Building and maintaining a portfolio

Allocation first, funds second, and a rule for what to do when it drifts.

Most people assemble a portfolio backwards — starting with individual funds that looked appealing, and ending up with a collection rather than a plan. The order that works runs the other way.

The order

1. What's the money for, and when do you need it? Separate pots for separate purposes. A house deposit in three years and a retirement in twenty-five are different problems and shouldn't be blended into one portfolio.

2. What wrapper? ISA, pension or general account, based on access and tax — see Investment wrappers.

3. What asset allocation? The split between shares, bonds and cash. This decision drives the large majority of how your portfolio behaves.

4. What holdings? Only now do individual funds come into it. This is the step people start with, and it matters least.

5. What's the maintenance rule? Decided in advance, while you're calm.

The simplest good answer

For a great many people, a single multi-asset fund at an appropriate risk level does the whole job.

One holding, diversified across asset classes and regions, rebalanced automatically by the manager, at a risk level you choose from a range. No allocation decisions, no rebalancing admin, no temptation to tinker.

It's less interesting than building your own, and for most investors it produces a better outcome, because the main risks — poor diversification, drift, and fiddling — are engineered out.

If you'd rather build it yourself, the rest of this page applies.

Building it yourself

A straightforward approach uses a small number of broad, cheap building blocks:

Global developed equity — the core
Emerging market equity — a modest slice
Bonds — government and corporate, sized to your risk level
Cash — for anything needed within a few years

Three to five holdings is often plenty. More holdings is not more diversification if they own the same things — check overlap before adding another fund.

Core and satellite is a common structure: a large diversified core in cheap trackers, with smaller deliberate positions around it. It gives you somewhere to express a view without putting the whole portfolio at risk.

Rebalancing

Over time the winners grow and the losers shrink, so a portfolio that started at 60% shares drifts towards 70% or more. Your risk level rises without you deciding anything.

Rebalancing means selling some of what's grown and buying what hasn't, to return to your target. It feels wrong — selling what's doing well to buy what isn't — and that's precisely why it works: it's a mechanical rule that makes you buy low and sell high without needing to predict anything.

Two approaches:

Calendar rebalancing — review annually on a set date. Simple and sufficient.

Threshold rebalancing — rebalance when an allocation drifts more than a set amount, say five percentage points, from target. Slightly more efficient, requires more monitoring.

Both work. Doing neither is the problem.

Rebalance with new money first. Directing new contributions towards whatever is underweight avoids selling anything at all, which is cheaper and avoids realising gains.

Rebalance inside tax wrappers where possible. Selling within an ISA or pension has no tax consequence. Selling in a general account can realise a capital gain, so where a portfolio spans several wrappers, do the adjusting inside the sheltered ones.

Reviewing

An annual review covering:

Has anything changed? New job, new child, an inheritance, a changed retirement date, a shifted goal. Circumstance changes matter far more than market movements.

Is the allocation still right? As a goal gets closer, the sensible risk level usually falls — a portfolio for a deposit needed in two years shouldn't look like one for a retirement in twenty.

What am I paying? See What you actually pay.

Does it still hold what I think it holds? Funds change mandates, merge and close.

Once a year. Resist more.

The common mistakes

Too many holdings. Fifteen funds owning the same global companies is one portfolio with fifteen sets of paperwork.

Never rebalancing, so risk creeps upward for a decade until a downturn reveals it.

Chasing performance, adding whatever did well last year — which systematically buys high.

Mismatched horizons. Money needed soon invested as though it isn't.

No plan for a fall. The rule has to exist before you need it, which is the subject of Time in the market.

What good looks like

A portfolio you can describe in a sentence, that matches when you need the money, that costs you a known and reasonable amount, and that you can leave alone for a year without anxiety.

Not sophisticated. Just deliberate.

This article is for general education only and isn't personal advice or a recommendation of any investment or strategy.