Bespoke Portfolio Management for HNW Clients: When It Earns Its Fee

A 62-year-old client walks in with GBP 3.4m. GBP 1.1m sits in a single AIM-listed share she received when her employer floated in 2014. GBP 600,000 is in a SIPP, GBP 950,000 across two ISAs and a GIA, and the rest in a discretionary trust she set up for her grandchildren. She wants income from age 65, full ESG screening, and absolutely no tobacco or defence exposure. No model portfolio service can run this. That is when bespoke earns its fee.

The bespoke discretionary conversation gets muddled because the industry sells it on prestige and the academic literature treats it as a footnote on MPS. Both miss the practical point. Bespoke is not a luxury tier; it is the right tool for a specific set of constraints. Where those constraints are absent, a well-run model portfolio service almost always delivers better net outcomes. Where they are present, an MPS quietly fails the suitability test even when the headline returns look fine.

This is a working note on where bespoke portfolio management actually pays off in 2026, what it should cost, and how to test whether the provider is delivering it.

What “bespoke” really means under the bonnet

Strip the marketing and bespoke discretionary management has three structural features that an MPS does not.

First, a single portfolio is built and run for one client, not a model run across many. The manager can buy stock A for client X and stock B for client Y on the same day, with no obligation that client Z holds either. Second, the mandate document captures constraints that a model cannot encode: realised-gain budgets, concentration limits, sector exclusions, currency overlays, named legacy holdings to retain, dividend-yield floors. Third, the rebalancing logic is tax-aware on a per-client basis, so the manager will sit on a position past a model trigger if realising it would crystallise a six-figure CGT charge with no offset.

These are real mechanical differences, not branding. They are also expensive to deliver, which is why bespoke fees sit where they do and why the threshold for switching from MPS is genuinely material.

The honest fee picture in 2026

UK practice as of mid-2026:

Service tier	AMC range	Fund or instrument cost	Custody and transaction	Typical all-in
MPS (passive)	0.10 to 0.20%	0.10 to 0.20%	0.05 to 0.10%	0.30 to 0.45%
MPS (active or blended)	0.20 to 0.30%	0.30 to 0.50%	0.05 to 0.15%	0.55 to 0.80%
Bespoke (entry, GBP 250k to 1m)	0.50 to 0.90%	0.10 to 0.30%	0.10 to 0.20%	0.70 to 1.30%
Bespoke (HNW, GBP 1m to 5m)	0.40 to 0.75%, tapering	0.10 to 0.25%	0.05 to 0.15%	0.55 to 1.05%
Bespoke (UHNW, GBP 5m+)	0.25 to 0.50%, tapering	nominal	negotiated	0.40 to 0.70%

A few things drop out of that table that the brochure version does not show. The all-in cost of premium MPS and entry-level bespoke can overlap. The marginal cost of moving an existing GBP 2m client from MPS to bespoke is usually 25 to 50 basis points, not the 75 basis points implied by headline AMC comparisons. And UHNW bespoke fees compress quickly above GBP 5m; if you are seeing flat 0.50 percent on a GBP 8m mandate, you are negotiating against an opening offer, not the floor.

The five suitability triggers that justify bespoke

Most clients do not need bespoke. The ones who do tend to share at least one of these triggers. None is a brochure flourish; each one is a feature an MPS cannot replicate.

1. Concentrated low-cost-base holding. A founder share, an inherited stake, a vested RSU position. The cost is realised CGT, the constraint is a phased disposal plan over multiple tax years, and the manager has to construct the rest of the portfolio around the residual concentration. No model can do this without forcing a sale.
2. Hard exclusions or values-based constraints. Specific tickers, sectors, jurisdictions or thematic exposures the client refuses to hold. Even SDR labelled funds rarely match a particular client’s exclusion list. A bespoke mandate can run a screen list against every trade ticket.
3. Multi-wrapper tax overlays. Where a client holds material assets across SIPP, ISA, GIA, onshore bond and trust, the rebalance logic should hit the GIA last and put income-generating assets in the SIPP. An MPS rebalances all wrappers identically.
4. Non-UK domicile or international tax exposure. US-connected clients, returning expats, dual-resident HNWs. The portfolio has to navigate PFIC rules, withholding-tax treaties and reporting status of underlying funds. This is a per-client construction problem, not a model setting.
5. Trustee or corporate ownership structure. Trusts, family investment companies and onshore investment bonds each carry their own tax wrapper rules, beneficiary considerations and reporting requirements. The mandate has to be drafted around the structure, not bolted on.

A client without any of these triggers is usually a textbook MPS case. The fair-value question under Consumer Duty is then whether the bespoke provider is doing more than charging a premium for a near-model portfolio. If the answer is no, the recommendation needs to change.

Where bespoke quietly fails the suitability test

The other side of the same regulatory question is rarer in conversation but more dangerous in practice. There are three patterns I see in file reviews where bespoke is being delivered but is failing on its own terms.

The first is the “model in disguise.” The client pays a 0.75 percent bespoke AMC, but the portfolio holds the same 18 funds as the firm’s growth model with two minor swaps. There is no concentrated position, no exclusion list, no tax overlay. The mandate document says “tailored” but the trade tickets show otherwise. Under Consumer Duty’s price and value outcome, this is the easiest fair-value case to lose.

