Segregated mandates held 58% of UK institutional assets in 2024 — but that average hides a split by client type: insurers ran 70% of their money through segregated accounts, corporate treasury clients just 17%. Most descriptions of asset management skip that entirely and assume a single shape: a fund, a pool of money, a manager picking stocks or bonds inside it.
A very large share of the money asset managers run for a living isn't sitting in a fund at all. It's a portfolio built for one client, under one contract, to that client's own rules — a segregated mandate, run for a pension scheme, an insurer, a sovereign investor, or any institution with enough scale to want its money handled alone rather than pooled with everyone else's.
The difference isn't cosmetic. It changes who the client is, what document governs the money, what a broken rule looks like, and — the part almost nobody explains clearly — whose problem it is when somebody else wants their money back.
A segregated mandate is one client's own portfolio
A segregated mandate is a portfolio run exclusively for one client, to that client's own brief. The benchmark is theirs, the risk limits are theirs, the reporting cadence is theirs, and often so are specific restrictions on what the manager can and can't hold. The assets sit in an account in the client's own name, at its own custodian, not mixed with anyone else's.
That last point is the one candidates most often miss. Because the account belongs to the client, not a shared vehicle, the client can change managers whenever it wants, for whatever reason it wants, and nothing happens to anyone else's money. No fund to redeem out of, no other holders to consider. One phone call, and the relationship ends.
What that independence buys
- A benchmark the client chose, not one a fund already publishes. Client and manager agree it between themselves, which is what makes a bespoke or liability-driven benchmark possible at all.
- Direct legal ownership of the underlying assets. If the manager runs into financial trouble, the client's holdings sit untouched in its own custody account.
- The ability to exclude a specific holding or sector. A pooled fund can't carve out one investor's restriction without changing the fund for everybody; a segregated account can.
- Real-time visibility into every position, rather than the periodic, standardised disclosure a prospectus requires.
A pooled fund is many clients sharing one dealing cycle
A pooled fund flips the arrangement. Many investors accept the same strategy, benchmark and cost base, built to one prospectus, for professional management and diversification at a scale none could buy alone. A saver with a modest sum owns a slice of hundreds of positions through one purchase — something a segregated account that size could never hold economically.
Nobody in a pooled fund can ask the manager to deviate from its stated strategy on their behalf. A scheme running a segregated mandate can ring-fence a sector it refuses to hold; the same scheme, in a pooled fund, gets whatever the prospectus already allows or stays out entirely.
What changes when the vehicle changes
| What happens | Segregated mandate | Pooled fund |
|---|---|---|
| Another investor wants their money back | Nothing. There's only one investor | The manager may have to sell something to pay them, and the cost of that sale lands on you too |
| The client wants to exclude a sector or a name | Written into the mandate directly | Only if the fund's existing prospectus already allows it |
| The manager underperforms for two years running | The client can terminate and move, alone, on its own timetable | You can only redeem your own units; the fund and its other holders carry on regardless |
| A guideline is breached | Resolved privately between the client and the manager | Governed by the fund's own prospectus and its regulator, for every holder at once |
| The client wants to know every position, in real time | Standard | Rare; holders get periodic, standardised disclosure instead |
Other people's redemptions are your problem
This is the single most useful idea in the whole subject, and almost nothing written for candidates says it plainly: in a pooled fund, meeting another investor's exit is not free, and the cost doesn't fall on them. It falls on the fund, which means it falls on you.
Why a redemption is never really "somebody else's transaction"
A fund's price on any day reflects what it holds, not the cost of turning that into cash. When investors redeem, the manager has to sell something to raise the money, and selling costs money: the bid-ask spread, the market impact, sometimes a fire-sale price if it has to happen fast.
Those costs come out of the fund itself, so every investor who stayed absorbs a slice of them, whether or not they had anything to do with the decision to leave.
The investor who redeemed, by contrast, gets out at the price struck that day, before the cost of raising their cash has fully hit the fund. That mismatch is sometimes called dilution, and two tools exist to correct it: swing pricing, which adjusts the fund's valuation on heavy-flow days so departing investors carry more of the cost themselves, and an anti-dilution levy, which charges the transacting investor directly instead of spreading it across everyone who stayed.
A segregated mandate has no version of this problem. There's one investor, so there's nobody else's exit to fund and no trading cost to absorb on their behalf. That isn't a claim that segregated portfolios never lose money; it's a claim that this specific failure mode, other people's decisions showing up in your return, structurally can't happen to you.
Test yourself
Interview levelA pooled fund holding illiquid stocks faces months of heavy redemptions. Why does that leave the investors who stayed worse off than the ones who already left?
