How to Run a Portfolio ReviewStart here

A structured pass over what you hold, in an order designed so the boring findings surface before the interesting ones. Most of what a review finds is administrative, and that is where the money is.

The order, and why it is this order

• Facts before opinions. Establish what you actually hold before deciding whether you should hold it. These are separate exercises and mixing them produces bad versions of both.
• Costs before performance. Costs are known and permanent; performance over any short window is mostly noise.
• Structure before selection. The allocation matters far more than which fund implements it.
• Boring findings first. Duplicated holdings, a forgotten expensive share class, an unhedged exposure nobody chose. This is where reviews reliably pay for themselves.

This page describes a process, not a set of targets. It contains no view on what anyone should hold.

Step 1: List what you actually own

• Every account, every wrapper, in one place. A portfolio spread across four platforms is one portfolio, and it has never been looked at as one.
• Look through the funds. Two funds with different names can hold the same forty companies. Concentration is a property of the underlying holdings, not of the fund names.
• Record the ISIN, not the name — share classes differ in ways names do not show, per the share-class comparison.
• Note what each holding is for. If a position has no stated job, that is the finding.

Step 2: Exposures, not labels

• Run the resulting weights through the multi-asset portfolio calculator: it returns the volatility, the diversification benefit, and which sleeve carries most of the risk. The last of those is usually the surprise.
• Then stress it. The scenario tool asks what an equity fall plus a rate rise plus a spread widening does simultaneously — which is how those three tend to arrive.

Step 3: Total cost, computed rather than quoted

• Add every layer: platform, ongoing charges, transaction costs, spreads, FX conversion, and anything taken from cash balances. The chain analysis lists them all.
• Compound it over your actual horizon with the cost-of-ownership calculator. The annual number is small and the compounded one is not.
• Check for a cheaper share class of the same fund. This is the most common finding in any first review and requires no view on anything.
• Check tracking difference, not just the ongoing charge, for index funds — the decomposition shows the part the fee sheet does not.

Step 4: The questions for each holding

• What job does this do that nothing else in the portfolio does?
• Would I buy it today at today's price, knowing what I now know? If not, the reason for holding is the purchase price, which is not a reason.
• What would make me sell it? If there is no answer, there is no exit plan.
• What am I paying, and to whom?
• What happens to it if the provider fails — segregated, cleared, or unsecured? See what the wrapper changes.
• Could I exit it in a week in a market that does not want to buy?

Step 5: Drift and the rebalancing question

• Compare current weights against intended ones. Drift is automatic and always in the direction of whatever has done best, which is also whatever is now most expensive.
• Use a band rather than a date. The rebalancing calculator shows the trade a given band implies, and whether one is required at all today.
• Count the cost of rebalancing against the risk of not doing it. Frequent rebalancing is a cost; never rebalancing is a drift into concentration.
• Tax consequences can dominate the decision entirely, and are jurisdiction-specific.

Step 6: Write down what you found and what you did not do

• Record the decisions and the reasons, including the decision to change nothing. A review with no written output is a review that will be repeated from scratch.
• Record what would change your mind for each holding you kept. This is the only reliable defence against reasoning backwards from the price later.
• Set the next review date, and make it infrequent. Reviewing more often produces more trading, and trading has a known cost and an unknown benefit — see behavioural finance.