Three Sample Plans, One Question: When Can I Retire?
planbend ships with three finished sample households so you can look at a real plan before building your own. I use them constantly while working on the app, and I noticed something while running all three through the same question: not one of them runs out of money. Zero shortfall years in all three, across sixty-odd years each.
That could have been a boring result. Instead it turned out to be the most useful thing about them, because it means the three plans differ in something other than whether they work — and what they differ in is the part a retirement number can't tell you.
Every figure below is output from the live projection, read on 29 October 2026. The households are invented. The arithmetic isn't, and you can open each one and reproduce it.
How to read these
Two things to hold onto before the numbers.
The dollars are nominal. All three plans assume 2.5% inflation and run to 92, so the end balances are in the money of the 2080s. Dani's $9.9 million at 92 buys roughly what $2.2 million buys today. A large terminal balance is mostly a statement about the length of the projection, which is why I've put almost no weight on it below.
All three draw proportionally. None of them is using a clever withdrawal order. That's deliberate — I wanted to see what the plans look like before anyone optimises them, because that's the state most real plans are actually in.
Dani, 31, single, $73,800 — the thirteen cheap years
A marketing coordinator in Ohio, renting, $24,800 of student debt left, no kids. She defers 6% into a 401(k) and saves about $13,500 a year all in. Nothing heroic.
Renting is the thing that shapes her whole plan, and it is worth saying plainly before the numbers. She has no mortgage to pay off, so nothing about her housing cost ever ends; it just goes up. Her $1,450 a month today is modelled rising at 3% a year for sixty-one years — a little ahead of the plan's 2.5% inflation, which is roughly how US rents have behaved over most long periods — and that single assumption is why her retirement spending below looks nothing like a homeowner's.
Look at those last two lines next to each other. Thirteen years of retirement cost Dani $56,915 in federal income tax — about $4,380 a year — and the eighteen years after them cost $471,464. That is not a tax cliff at 75; it's a required minimum distribution arriving at a 401(k) that has been compounding untouched for forty-four years.
By 75 that account holds $2.72 million and the first RMD is $47,058, against annual spending of $98,482. The distribution does not even cover half of what she spends, so none of it is wasted — but it arrives as ordinary income stacked on everything else, and her federal bill goes from $6,492 the year before to $13,570 the year it starts. Every year after that, more.
Meanwhile, for those thirteen quiet years, her marginal rate never leaves 10% or 12% and the plan shows $848,147 of unused room in the bracket she is sitting in. She could have been moving money out of the 401(k) at the lowest rate she will ever see, and the plan as it ships moves none.
I ran it both ways. Filling that room each year with Roth conversions:
She pays an extra hundred and twenty-eight thousand of tax during the window to take a hundred and forty-three thousand off the lifetime bill, and ends with $436,764 more. The 4% rule has nothing to say about this, because the question isn't how much she can withdraw. It's which account the money is sitting in when the IRS starts insisting.
One more thing in Dani's plan that I only saw because I was looking: her HSA. It's $3,100 today. By the time she retires it's large enough to pay $57,059 of healthcare across the three years between 62 and Medicare — entirely untaxed, because qualified medical withdrawals from an HSA always are.
The Hendricks family, 44 and 42, $163,800 — college is bigger than you think
Marcus and Priya, North Carolina, two children — Nora is 13, Theo is 10. A $232,000 mortgage, a car loan, a credit card balance, a student loan of Priya's, a small pension of hers from 62. Two 529s with $39,500 between them. Both retire at 60.
This is the household that taught me something about my own code this week, so I'll start there. The two 529s look healthy. They are not close to enough.
Thirty-six per cent from the 529s. Of the rest, $206,015 comes out of the taxable brokerage account they were building for retirement, and the last $1,981 out of plain cash in the final year, once both 529s are down to their last few hundred dollars. And the overlap year is the one that hurts: in 2034 Nora is a senior and Theo is a freshman, and the household writes $80,746 of cheques in twelve months.
That matters more than it looks, because this is money leaving the portfolio in their early fifties — the decade when compounding does the most work. The plan still succeeds, but it ends at $9.2 million. Take the children out of the plan altogether — no college, and no 529 contributions either — and the same household ends at $11.3 million. Not all of that $2.1 million is college; some of it is the money that went into the 529s in the first place, which is the point. A 529 shortfall doesn't show up as a crisis. It shows up as a smaller number sixty years later.
(If you want the mechanics of why one 529 can't just pay for both children in an overlap year, I wrote that up separately: 529 plans when multiple kids overlap in college.)
Then the five-year gap
They retire at 60 and Medicare starts at 65. Five years of buying their own health insurance, and their HSA — $19,400 today — covers $101,301 of it, tax-free, which is essentially the whole healthcare line for those five years.
