The plan did not stop changing when we published it.

Smaller pots. More time?

Over the last couple of weeks, the question has changed.

For months, we’ve been asking how much money we might have when we retire. We’ve tracked down pensions, compared savings options and spent far too much time in the spreadsheet.

More recently, Karen and I have started asking something else.

If we stopped working earlier, would we prefer it?

Earlier retirement means fewer years of contributions, less time for investments to grow and more years for the money to cover. We’d expect to start with smaller pots.

But we’d also have more time together while we hope to be healthy enough to enjoy it.

We can’t guarantee those extra years will be healthy. That uncertainty is part of why the question matters.

What would the extra money actually change?

We’ve been talking about what we want an ordinary retirement month to look like. The bills, the activities we already enjoy, time with family, weekends away and bigger trips.

Then we’ve asked what working longer would add.

Would the extra money let us do things we genuinely want to do? Would it mostly give us more reassurance? How much reassurance would make another year at work worthwhile?

There’s no clever formula for that bit.

A bigger retirement pot has value. So does a year we can spend together.

We’re trying to work out the balance we would actually choose.

The date is still optional

The protection we’ve always kept in the plan hasn’t changed.

A retirement date in a spreadsheet is an option. We don’t have to hand in our notice because a cell says it’s time.

If markets haven’t played nicely when we get there, we can reassess. Working longer remains one of our options, provided our health and circumstances allow it.

That wouldn’t make a difficult market harmless. It gives us a decision to make before we start drawing on the pots.

A different idea of progress

We haven’t settled on a final date.

But we’re now willing to explore deliberately retiring with less money in exchange for more time.

That feels like a significant change. Progress used to mean making the projected balances bigger. Now it can mean finding an earlier stopping point that we would prefer—and understanding what we’d be giving up to take it.

The question is becoming: how much is enough for the retirement we actually want?

October 2026: Bigger pension pots, better decisions and our first real progress

A few more awkward questions sent us back to the spreadsheet. This time, we also started putting the plan into practice.

Is a bigger pension actually better?

A conversation at work sent me straight back to the spreadsheets.

I'd been challenged on whether putting more money into my pension through salary sacrifice would be better than our current plan of building up Stocks & Shares ISAs after paying off the mortgage.

It's a fair challenge. Pension contributions can be extremely tax-efficient, particularly when salary sacrifice is available.

So we started testing again.

We compared putting the same amount of take-home money into different savings arrangements and explored what that might mean for our retirement finances.

The alternative scenarios could produce significantly bigger pension pots. We even explored what buying an annuity with around £810,000 might look like.

But we kept coming back to the same question.

What's the point of having more money if it's in the wrong place, or arrives at the wrong time?

Our retirement plans aren't about becoming as wealthy as possible at 85. They're about having enough money to enjoy the earlier years of retirement, while still being financially secure when we're older.

Buying a large annuity could provide welcome guaranteed income. But it could also leave us with significantly less accessible capital, particularly in the early years when we hope to do most of our travelling.

And later, when State Pensions and other guaranteed income begin, we might find ourselves receiving considerably more income than we actually need.

That's not necessarily a bad position to be in, but it isn't quite what we're trying to achieve.

Our decision? Stick with our existing strategy for now.

Not because it's guaranteed to produce the biggest overall return, but because it appears better suited to how we actually want to retire.

What happens if the markets don’t cooperate?

Of course, there's a fairly obvious problem with relying on Stocks & Shares ISAs to help fund retirement.

The stock market doesn't care about our spreadsheet.

We've spent more time examining alternative investment returns, early market falls and what happens when the sequence of returns works against us.

Some of the results have been uncomfortable.

One particularly difficult scenario brought our projected fund exhaustion date forward to approximately 2060/61. That was a stress-test result, not our central forecast.

It's a useful reminder that averaging investment returns over a retirement lasting several decades can conceal substantial risk.

A poor run of returns early in retirement can be especially damaging when you're also withdrawing money to live on.

So we've continued exploring how much flexibility we need, how to manage withdrawals, and whether we might want to reduce our exposure to investment risk later in life.

One possibility is to enjoy the investment flexibility through our sixties and revisit guaranteed income when we're around 70.

Another is to gradually move some savings into less volatile assets as retirement progresses.

Neither is a final decision. They're options we want available to us.

The objective isn't to predict the future. It's to have a plan that can adapt when reality refuses to follow the forecast.

Our first small financial milestone

This month we achieved something that probably sounds rather unimpressive.

We had £230 left over.

We'd set ourselves a household discretionary spending budget of around £2,000 per month, separate from our regular bills, savings and other commitments.

Rather than letting any leftover money accumulate in the current account, we've decided to sweep it into our mortgage offset account each payday.

The idea is simple: if we managed perfectly well without spending that money last month, we don't need to give ourselves permission to spend it next month.

