Every portfolio rests on a set of assumptions. Some were chosen deliberately. Others arrived quietly, carried in with a fund selection or a rule of thumb picked up years ago and never revisited.
Assumptions age. A portfolio built on sensible reasoning in 2019 may still hold the same positions while the reasoning underneath has shifted, because markets change composition faster than most people update their thinking. Checking those foundations periodically is ordinary maintenance, and it’s one of the higher-return uses of an investor’s time.
Turning definitions into a checklist
Basic material reads very differently with a live portfolio open beside it. The first time through, terms like diversification, correlation and market weighting are definitions to memorise. The second time, they’re tests to apply.
Working back through investing 101 material with real holdings in view converts familiar concepts into a practical audit. Each definition becomes a question. Does this portfolio actually meet the definition of diversified? Are the correlations the allocation assumed still holding? Is position sizing consistent with the stated risk tolerance, or has it drifted with performance?
Those questions take an afternoon. The answers tend to be more useful than any new strategy.
Assumption one: the index is diversified
This is the assumption most worth testing, because it’s the one that has changed the most.
Broad index exposure is often treated as a complete diversification solution. The arithmetic has moved. The ten largest companies in the S&P 500 now account for close to 40% of the index, a level that has roughly doubled over the past decade.
None of this makes index investing a poor choice. It does mean the diversification an investor believes they hold and the diversification they actually hold may be two different things, and that gap is worth measuring rather than assuming.
Assumption two: holdings don’t overlap
Investors who deliberately spread across several funds sometimes end up with more overlap than intended. Growth funds, value funds and thematic funds can all hold the same large companies, arriving at similar exposure through different labels.
The check is simple. Pull the top ten holdings of every fund in the portfolio, list them together, and count how often the same names repeat. Most platforms publish this data, and the exercise takes minutes.
It’s also worth knowing which direction concentration is moving. Top-ten weighting has eased slightly from its 2025 peak of around 41%, though it remains well above the levels of the previous two decades. A portfolio built when the figure sat near 20% was solving a different problem.
Assumption three: risk tolerance is fixed
Risk tolerance is usually set once, often through a questionnaire, then treated as a permanent setting. In practice it moves with circumstances. A stable income, a mortgage, a change in time horizon or simply a few years of experience all shift what an investor can genuinely sit through.
A useful test asks about behaviour rather than preference. What did this portfolio’s owner actually do during the last significant drawdown? That answer describes real tolerance more accurately than any self-rating.
Assumption four: the plan is still being followed
Portfolios drift. Winners grow into oversized positions, contributions land in whatever looked appealing that month, and the allocation on paper gradually separates from the allocation in the account.
Worth confirming at each review:
- Current weights against target weights, position by position
- Rebalancing frequency, and whether the last one actually happened
- Contribution destinations over the past year
- Cash levels, including cash that accumulated by accident rather than by decision
Assumption five: the costs are known
Fees are among the easiest things to audit and among the most commonly estimated rather than checked. Fund charges, platform fees, currency conversion on foreign holdings and transaction costs all sit in different places, and few investors have added them into a single annual figure.
Doing so is straightforward and produces a number that compounds against returns every year without exception. That makes it one of the few portfolio inputs an investor controls directly.
Running the audit
A workable rhythm looks like this:
- Annually: full review of allocation, overlap, costs and stated goals
- After major life changes: revisit time horizon and risk tolerance
- After significant market moves: check whether weights have drifted past tolerance bands
- Continuously: note the reasoning behind each new position, so future audits have something to check against
That last habit does most of the work over time. A portfolio where every holding has a written reason is straightforward to audit. One where the reasons live in memory usually isn’t.
What the audit is really measuring
The point isn’t to find errors. Most audits confirm that a portfolio is broadly doing what its owner intended, which is a useful result in itself.
What the process reliably surfaces is drift between intention and reality, and drift is quiet by nature. It accumulates through reasonable individual decisions that were never reviewed as a set. Whether an annual check is frequent enough depends on how much the portfolio changes between reviews, which is something an investor only learns by starting to measure it.