Who gets which, and why
There's no official size at which a scheme "graduates" into a segregated mandate, and treating any number you hear as a fixed rule is a mistake — the trade-off has moved for years and will keep moving as custodians make bespoke setups cheaper to run.
What decides it is a comparison: the cost of running one client's own account, with its own reporting and custody relationship, against the value of a portfolio built to its exact brief. Below a certain scale, that comparison doesn't clear, and pooling wins on cost.
What the split looks like today
The Investment Association's 2024-25 survey found UK institutional investors, taken together, kept the majority of their assets in segregated mandates rather than pooled vehicles — 58% segregated against 42% pooled in 2024, up slightly from 41% pooled the year before. The average hides a useful pattern once you split it by client type.
Insurers run liability-matched, cash-flow-aware portfolios that suit a bespoke mandate. Corporate treasury clients mostly want same-day liquidity, which a pooled cash fund delivers more cheaply than a segregated account ever could.
Insurers were already the most segregated client type, at 70% in 2024, up from 65% the year before, because their portfolios exist to match a specific book of liabilities.
Corporate treasury clients sat at the other end, just 17% segregated, because what they mostly need is same-day liquidity, which a pooled cash or money-market fund delivers more cheaply than a bespoke account ever could.
In round terms, insurers are four times as likely to sit in a segregated mandate as a treasury client — the same skill, applied to two different jobs.
Pension schemes sit in the middle, and the line is blurring
Pension funds sat closer to the middle, at 51% segregated, and the trend is worth knowing: the pooled share of pension assets rose to 49% in 2024, from 44%, driven largely by shrinking liability-driven books since their 2021 peak and by the growing use of what the Investment Association calls "fund-of-one" structures — a bespoke portfolio for a single client, wrapped inside an authorised fund for administrative reasons.
A fund-of-one behaves like a segregated mandate in every way that matters to the client; it just gets counted as pooled, a reminder that the two categories blur more than a glossary admits.
Signs a scheme has outgrown a pooled fund
- A benchmark that doesn't fit any existing pooled range. A liability-based target, in particular, usually can't be bought off the shelf.
- A restriction or exclusion the pooled range can't accommodate. Ethical screens, sector exclusions and single-name limits are all easier to write into a bespoke mandate than to negotiate into an existing fund.
- Enough governance capacity to run a direct custodian relationship. A segregated mandate hands the client more control and more to manage.
- Enough scale that the extra administrative cost is worth it. This is the one that keeps moving, as platforms and custodians make bespoke setups cheaper to run than they used to be.
Test yourself
Partner levelIn the UK's 2024 institutional data, insurers used segregated mandates far more than corporate treasury clients did. What best explains that gap?
The Investment Management Agreement is the actual document
Every segregated mandate is governed by a contract called an Investment Management Agreement, or IMA — worth reading as an actual document rather than an abstraction, because a real one looks nothing like a prospectus.
Franklin Resources filed a representative version of its own institutional IMA publicly, as an exhibit to a regulatory filing. It runs to 21 sections covering the manager's authority, the client's guidelines, fees, liability, reporting and termination. Two details in it explain how the relationship runs day to day, not how it looks on paper.
How the document works
| What the IMA covers | What it actually specifies |
|---|---|
| Authority | Whether the manager has full discretion, or needs sign-off before certain trades |
| Guidelines and restrictions | A schedule the client sets and can amend at any time, covering permitted assets, concentration limits and any exclusions |
| Benchmark and objective | What the manager is being measured against |
| Fees | Negotiated per client, usually as a tiered schedule that steps down as assets grow |
| Reporting and valuation | How often the client is updated, in what format, and how the account is priced |
| Liability | What the manager is, and isn't, on the hook for if something goes wrong |
| Termination | How much notice either side has to give to end the relationship |
First: the client's investment guidelines aren't fixed at signing. They're a schedule the client can amend at any time in writing, and the manager gets a reasonable period to bring the account into line, rather than being expected to rebalance instantly regardless of cost. That language exists because an instant, cost-blind unwind can hurt the client more than a short, orderly one.
Second: the fee schedule in a real, filed representative agreement is left blank — a tiered template reading "a percentage of the first tranche of assets, a lower percentage of the next tranche, a lower percentage on the balance" — with the actual numbers negotiated client by client, not published anywhere.
The Investment Association publishes its own model version for UK pension trustees to adapt, built by a working group spanning dealing, compliance, legal and operations — proof of how standard the document's shape is, even though every number inside it gets negotiated fresh.
