The Concentration Signal: What the FCA’s Wealth Survey Says About where your client lands

Reacting to: FCA warns wealth managers over fees, financial crime and AI risks (Money Marketing, 18 August 2026) →

The FCA published its Wealth Management Survey Report 2026 this morning, and the headline the trade press has taken from it is about fees, financial crime controls and artificial intelligence. Those are the right findings for a wealth manager to read. They are not the most useful finding for an accountant. The most useful finding is buried in the market structure section, and it is this: the ten largest firms, by client numbers, now serve 89% of discretionary management clients — a share the FCA describes as “up 19% since we conducted the first survey in 2022”.

That single number changes what the sentence “I know someone” is actually doing. When an accountant points a client towards regulated advice, the assumption underneath is that a broad, competitive market is waiting on the other side and the client will find their level in it. The regulator's own data says the market is narrowing, not broadening, and that the firms doing the absorbing are the largest ones — whose propositions are built around investable assets. Your client, whose wealth is mostly a trading company, a director's loan account and a pension nobody has looked at since 2019, is not the client that market is designed around. Knowing that does not make the introduction less worth making. It makes the question of who, and of what the accountant contributes to the conversation, considerably more important than it was.

What the FCA actually published, and what it covers

The survey covers the firms in the FCA's wealth management supervision portfolio: on the regulator's own figures, more than 5.5 million retail clients and almost £1 trillion of assets, with around 400 firms responding and 320 of them carrying out portfolio management. It is the third iteration — 234 firms have now completed all three rounds since 2022, which is what makes the trend readings possible.

One caveat matters and is easy to miss in the coverage: the data runs to submissions made up to 31 December 2024, filed by May 2025. The report is new; the picture in it is roughly eighteen months old. Every structural trend it describes has had another year and a half to run since the numbers were collected. That cuts in one direction only, because consolidation in this market has not slowed since 2024.

Ten firms, 89% of the clients — and why that lands on your desk

Concentration in an advice market is not automatically bad for the people in it. Large firms tend to have deeper compliance functions, better systems and more resilience than a two-adviser practice. The FCA's own report notes that all firms now refresh their Know Your Client checks, against 8% failing to do so in 2023/24 — scale helps with that sort of thing.

What concentration changes is fit. The larger the firm, the more its service has to be standardised to be deliverable: a centralised investment proposition, model portfolios, a relationship manager holding several hundred clients, and an onboarding process that asks first what a prospective client has to invest. None of that is designed to interrogate whether a company should be making an employer pension contribution before the year-end, whether the surplus sitting at the bank is working capital or dead money, or whether the shareholder protection nobody arranged is a bigger exposure than the asset allocation. Those questions are business questions, and they are visible in the accounts long before they are visible on a fact-find. We set out the six most common of them in the signals already sitting in your client file.

This is the practical case for the accountant staying in the conversation rather than handing it over and stepping back. Not because the wealth manager is deficient, but because the two professions can see different things. The accountant holds the history, the timing and the corporate structure. The regulated adviser holds the analysis, the recommendation and the responsibility for it. Who does what sets out that split properly.

The fee question your client probably cannot answer

The FCA's fair-value finding is worth reading precisely, because it is being reported loosely. The regulator does not say wealth managers charge too much. It says that pricing “is not always clear, easy to compare or applied consistently across different services and client groups”, and it points at the FCA's own Financial Lives 2024 survey, in which 17% of adults holding £100,000 or more of investable assets who used a named wealth management firm were concerned that fees were high, hidden or complex. The regulator also asks firms to consider whether portfolio turnover is delivering genuine value, rather than activity for its own sake.

For an accountant this is unusually actionable, because clarity of cost is your native territory. You already ask what a lease costs, what the finance charge on the van comes to, what the bank is taking. Very few clients can answer the equivalent question about their investments in pounds. They can usually quote a percentage, and a percentage is precisely the format in which a cost stops feeling like money.

Putting real numbers on it

Take the lang cat's State of Advice Report 2025 figure for the average new advised client portfolio: £411,000. On that amount, every 0.25% of annual charge is £1,028 a year. Over ten years, before any growth at all, that quarter of a percent is £10,280 out of the client's money.

Now the composite. It is illustrative — not a real client, and not a comment on any particular firm's charges.

