Almost everything I write is about what it costs to do nothing. Money sitting still. Ten years not started. Cover that paid out to the wrong place. There is a reason for that, which is that the cost of inaction is the most expensive thing in personal finance and almost nobody measures it.
But it makes for a one-sided library, and it leaves out the part that actually matters. So this one is the other side.
It is one real client, with his permission and his name removed. Six years, three pots of money, and a year where everything got difficult. The figures come off his Ninety One statements dated 15 September 2026, not off a projection.
Where he started
October 2020. Seven months after the March crash, when the recovery was not remotely obvious and most people were still waiting to see what happened next.
He opened two things five weeks apart. A discretionary investment portfolio, starting at zero. And a retirement annuity, into which he transferred R88,000 from an older fund he had stopped thinking about.
Then he set up debit orders and left them running.
Three pots, three jobs
This is the part I want you to notice, because it is the whole reason the rest of the story works.
The money was never one pile. It was split by what it was for, and each piece was invested according to when it would be needed.
The education money went into a growth mandate. His children are years away from university, so that money has time, and time is what lets you carry equity risk.
The retirement money went into a managed mandate inside the retirement annuity. Longer horizon still, plus the tax deduction on the way in.
The short-term money went into an income mandate. Nearly all cash and bonds, almost no equity. It was not there to grow. It was there to be available.
Three pots, three time horizons, three different levels of risk. If you have ever wondered what a financial plan actually is, underneath the paperwork, it is mostly this.
Six years of arithmetic
Between October 2020 and September 2026 he contributed R412,000 across the three, most of it by monthly debit order. Today he is putting in R5,500 a month.
Over that period the money earned R158,000, after every fee. His internal rate of return since inception is 9.76% a year.
For context on the same contributions over the same period, inflation would have given him 4.66% a year and cash in a money market would have given him 6.72%. So he is roughly five percentage points a year ahead of inflation and three ahead of cash.
I will come back to what he did not beat.
Then the year went wrong
In 2026 his household income dropped sharply and stayed down for most of the year. The reason is private, and it is not mine to publish. What matters is the shape of it, and the shape is one most families will recognise: money going out at the same rate, considerably less coming in, and no clear date for when that ends.
This is the moment every plan is actually built for, and it is the moment most plans get abandoned.
Here is what he did.
He reduced his contributions. He did not stop them. The debit orders came down to a level he could sustain, and they kept running every month through the worst of it.
He drew on the short-term pot first. That is what it was there for. He had put R63,000 into it over the years. He took R60,000 back out.
Then he took what else he needed, and the rest came from the education money, which is the uncomfortable part and I am not going to dress it up. Some of what his children's fund had built went to keeping the household steady. He will have to rebuild it.
In total he withdrew R176,000.
The number that makes this worth writing
He withdrew R176,000.
The money had earned R158,000.
Almost every rand his family needed in the hardest year they have had, the plan had already produced. He did not eat into what he had put away so much as spend what the putting away had made.
And after taking out R176,000, he is still sitting on R395,000.
The pot that ran dry is the one that worked
That short-term pot has R8,000 left in it, down from R63,000.
On a statement that looks like failure. It is the opposite. That money was parked in a low-risk mandate precisely so it would be there, whole and available, on a month nobody saw coming. It was not supposed to grow into anything. It was supposed to be reachable.
It was reachable. That is the entire job description, and it did it.
An emergency fund that never gets used has not proven anything. One that gets used and holds up has.
What this is not
Three honest caveats, because an article that only shows the good half is just advertising.
He did not beat either index. Over the same period the FTSE/JSE All Share returned around 16% a year and MSCI World about 15.7% in rand, against his 9.76%. Both of those are real and both deserve saying out loud.
But neither of those is what his money was measured against, and neither is what it was built to be.
His retirement money sits in a portfolio that targets inflation plus 5% and is limited to 75% equity by regulation. His education money sits in one targeting inflation plus 6%. Both hold cash, bonds and listed property alongside shares, on purpose. Measured against the categories they actually compete in, both beat their peer group over five years, and the short-term pot beat its own benchmark over every period on its fact sheet.
The reason portfolios like that lag a pure equity index in a good decade is the same reason they hold together in a bad year. In 2022, when almost everything fell, his retirement mandate finished the year up 1.14%.
It is also worth knowing that nobody actually earns an index return. The cheapest way to own MSCI World in South Africa returned 15.55% over five years against the index's 15.74%, and that is before a cent of advice, platform or administration cost. An index is a measuring stick. It is not something you can buy.
This is one household. One client is a story, not evidence. It does not prove a strategy and it does not predict yours.
Past returns say nothing about future ones. The next six years will not look like the last six. They never do.
What actually made the difference
Not the fund selection. Not the market. Not timing, since he started seven months after a crash and could not have known what came next.
Three things, and none of them are clever.
He gave every rand a job before he invested it. So when the year turned, he was not deciding what to sell under pressure. He already knew which pot to open and in what order.
He kept going at a smaller number instead of stopping at zero. A reduced debit order keeps the habit, keeps the compounding running on everything already invested, and means there is no decision to make again later about restarting. Stopping is not a pause. Stopping is usually the end.
He started before he was ready. October 2020 was not a comfortable month to begin investing. He began anyway, with an amount that felt small.
That is the whole method. It is not sophisticated and it was never supposed to be.
The part I keep thinking about
If he had not started in October 2020, he would have arrived at 2026 with the same drop in income, the same bills, and R176,000 less to meet them with.
He would have had to find that money somewhere. Credit, most likely, at 15% or more, which he would still be paying off now and for years afterwards. That is the version of this story where nothing dramatic happens and the damage takes a decade to undo.
The debit order he set up in 2020 was not really an investment decision. It turned out to be the thing that stopped 2026 from becoming 2031.
He could not have known that. Nobody can. That is precisely why you start before you know.