The second is stale construction. A bespoke portfolio set up in 2018 around a now-irrelevant constraint, never rebalanced beyond income reinvestment, with the manager still charging full AMC. The portfolio drifted into a passive concentrated holding several years ago. The right answer is either to refresh the mandate or move the client to a cheaper service.

The third is scale mismatch. A bespoke service designed for GBP 1m+ clients gets sold to a GBP 350,000 client because the firm has no MPS option. The client pays bespoke fees on a portfolio that runs as a near-MPS by necessity. A practice that recommends bespoke below the firm’s stated bespoke threshold should expect questions in any FCA review.

The eight due diligence questions that work

If you are evaluating a bespoke discretionary partner, written brochures will not separate them. These questions will. Send them in advance and read the written answers.

1. Who personally manages this portfolio, and how many other portfolios does that individual run? A target of 80 to 120 portfolios per lead manager is workable. Above 200, the service is operationally an MPS even if the wrapper says otherwise.
2. What is the average and median client size on this strategy? If your client sits below the median, the service will be designed around someone else’s needs.
3. Show me a redacted current statement, factsheet and proposal for a comparable client. Read the trades, look at concentration and turnover, check whether the construction is genuinely client-specific.
4. How are exclusions, concentrated holdings and tax constraints implemented? Manual flag, system rule, or a “we tell the manager”? Manual flagging breaks under volume.
5. What is the realised-gain budget policy for taxable accounts, and how is it set? A defensible answer is a pounds-or-percent annual cap negotiated with the adviser. “We try to be tax-aware” is not an answer.
6. What is the all-in cost, broken down by AMC, fund OCFs, transaction costs, custody and any platform overlay? A provider that cannot produce this in writing in 24 hours is not ready for fair-value scrutiny.
7. What is the regular reporting cycle and what does it actually contain? Quarterly performance versus a benchmark and a generic commentary is the floor. A bespoke service should report against the client’s specific objectives, exclusions and tax position.
8. What is the custody arrangement and the counterparty? This matters more than people think; see the role of institutional custody in client confidence for the longer treatment.

A provider that answers seven of these well and ducks one is probably worth a meeting. A provider that ducks three should not progress.

How regulators frame the assessment

The regulatory floor is set by the FCA’s Consumer Duty rules and Policy Statement PS22/9, which require that the price the client pays is reasonable relative to the benefits provided. For bespoke, those benefits have to be visible in the portfolio, not just the marketing. The suitability and appropriateness rules in COBS 9A apply directly to discretionary portfolio management; the firm offering the service has to demonstrate, on file, that the mandate matches the client’s circumstances. The wider SYSC rules on systems and controls govern how that demonstration is recorded.

For the broader policy frame on discretionary services, the FCA’s MiFID II pages remain the underlying source for inducements, costs and charges disclosure, and best execution.

A practical segmentation rule for advice firms

Most well-run UK practices end up with a three-tier segmentation that names the bespoke trigger explicitly. Something like:

• Tier A (MPS, risk-graded). Clients with portfolios below GBP 500,000, single risk profile, no concentrated holdings, no exclusions, single or two-wrapper structure.
• Tier B (premium MPS or hybrid). Clients GBP 500,000 to GBP 1.5m, possibly with mild ESG preferences and multi-wrapper holdings, served by an enhanced MPS or a small set of model overlays.
• Tier C (bespoke discretionary). Clients with at least one of the five suitability triggers, regardless of size, plus default bespoke above GBP 1.5m.

Spell it out in the firm’s centralised investment proposition, record it in the client file note, and review whenever a Tier B client crosses a Tier C trigger. That is the documentation pattern that survives a Consumer Duty review.

When the answer is “do nothing”

Two situations where the right call is to leave a bespoke arrangement alone, even if the rate-card maths looks ugly.

A long-running mandate with a concentrated legacy holding the client refuses to sell. Switching out crystallises the holding or forces a transfer-in-specie that complicates the new manager’s construction. The current arrangement may not be ideal, but the cost of moving is real.

A bespoke arrangement set up around a complex trust where the trustee, settlor and beneficiaries are all in agreement and the manager has built bespoke reporting around the structure. Tearing this up to capture 25 basis points of fee saving is rarely net positive once trustee meetings, legal review and re-papering are priced in.

Both are situations where the tax implications of the existing arrangement outweigh the headline fee saving. The right Consumer Duty file note records the trade-off and reviews it annually.

What I would actually do this quarter

If you have not done so since the start of the 2026/27 tax year, three practical steps.

Pull a list of every client paying bespoke fees with portfolios under your firm’s stated bespoke threshold. For each, record the suitability trigger or document the move to a cheaper tier. This is the highest-priority file remediation under Consumer Duty.

Pick the three largest bespoke clients by AUM and ask the provider for the eight due diligence answers above in writing. The exercise is more useful than any annual review meeting.

Update the firm’s segmentation document so it names the five triggers explicitly. The most common file-review failure is a segmentation written in 2023 that has not been updated for the post-Consumer Duty fair-value regime.

Bespoke portfolio management is the right answer for a meaningful minority of HNW clients. It is the wrong answer for the rest. The work is being clear about which is which and being able to show it on the file.