• The client. A manufacturing company owner, 58, who sold a minority stake three years ago and put the proceeds with a discretionary manager. The accountant sees the portfolio once a year, on the tax return, as a dividend figure and a realised gains schedule.
• The question asked at the year-end meeting. Not “are your investments performing?” — that is not the accountant's question to ask — but “what did that arrangement cost you in pounds last year, all in?”
• The answer. The client says about 1%. The accountant asks for the annual statement, and finds that the 1% is the discretionary management fee only. Platform custody and the underlying fund charges sit in a separate schedule.
• The arithmetic. On £411,000, the difference between a total annual cost of 1% and one of 1.75% is £3,083 a year. Nobody in this story has done anything wrong. The client simply did not know the number, and could not have compared it with anything.
• What the accountant does next. Records the figure in the file, and — because judging whether that cost is good value for this client is regulated work — raises with the client whether it should be looked at by someone authorised to give that opinion.

The point of the exercise is not to find an overcharged client. Most of the time you will not. The point is that a client who cannot state the cost in pounds cannot exercise any judgement about it, and the FCA has now said in terms that the market is not making that easy for them.

A percentage is a number a client can repeat. A figure in pounds is a number a client can actually weigh.

The 18% heading for the exit

The other structural finding deserves more attention than it has had. The FCA reports that 41% of the firms surveyed plan to acquire another firm, grow revenue or increase their client base by more than 25% over the next two years — while 18% are considering winding down or selling all or part of their client base.

Read those two figures together and you have the mechanics of the concentration trend, running in real time. Roughly one firm in five is contemplating an exit; two in five are shopping. The clients in between get moved, and the move is rarely their idea.

An accountant sees the consequences of that transfer before almost anyone else, because they show up in the numbers. Charges change when a book moves onto the acquirer's platform. Holdings change when clients are migrated onto a house model portfolio, and a migration can crystallise gains that land on a tax return the client was not expecting. The named adviser often leaves within the earn-out period. None of that is a scandal; it is ordinary consolidation. But a client who has been told “nothing changes for you” and whose annual costs and holdings have both changed is a client whose accountant is in a position to notice.

What AI is, and is not, doing about the gap

The AI finding is modest and the regulator presents it modestly: the FCA recorded 13% of surveyed wealth managers using in-house or third-party AI tools, rising to 45% including those considering it. The FCA's framing is that used well, AI “can reduce friction, improve efficiency and help close the advice gap so more consumers who could benefit from support and advice receive it” — alongside warnings about fraud, cyber security and consumer harm.

Set that against the size of the gap. The lang cat's State of Advice Report 2025 found that only 9% of UK adults had paid for financial advice in the previous two years, while 91% of those who did take advice found it valuable. Efficiency gains reduce the cost of serving someone who has already walked through the door. They do nothing about the far larger group who never considered walking through it, and who will not be reached by a cheaper process. That group is reached by someone they already trust saying something at the right moment — which is the entire argument set out on the advice gap, and the reason the FCA's separate targeted support reforms matter, as covered in targeted support is live.

What is still open, and when we will know

Three things in this report are genuinely unresolved.

First, the vintage. The findings describe the market as at 31 December 2024. Whether the ten largest firms have gone past 89% since then will not be known until the next survey round, which on the pattern of 2022, the intervening round and this one is the point at which the trend either continues or breaks.

Second, fair value. The regulator has set out an expectation about pricing clarity rather than a rule change, and expectations of that kind usually precede supervisory work rather than replacing it. What follows — and whether it becomes anything more formal — is not yet published.

Third, AI. The 13% figure is a starting point measured eighteen months ago against a stated ambition of narrowing the advice gap. Whether adoption converts into more people served, rather than the same people served more cheaply, is exactly the sort of thing the next survey will show and this one cannot.

Two things worth doing this week

One: add the cost question to the year-end agenda. For every client you know holds an investment portfolio or a personal pension of any size, ask what it cost them last year in pounds, all in. Record the answer, and record it when there isn't one. It takes a minute, it is a factual question about their affairs rather than advice, and on a £411,000 portfolio the quarter-percent nobody could account for is £1,028 a year.

Two: check which of your clients' advisers have changed hands. Where a client mentions their wealth manager has been acquired or their adviser has moved on, flag the file for the next return. Migrations move charging structures and crystallise gains, and with the FCA recording 41% of surveyed wealth managers planning acquisitions or 25%-plus growth, the odds of this touching your client book in the next two years are not small.

Neither of those is a regulated activity, and neither requires an opinion on anybody's investments. Both put an accountant in a position to see something the client cannot — which is, in the end, the whole basis on which the two professions are useful to each other.

Common questions