Community credibility compounds. Reach does not.

Reach is the easiest thing to buy and the easiest thing to report. A media plan produces impressions, impressions produce a number, and the number goes in the deck. It is also, in most African markets, the least reliable predictor of whether anyone will buy.

What predicts buying here is whether the people a prospect already trusts have bought, and said so. That is true almost everywhere to some degree. It is decisive here, and the practical consequences for a growth budget are large enough that ignoring them wastes most of the spend.

Start with what people actually believe

Nielsen has asked the same question for two decades. In its 2021 Global Trust in Advertising study, covering 40,000 respondents across 56 countries, 88 percent said they trusted recommendations from people they know, ahead of every paid format tested. Paid formats do not fail. They start from further back, and the gap is roughly 50 percent between a personal recommendation and the lower ranked channels such as banner and mobile advertising.

Now add the local reading. Edelman's 2025 Trust Barometer found 77 percent of Kenyan employees trust their employer to do the right thing, with trust in business overall at 72 percent. It also found 71 percent of South Africans and 70 percent of Kenyans holding a moderate or higher sense of grievance, defined as the belief that business and government serve narrow interests at the expense of ordinary people. Both things are true at once. Institutions close to people are trusted. Institutions at a distance are not.

A brand that arrives as an institution is starting an argument. A brand that arrives through someone the buyer already trusts is finishing one.

The channel map is not the one on the media plan

Two structural facts decide where the work has to happen.

The first is distribution. Across most of the continent, traditional and informal retail still carries the overwhelming majority of consumer goods, more than 70 percent in aggregate and above 85 percent of total sales in markets such as Nigeria. South Africa is the exception, with modern retail carrying more than 65 percent. If you are selling into Lagos or Nairobi, the shopkeeper is a channel, and in many categories a more important one than any platform.

The second is connectivity. GSMA Intelligence put smartphone adoption in Sub-Saharan Africa at 54 percent in 2024, the lowest of any region, forecast to reach 81 percent by 2030. Mobile internet penetration reached 27 percent by the end of 2023 with a usage gap of around 60 percent, meaning most people who could connect do not, mostly because handsets cost too much. A digital-only plan in these markets is a plan that reaches the wealthiest slice of the addressable audience and calls it the market.

What compounding actually looks like

Community credibility compounds because each interaction that goes well produces a person willing to vouch, and each person willing to vouch reduces the cost of persuading the next one. Reach does not compound. It resets the moment the flight ends. This is the same asymmetry Les Binet and Peter Field found in the IPA Databank, analysing 996 effectiveness case studies across 700 brands and more than 30 years: brand building creates slow effects that accumulate, sales activation creates fast effects that decay, and the brands that grew long-term market share split their budgets around 60 percent to the first and 40 percent to the second.

In African markets that 60 percent is not only advertising. A large share of it is the work of earning the right to be recommended:

• Relationships with the traders, stockists and distributors who decide whether a customer ever sees the product.
• Presence at the gatherings, associations and community structures where category decisions get discussed before they get made.
• Earned media in the outlets a buyer already reads, which behaves like an endorsement rather than an interruption.
• Customers who are visibly looked after, because a visible resolution is worth more than the complaint cost.
• Founders and executives who are known, quotable and reachable in their sector.

How to run it as a commercial programme

The objection to all of this is that it cannot be measured, so it cannot be managed. That is a reporting failure, not a truth about the channel. Credibility work produces measurable leading indicators if you decide in advance what they are.

Set a referral rate and track it monthly. Count named stockists and the reorder rate within each. Track share of category conversation in the outlets your buyers read, not total mentions. Ask every inbound lead how they first heard of you and record the answer in the CRM instead of in someone's memory. Watch the cost of acquisition over four quarters rather than one, because the whole claim is that it should fall as credibility accumulates. If it does not fall, the programme is not working and you should say so.

None of this argues against paid media. Paid media is how a credible brand gets in front of people faster than word of mouth alone can carry it. The order matters. Buying reach for a brand nobody can vouch for is paying to accelerate an argument you have not won yet.

The test is simple enough to apply this week. Ask your last ten customers who or what made them comfortable enough to buy. If the answer is a person, your growth budget is probably pointed at the wrong thing.