Sign up
Book a Demo

The FAIS Referral Fee Judgment Might Be the Best Thing to Happen to South African Lead Generation

Derek Notman

August 27, 2026

Originally published August 27, 2026

By Derek Notman, CFP®, Founder and CEO, Couplr · 24 August 2026

A Johannesburg High Court judgment in April made a percentage-of-premium referral agreement unenforceable under FAIS, and the reaction across South African distribution has been alarm. I think that reaction is wrong. Not because the commercial disruption is not real, it is, but because the arrangement the court declined to enforce was one our industry should have been moving away from anyway. My view is straightforward: a company that generates leads should be paid for the lead, not for the product that gets sold afterwards. The moment a referrer’s income depends on a policy being written, that referrer has an interest in the policy being written, whether or not it suits the person buying it. Paying a flat fee for an introduction is a less conflicted arrangement than paying a share of revenue. That is true before you get anywhere near FAIS, and it stays true regardless of how this judgment is eventually treated.

What did the court actually decide?

Briefly, because the facts matter and I do not want to overstate them.

In The Raspberry Academy (Pty) Ltd v Oaksure Financial Services (Pty) Ltd, handed down on 14 April 2026, Raspberry sued Oaksure for R1,214,698.62 in unpaid referral commissions under a 2019 Lead Referral Agreement. Raspberry referred potential customers to Oaksure for short-term insurance. It was paid a percentage of net premium, due only once premiums were received, and payable for as long as the client kept the policy. Rates ran from 10% on passenger liability cover to 50% of statutory commissions on other lines.

Oaksure did not defend the claim. Acting Judge DJ Smit raised the regulatory question himself. He looked past the contract’s description of Raspberry as a marketing agent, examined what the arrangement actually did, and held that Raspberry had performed intermediary services under the Financial Advisory and Intermediary Services Act 37 of 2002 without being an authorised provider or an appointed representative. The agreement was illegal and unenforceable. Raspberry recovered nothing.

Writing in Moonstone on 18 May, Mark Bechard drew the practical line as clearly as anyone has: flat-fee, purely mechanical contact-passing arrangements might escape regulation, while commission-based referrals tied to policy outcomes trigger intermediary services requirements. Ayanda Nondwana and Jade Werner of Eversheds Sutherland made the same point from the other side on 14 May: a firm cannot contract its way out of FAIS by choosing a different label.

I should say plainly that this is not settled law. Cliffe Dekker Hofmeyr, writing on 5 August, question the reasoning and point to tension with the Supreme Court of Appeal’s decision in Tristar Investments, which cautioned against stretching the intermediary definition to cover activities only indirectly connected to a transaction. It was also an undefended matter decided by a single judge in a Local Division. Anyone telling you the question is closed is ahead of the evidence.

But the commercial signal does not depend on how the appellate question resolves. Firms are looking hard at their referral arrangements right now. That is the part worth talking about.

Why do I think this is good news?

Because the thing the court pulled on is a thread the industry has been avoiding.

Set FAIS aside for a moment and just look at the economics. If I pay you a percentage of premium every time one of your introductions turns into a policy, I have told you exactly what I want. I want policies. I have not asked you to care whether the person on the other end of your introduction was well suited to that policy, or to that adviser, or whether they needed it at all. I have made your income a function of transactions, and then I have hoped you will optimise for something else.

People respond to what they are paid for. That is not cynicism, it is just how incentives work, and it applies to every industry, not only ours.

I want to be careful here, because I am not accusing anyone of anything. A great many South African firms built these arrangements in good faith, at a time when they were the market norm, and are running them today with genuine care for the client outcome. The people are not the problem. The structure is. And a structure that requires everyone inside it to act against their own financial interest in order to produce a good result is a badly designed structure, however good the people are.

So when a court says that particular structure will not be enforced, my honest reaction is that the industry has been handed a reason to fix something it already knew was uncomfortable.

What does a fee structure actually do to behaviour?

This is the part I care most about, and it is where I think the conversation has been too legal and not human enough.

Pay a lead generator on the product sold, and the rational move is volume plus pressure. Send more introductions, send them faster, and push the ones that look like they will convert. Fit is somebody else’s problem, and usually it becomes the adviser’s problem, and eventually the client’s.

