Why "Africa coverage" usually means four countries

Ask a global expert network whether it covers Africa and the answer is yes. Ask it for someone who has run a distribution business in Côte d'Ivoire, or administered a mining licence in Zambia, and the answer arrives more slowly, or arrives as someone based in London who advised on the region some years ago.

This is not dishonesty. It is the predictable output of how a database network is built.

Databases accumulate where recruitment is cheap

A database network grows by recruiting experts who can be recruited efficiently: reachable by email, discoverable on professional platforms, comfortable with a self-service sign-up flow, and already familiar with the expert-call format. Those conditions are unevenly distributed. They cluster in a few markets with large formal corporate sectors, deep professional services industries and long exposure to institutional research.

Over a decade, the database fills unevenly. The network is not choosing to under-cover the rest of the continent; the recruitment mechanics chose for it.

Retrieval then hides the gap

The second mechanism is retrieval. When a brief arrives, the system searches what exists. A search always returns something. What it returns for an under-covered market is the nearest available match: a regional generalist, a diaspora professional whose direct experience ended some years ago, a consultant who has written about the market rather than operated in it.

The client sees a shortlist, not an absence. The gap is invisible precisely at the moment it matters.

Why the nearest match is not a small compromise

In markets where institutions are consistent and rules are applied uniformly, a regional generalist can be genuinely useful. The variance between neighbouring markets is low enough that transferred knowledge holds.

Across African markets that assumption is weak. Licensing practice, customs administration, local-content enforcement, land tenure, currency access and the informal share of a value chain differ sharply between countries that share a border, a language and a trading bloc. A view formed in one market can be confidently wrong in the next, and confidence is the problem: the client has no way to detect that the expert has crossed a boundary their experience does not cover.

What actually distinguishes coverage

Three questions separate real coverage from advertised coverage.

First, when was this person last in the market, in an operating or regulatory role rather than an advisory one. Recency matters more where rules change faster.

Second, at what level did they operate. Sector-level familiarity is not the same as having held a licence, negotiated with the ministry, or run the depot.

Third, what is the specific boundary of their knowledge. A good expert will tell you unprompted: "I can speak to the northern corridor; I cannot speak to the coastal market, that is a different set of relationships."

The alternative is to source, not retrieve

If the pool for a given question does not exist yet, the honest answer is to build it for that question: identify the seats that would produce the right experience, find the people who held them, approach them individually, screen them, and brief them. It is slower than a database query and it produces a materially different call.

It also means the coverage claim becomes testable. "Forty-seven countries" is only a meaningful statement if it describes where you are willing to source, with a stated turnaround, rather than where you happen to have profiles already sitting in a table.

What to ask a network

Ask for the last three experts they placed in the specific market you care about, at the seniority you need, with the recency of their experience. Ask whether those experts were retrieved or recruited. The answer tells you what "Africa coverage" means in that particular case, which is the only sense in which the phrase means anything at all.