That is the single best use of an HSA and almost nobody plans for it. A dollar in an HSA spent on a medical bill is never taxed: not going in, not growing, not coming out. A dollar in a Traditional 401(k) spent on the same bill is ordinary income. In the pre-Medicare years, when healthcare is both expensive and entirely self-funded, those are very different dollars.
There's also a nice bit of sequencing visible in their tax line. They both "retire at 60", but Priya is two years younger, so she's still working for the first two years of Marcus's retirement. Their tax bill is $12,852 at 61 and $1,392 at 62 — the year her salary stops. Their Roth conversion window doesn't really open until then. The plan converts $25,000 a year from 60 to 64 regardless, which means two of those five years are conversions made at a much higher marginal rate than necessary. Same ladder, wrong five years.
The Okonkwo household, 46 and 45, $381,600 — eleven years to bridge
Ade and Clare, Washington State, one daughter. An engineering director and a product designer, $42,000 of bonus, an RSU grant with 1,300 shares still to vest, and a $940,000 house. Ade retires at 54, Clare at 56. They are the household with the most money and by far the hardest problem.
Eleven years is the whole story. Retiring at 54 means eleven years of self-funded health insurance, eleven years before Social Security, and — for most of it — a 401(k) they can't touch without a penalty. Their $4.4 million is not the constraint. Reachability is.
Their HSA carries $193,717 of that healthcare bill without a cent of tax. There's a detail in there I like: the year their daughter turns 26 and comes off the family plan, the household's healthcare line drops by about a third, from $21,773 to $14,878. That is Ade at 57, in 2037, because Zara was born in 2011 — the plan knows the year because it knows her birthday. A flat healthcare assumption would have missed it in both directions.
The part they can't convert away
This household does what you're supposed to do. It converts $560,000 to Roth across eight years, starting the moment Ade's income stops. And it still isn't enough.
A required distribution of $163,989 against spending of $106,802. Half again as much as they need, as ordinary income, every year from 75. That income then pushes them over the first IRMAA threshold, so their Medicare premiums carry a surcharge for seventeen years — $122,200 of it, which is a cost most people don't know exists until it appears on a statement.
This is the honest version of a high-income plan: the conversion window is real, it's worth taking, and it isn't a cure. Eight years at $70,000 a year moves $560,000 out of a pre-tax balance that is growing faster than that. The lesson isn't "convert more" — converting more means paying tax at a higher rate now. It's that a plan this pre-tax-heavy has a tax tail you can shorten but not remove, and you want to know that at 46 rather than at 75.
What was true in all three
Three households, an income range of more than five to one, and the same three things came up every time.
The binding constraint was never the total. Not once. Dani's problem is which account her money is in. The Hendricks' problem is a seven-year stretch in their fifties. The Okonkwos' problem is eleven years of being asset-rich and access-poor. A target number — 25× spending, or any other multiple — answers a question none of them was actually facing.
The HSA did more work than its balance suggests. It's the smallest account in all three plans and it paid for healthcare in the most expensive, least-covered years, untaxed, in every one of them. Up to 65 it's a medical account — though that's wider than it sounds, because one spouse's HSA can pay the other's bills and the children's, whoever is on which health plan. It can be a bridge account as well, if you tell the plan you're keeping receipts: a medical expense you paid out of pocket years ago can still be reimbursed from the HSA tax-free at any age, and spent on anything. From 65 the non-medical part behaves like a Traditional IRA. That makes it the only account that is better than a Roth for one specific purpose, and that purpose is the exact gap an early retiree has to fund.
Every one of them had a tax window, and two of the three were using it badly. The quiet years between retiring and the first required distribution are the cheapest years of your life, tax-wise. Dani spends thirteen of them in the 10% and 12% brackets and converts nothing. The Hendricks' five are real but start two years too early. Only the Okonkwos are converting hard, and they still don't catch up. The window is the most valuable thing in a retirement plan and the easiest to miss, because nothing goes wrong while you're wasting it.
Go and break them
Each of those links opens the real application against that household's data. Editing is fully enabled and nothing is saved anywhere — drag Ade's retirement age to 58 and watch eleven years of bridge become seven; turn on a Roth ladder in Dani's empty window; move the Hendricks' 529 contributions up and see how much of that $206,015 you can take off the brokerage.
That is genuinely how I find most of the problems in my own plan, and, as it happens, most of the problems in the app. Writing this post is what turned up the fact that the projection had been charging college at tuition only while the 529 card was collecting room and board — which understated the Hendricks family's two children by about $160,000. It's fixed now, and the numbers above are the corrected ones. Sample plans earn their keep.
When you're ready to do it with your own figures, planbend is free to start and doesn't ask for a bank login. If you'd like a professional's read on any of this, there are directories on the Connect with a professional page.