Pay a lead generator a flat fee for an introduction, and the rational move changes completely. Now the only thing worth optimising is whether the introduction is any good, because a bad introduction costs the buyer money and earns the generator nothing extra. Suddenly the generator and the adviser want the same thing.

Two variables do almost all of the work here: what triggers the payment, and who makes first contact.

Arrangement What triggers payment Who makes first contact Position relative to the judgment
Percentage of premium Policy concluded, premiums received The referrer The arrangement the court declined to enforce
Share of commission Policy concluded The referrer Materially the same analysis
Conversion-based fee Sale completed The referrer Named by Cliffe Dekker Hofmeyr as exposed
Cost per lead Contact details delivered The referrer Weaker causal link, but nobody asked to be contacted
Flat fee, consumer-initiated The consumer chooses to start a conversation The consumer Furthest from the facts of the judgment

This is the distinction I keep coming back to in my own work: lead generation and lead conversion are not the same business, and paying for one while measuring the other is how firms end up disappointed in both. I have written elsewhere about the alternatives to pay-per-lead models and why the pricing question is really a design question.

What do the acquisition economics actually say?

There is a set of numbers I find clarifying, and they have nothing to do with regulation.

Kitces Research surveyed roughly 1,000 advisers for its 2022 study on how financial planners actually market their services. The average all-in cost to acquire a single client was $3,119, and only $519 of that was hard-dollar spend. The other $2,600, about 83%, was the adviser’s own time. Client referrals cost $338. Social media cost $11,937.

Cost to acquire one client, by channel

Social media · $11,937

Average, all channels · $3,119

Client referral · $338

Source: Kitces Research, How Financial Planners Actually Market Their Services, Vol. 1 (2022), n = approximately 1,000 advisers. Figures in US dollars. Of the $3,119 average, $2,600 (83%) is the adviser’s own time rather than hard-dollar spend.

Read that again. A referral is not cheap because it is free. It is cheap because fit and intent are already established before anyone spends an hour. Every rand a firm saves on a cheaper lead is dwarfed by the hours its advisers burn discovering that the lead was never a fit. The industry has spent a decade negotiating the small number and ignoring the large one.

Does this mean commission is the problem?

No, and I want to be unambiguous about that, because it would be easy to read this piece as an argument against commission generally. It is not.

An adviser earning commission on a product they have recommended, to a client they know, under a suitability and conduct framework, is a completely different arrangement from an unlicensed third party being paid out of the proceeds of a transaction it introduced but had no responsibility for. The first is a long-settled question with a regulatory regime around it. The second is what this judgment was about.

Commission-based advisers are not the issue here. Some of the best advisers I know are commission-based, and some of the worst client outcomes I have seen came from fee-based arrangements. How someone is paid tells you about the incentives they face, not about their character or their competence.

What the judgment questioned is narrower and, I think, fairer: whether someone who does none of the advice work, carries none of the licensing obligations, and bears none of the ongoing responsibility should nonetheless be paid a slice of the product revenue for as long as the client keeps paying.

What changes when the consumer opts in first?

Everything downstream of it, in my experience.

Most lead arrangements begin with a consumer who did not ask to be contacted. Their details move first, and the conversation is initiated by someone with a commercial reason to initiate it. Every problem the industry then complains about, low conversion, high complaints, short relationships, poor persistency, follows from that opening move.

Flip it. Let the consumer see the advisers who suit them, and let nothing happen until the consumer decides to reach out. Now the adviser is not prospecting, they are responding to somebody who has already chosen them. The introduction is warm because the consumer made it warm.

That design also does something useful for conduct risk, and it does it structurally rather than through a policy document. If no personally identifiable information reaches an adviser until the consumer has acted, then nobody is being marketed to without asking. That is not a clever compliance workaround. It is just what happens when you let the person with the least power in the transaction decide when it starts. We have written about how consumer-initiated engagement changes the compliance picture, and about what compliant advisor-client matching looks like before trust exists.

Combine the two ideas and you get the arrangement I would argue for even if this judgment had never been handed down: a flat fee, paid for an introduction, where the consumer opted in. The generator is paid for the thing they actually did. Nobody’s income depends on a product being placed. And the person the whole system exists to serve is the one who chose to start the conversation.

What did this look like when I was the adviser on the other end?

I have South African roots, and I have spent a lot of time with advisers there, but I want to be honest that my own years in practice were in the United States. The mechanics travel better than most people expect.

