Fifty cents.
That was the difference between what the sales agent had promised verbally and what was actually deducted from the customer’s first premium.
In the boardroom, it was a rounding error. In the market where the customer actually lived, it was a betrayal. And it was the reason customers were canceling their insurance policies one month after signing up ; a pattern consistent enough that someone had finally put it on a slide deck and asked me to figure out what was going on.
Unfortunately, the customers we were losing weren’t a niche. They were the majority of the country.
Ghana’s economy runs on informal-sector workers: market women, taxi drivers, seamstresses, hairdressers, mobile money agents, mechanics, day laborers. People who pay close attention to small amounts of money because small amounts of money are how they survive. To the market woman selling tomatoes under the scorching Accra sun, fifty cents isn’t a fee. It’s a meal. It’s the difference between buying onions for dinner and not. And when it’s unexplained, it’s not a discrepancy, it’s a signal: "They lied to me. They always lie. I should never have trusted this." So she walks. She tells her neighbors. Her neighbors never sign up. And somewhere in a quarterly report, the company decides this segment is “low-value” and “hard to retain” without ever asking why.
The 50-cent discrepancy was happening because the products had been designed for a customer profile that didn’t exist in any meaningful volume, the salaried, English-fluent, branch-walking customer who looks past small fees because their cash flow can absorb them. The actual customer base was the opposite. Cash-flow-tight. Often non-literate or low-literate in English, though fluent in Twi, Ga, Ewe. Conducting business in person, in markets, with handwritten receipts. The product wasn’t being rejected. It was rejecting them through a hundred tiny design decisions that assumed away who they were.
We didn’t just fix the 50 cents. We rebuilt the system that produced it.
- We redesigned the onboarding forms. Out went the legal jargon, the dense paragraphs, the assumed reading level. We rewrote everything so a six-year-old could follow it. Not because our customers were children, they were grown adults running their own businesses, but because the literacy the original forms assumed wasn’t the literacy our customers actually had. Plain language is not condescension. Plain language is respect.
- We gamified the premium calculation. The old way: the agent quotes a number, the customer either accepts it or doesn’t. The new way: a simple, tactile artifact the agent walks the customer through, step by step, that turns the premium calculation into a conversation. Customers could see the math. They could ask questions. They could spot the 50 cents before it became a betrayal. The artifact became the most important thing in the agent’s bag because it made the customer feel like a participant instead of a target.
- We listened to what the agents were actually working with. Our research surfaced something the head office had completely missed: about 80% of the insurance agents in the field were using brick phones. Not smartphones. Brick phones. Any “digital transformation” strategy that began with an app was dead on arrival. the agents were not resistant to technology, the technology being designed for them assumed a device they didn’t own and couldn’t afford. We designed for the phones they actually had. Then we designed for the literacy levels our customers actually had. Then we designed for the conversation that actually happened in a market stall.
- We built insurance products that fit the customers and not the other way around. This was the hardest fight and the most important one. Instead of taking the existing product portfolio and trying to retrofit it for blue-collar customers, we built new products from the ground up around how informal-sector workers actually earn money, save money, and absorb risk. Different premium cadences. Different claim flows. Different language. Different everything.
The outcome: millions in new premium revenue in the first three months after rollout. The segment the boardroom had written off as “flighty” and “low-value” turned out to be one of the most profitable customer bases the company had ever activated.
Why “small” segments get ignored, and why the business case for ignoring them is almost always wrong. Every company I’ve worked with has a version of this conversation: “That segment is too small to prioritize.” Sometimes it’s true. Often it isn’t. Here’s what I’ve learned to listen for when someone says it:
- “Small” is usually measured in revenue per customer, not lifetime risk to the brand. A customer who feels cheated by 50 cents doesn’t just leave. They warn their network. In markets where word-of-mouth is the primary acquisition channel, which is most of the world, the cost of one betrayed customer is dozens of customers who never sign up. The companies dismissing this segment as “low-value” are paying for it in customer acquisition costs they never trace back to the source.
- “Hard to serve” is often “we never designed for them.” Insurance, banking, telecom, they all complain that low-income, informal-sector customers are expensive to acquire and retain. What’s actually expensive is serving them with products designed for someone else. When you redesign the product itself for the customer who actually shows up, the unit economics shift. The “problem segment” becomes one of the most predictable ones.
- The segment that’s “too small to prioritize” is almost always the segment that compounds the fastest. Informal-sector workers in West Africa aren’t a fringe market, they’re 80% of the workforce. The companies that figured this out in mobile money (M-Pesa, MTN Mobile Money) didn’t get rich serving the salaried elite. They got rich because they finally noticed that the “small” segment was actually the whole market wearing a costume.
- The 50 cents is never about the 50 cents. It’s about whether the institution is paying attention. The customers who cancel over 50 cents are telling you something diagnostic about your entire product. If you fix the 50 cents and not the underlying disrespect that produced it, you’ve solved nothing. The next discrepancy will surface the same exit behavior.
A note on who’s in the room.
I want to be honest about something the FASPE application I wrote this year forced me to confront.
I am often the only person in the room advocating for customers like the market woman. Not because my colleagues are bad people — most of them are not — but because most of them have never been in the shoes of a customer or blue-collar worker in this particular scenario. They’ve never had to calculate, in real time, whether a fifty-cent unexplained charge means they can still buy onions for dinner. The empathy they bring to the customer is intellectual. Mine is structural. This makes me valuable. It also makes me vulnerable.
Valuable because I see what gets dismissed - the small, the informal, the inconvenient - and I argue for it before the data forces the issue. Vulnerable because the same training that taught me to see the market woman also taught me to be measured when I want to be loud. To present the business case calmly. To make the ROI argument again and again. To be the kind of advocate the room will actually hear.
The product changed. Early cancellations dropped. The company made millions in premium revenue from a segment they had previously dismissed. The boardroom moved on, But the framework that produced the original design, the assumption that the salaried English-speaker was the default customer and everyone else was a special case, that framework is still alive in every product I get asked to consult on. I can win the 50-cent fight a hundred times. Until someone changes the framework, the next 50 cents is already being designed somewhere by someone who has never met the market woman and never will.
That’s the part of the job I’m still figuring out.
________________
If you’re working on inclusive design in financial services, frontier-market product strategy, or the design of products for customers who’ve been written off as “too small” to prioritize — I’d love to talk. Get in touch.
