decision you made using practitioner intelligence — did the brief name the decision, or just the topic? And would a sharper brief have produced a more useful answer?

Aminu Rabiu

Founder, Kofa Insights Limited

Start a brief. The platform runs everything around it.

The brief form guides you through exactly the questions above — what decision you are making, what experience the practitioner needs, and what only they can answer. A structured decision asset in 72 hours. Where our network is active, Kofa finds the right people; anywhere else, bring your own.

Start a Brief→

Every brief that runs on the Kofa platform is reviewed before matching begins. If a brief is asking a question that a practitioner session is not built to answer well, Kofa will flag it — and work with you to refocus it before the engagement begins.