I came into this business through an agency distribution system at twenty-five. I was taught to sell before I was taught to advise, and I was handed lists. I once drove three hours each way to write a ten dollar a month term policy, because the activity was the metric and nobody was measuring whether that was a sensible use of a day.

What I remember most is not the driving. It is the number of conversations that were dead before I opened my mouth, because the person across from me had never asked to meet me. I was not bad at my job. I was operating inside a structure that generated introductions nobody had consented to, and then measured me on how many of them I could convert.

Roughly seven in ten new advisers leave the business within five years, according to Cerulli’s January 2024 research. We call that a talent problem. Having lived it, I think a meaningful share of it is a lead quality problem wearing a talent problem’s clothes. You cannot hire your way out of a distribution model that hands new people conversations that were never going to work.

What should a South African firm take from this?

Three things, and only the first is urgent.

  1. Review the arrangements, and start with the payment trigger. Inventory every deal where an outside party earns money in connection with new business. For each one, write down what has to happen before the money is owed. If the answer is a concluded policy, premiums received, or a share of commission, that arrangement belongs at the top of the pile. Then check whether the counterparty is an authorised FSP or a properly appointed representative, because that changes the analysis entirely. Take advice on the specific arrangements rather than the general question, because the facts differ from deal to deal.
  2. Treat this as a design question, not a paperwork question. The temptation will be to redraft contracts until they look defensible. That may be necessary, but it is not sufficient, and it leaves the underlying incentive exactly where it was. If an arrangement has to be restructured anyway, restructure it into something that also produces better conversations.
  3. Watch what happens to lead quality. This is the part I would genuinely like the South African market to measure, because I do not think anyone has good public data on it. If firms move from revenue-share to flat-fee introductions, does conversion improve? Does persistency? My strong expectation is yes, and I would rather be shown evidence than be right by assertion.

Where does this leave us?

South Africa may end up somewhere the rest of the world has not reached yet, and that is worth saying out loud.

That is not a new thought for me. On the Rethink. Financial Advice podcast, Bekithemba Mafulela of Momentum pushed back on the assumption that the United States leads the world in how advice gets delivered. His argument was that South Africa’s market structure, a large population with income but few assets, forces a harder and in some ways more advanced answer to the question of who can be served profitably. That conversation changed how I think about which markets are actually ahead.

The judgment did not hold that third-party distribution is unlawful. It held that an unlicensed party cannot be paid out of the proceeds of a regulated transaction it helped bring about. Those are very different propositions, and the space between them is where the better models live.

If this pushes the market toward paying for introductions rather than for products, then consumers get contacted less and matched better, advisers spend their hours on people who actually want to talk to them, and firms stop paying for volume they cannot convert. I do not think that is a compliance burden. I think it is a correction that was coming anyway, and South Africa may simply get there first.

For what it is worth, this is how we have built Couplr, and we built it this way before this judgment existed. Consumers complete a short set of questions, see the advisers we think they would match well with, and nothing happens until they choose to start a conversation. Firms pay a flat amount. It is not a percentage of premium, not a share of commission, and not linked to any product sold or revenue earned. Where we measure success, we measure it on conversations a consumer chose to start and meetings that got scheduled, never on products placed.

I am not going to tell you that makes any particular arrangement FAIS compliant. That is a legal conclusion, it turns on facts specific to your firm, and it is not mine to certify. What I will say is that we did not arrive at this structure to solve a legal problem. We arrived at it because paying for introductions rather than outcomes is the only version of this business that makes the incentives point the same direction for everyone in it.

If you want the wider picture on this market, our South African advisor-client matching page sets out how the model fits an enterprise stack, and we have published research on what the South African wealth consumer actually looks like. For insurance distribution specifically, see advisor matching software for insurance distributors.

In the spirit of Ubuntu,

Derek Notman

Rethinking how your firm generates new business in South Africa?

See how consumer-initiated matching works alongside what you have already built. A flat fee for the introduction, never a share of premium or commission, and no adviser receives a person’s details until that person chooses to reach out. Bring your compliance team to the call, we would rather have the conversation with them in the room.

Book a demo

This is an opinion piece about a South African court judgment and what it means commercially. It is not legal advice, and Couplr is not a law firm. Any firm reviewing its own arrangements should take advice from its own legal and compliance advisers.

Sources

Scroll to Top
Financial Advisor Directory