Our first sweep was £230.

It's hardly going to transform our retirement by itself, but every extra pound in the offset reduces the mortgage balance on which we pay interest.

More importantly, it's the first evidence that our day-to-day spending plan is achievable.

We've spent months building financial forecasts. It's satisfying to see even a small part of the strategy delivering results in real life.

Finally putting some money into the market

We've also taken another step and decided to start investing into a Stocks & Shares ISA.

We're starting with £500, followed by £200 per month for our initial trial period.

And we've chosen an adventurous investment approach.

This is deliberately a small experiment rather than a wholesale change to our retirement arrangements.

We want to learn how the platform works, understand the charges, see how investments behave, and find out how comfortable we really are watching our own money fluctuate.

There's a difference between putting a 5% or 6% growth assumption into Excel and seeing an actual investment fall by 10%.

It's easy to be adventurous with imaginary money!

We're not expecting every month to show a profit, and we're certainly not judging the success of long-term investing on a few months of returns.

But it feels like another significant moment.

After months of modelling how ISAs might help fund our retirement, we've finally started learning what investing in one actually feels like.

Making the spreadsheet follow real life

Another small but important development has been improving how we manage our retirement model.

Until recently, much of the spreadsheet was based on starting balances, assumptions and projections stretching decades into the future.

That's useful for planning, but reality changes every month.

Pension values rise and fall. We might save a little more than expected. An unexpected expense might mean saving less.

And sometimes we have an extra £230 to put into the mortgage.

We've been improving the model so that we can update actual balances and have future projections continue from those points, without rewriting our financial history.

We've also made the monthly calculations easier to review through annual summaries.

The aim is to make the spreadsheet a living financial plan rather than something we build once and then forget.

Because if we're going to make decisions using it, it needs to reflect what's actually happening.

The plan changed again. This time, the annuity lost.

I knew I wasn’t done. That was probably the problem.

The retirement spreadsheet has changed quite a lot over the last few days.

We tested earlier retirement dates. We modelled Karen adding a small SIPP. We fixed pension contribution assumptions. We looked at different annuity sizes, different annuity dates, tax-efficient withdrawal strategies, ISA withdrawals, pension withdrawals, and what happens if markets behave badly at exactly the wrong time.

For a while, an annuity looked like the obvious answer.

Guaranteed income. Less exposure to market volatility. Less pressure on the investment pots.

Then we modelled not buying one.

And annoyingly, that started to look better.

Keeping the pension invested gives us more flexibility. It preserves more of the ISA. It avoids handing over a large chunk of capital in exchange for an income stream we may not actually need once our State Pensions and defined-benefit pensions start arriving.

There are still risks.

For about eight years, our planned withdrawals are higher than the assumed investment growth. A bad market sequence during that period could hurt.

But our biggest protection against that risk is also the simplest one:

We’re still working.

If markets have fallen badly when we reach the point where we hoped to retire, we don’t have to pretend the spreadsheet is a contract.

We can work another year.

That realisation has changed how I think about the whole plan.

So, for now, the preferred version is:

No annuity. Build the pension and ISAs. Retire when the numbers and the real world both say we can. Then reassess.

The difficult bit now?

Leaving the bloody spreadsheet alone.

I’ve told myself I’m parking it for six months unless something genuinely changes.

I give myself about three days.

We tried turning our retirement spreadsheet into an app. It failed.

This started because I changed my mind about one line on the site: “please don’t ask for the spreadsheet.”

The spreadsheet had become useful to us because we understood every assumption, compromise and oddity inside it. So the obvious question was: could we make something cleaner that other people could use without needing our spreadsheet?

At first, it looked promising. We separated inputs from assumptions, built clearer output pages and started turning the logic into something more general. Then testing did what testing is supposed to do.

Every time we made it more general, another real-world complication arrived: pension crystallisation, annuities, survivor scenarios, ownership, tax, timing, withdrawals, strange inputs and life events. The problem had quietly changed from “make our spreadsheet easier to share” into “build retirement-planning software”.

That was the point to stop. Not because the idea was worthless, and not because nothing worked. Quite a lot of it worked. But the effort required to make it trustworthy for somebody else was no longer proportionate to what we were trying to achieve.

Back to why we built the spreadsheet →

HMRC found 24 employers. That helped us find 9 pension pots.

The Subject Access Request eventually came back, and the employment history was more useful than I expected.

The response listed 24 employers going back to 1991. That gave us another way to rebuild the employment trail and check the places where a pension might have been left behind.

Following those trails helped us identify 9 pension pots of various sizes. Some were easy to remember. Others were exactly the sort of thing that can disappear into decades of job changes, provider changes, old addresses and paperwork.

It reinforced one of the earliest lessons from this whole exercise: before worrying about optimisation, drawdown or tax strategy, first work out what you actually own.

Read how we searched for the old pensions →