Guidelines and breaches: a daily discipline, not paperwork
Every segregated mandate has a compliance function, sometimes called guideline monitoring, whose entire job is watching whether the portfolio still sits inside the IMA's limits, in real time, every trading day. It's one of the least glamorous jobs in the industry and one of the most consequential: a breach unnoticed for a week is a far bigger problem than one caught in an hour.
Passive breaches and active breaches aren't the same thing
The industry draws a sharp line between two kinds of breach, worth having ready for an interview, because it shows you understand that a crossed limit isn't automatically the manager's fault.
- A passive breach is caused by the market, not a decision. One holding's price rises sharply and a concentration limit is crossed purely as a result, with no trade behind it.
- An active breach is caused by the manager. Either a trade that pushes a position over its limit, or a failure to act when the guidelines required one.
- The two get monitored differently. A passive breach is watched and brought back within limits over a reasonable period; an active breach is treated as an operational incident and escalated straight away.
- Neither is automatically a disaster. What matters is how fast it's caught, how it's explained, and how it's fixed.
A candidate who explains that distinction unprompted, in a mandate-facing or operations interview, shows something more useful than memorised vocabulary: an understanding that a real portfolio runs inside rules somebody is watching, every day, not just when a client asks.
Test yourself
Interview levelA pension scheme changes its investment guidelines partway through an existing segregated mandate. Under a typical institutional IMA, what happens next?
Bespoke benchmarks: when the benchmark is the client's own liabilities
Every benchmark discussed so far in this industry is, at heart, a published index: a basket of stocks or bonds somebody else built. A segregated mandate can throw that assumption out entirely, and the clearest example is a pension scheme whose benchmark is its own future liabilities rather than any market at all.
What a liability-driven benchmark is built from
A liability-driven investment, or LDI, mandate exists to hedge a pension scheme's exposure to interest rates and inflation, because what the scheme has promised to pay its members moves with both. The benchmark isn't the FTSE or a bond index; it's the scheme's own projected cashflows, discounted using gilt or swap rates plus a fixed margin.
Some schemes build a more dynamic version that also accounts for credit-spread sensitivity and how inflation-linked payments adjust, so the hedge tracks the liability as closely as the actuarial picture allows.
This is, in a literal sense, LGIM's business. Now folded into what its parent group calls simply Asset Management, the firm built one of the UK's largest liability-driven investment books on exactly this idea, benchmarked to what a pension scheme actually owes rather than any index it could otherwise have chosen. A candidate who can explain what a liability benchmark is made of, rather than just naming LDI, is answering the question the seat is built around.
Test yourself
Warm-upA segregated portfolio drifts over a concentration limit purely because one holding's price rose sharply, not because the manager traded. What kind of breach is that?
Fee differences, and why a large mandate negotiates
A pooled fund's fee is public: printed in a factsheet as an expense ratio, identical for every unit-holder, whether they invested a thousand pounds or a hundred million. A segregated mandate's fee doesn't work like that, and understanding why is worth more in an interview than any specific number.
The fee and the cost of running the account are two different things
Running a bespoke account, with its own custody relationship, reporting and operational setup, costs a manager more per pound managed than adding one more client to an existing pooled fund. That's simple arithmetic: a pooled fund spreads its fixed running costs across every holder, and a segregated account has nobody to share them with.
Despite that, a large segregated client can still pay a lower rate than a pooled fund's published fee, for two reasons that have nothing to do with the cost of administering the account.
First, a segregated fee is negotiated bilaterally, not accepted off a rate card, and a mandate big enough to matter to the manager's own revenue has real leverage. Second, a pooled retail fund's published fee usually carries a layer of distribution and platform cost that a direct institutional relationship skips entirely.
Test yourself
Warm-upA pension scheme's segregated mandate is benchmarked to its own liabilities rather than a market index. What is that benchmark actually built from?
What a junior does on the mandate side
The distinction isn't just a definition to recite; it maps onto different jobs, and knowing which one a role sits behind changes what to prepare for.
- Client service on a segregated book. Reconciling one client's portfolio against its own bespoke benchmark, preparing bespoke reporting, and answering directly to a pension scheme's investment committee, not an anonymous unit-holder base.
- Consultant relations. Responding to institutional tenders and managing the investment consultants who advise schemes on which manager to hire, knowing the consultant is rarely the client, only the adviser beside them.
- Guideline and compliance monitoring. Watching portfolios against their limits in real time, distinguishing a passive drift from an active decision, and escalating the ones that need it.
- Pooled fund and product roles. NAV and unit-pricing processes that run identically for every investor, fund-wide compliance, and distribution work aimed at growing a shared vehicle rather than servicing one relationship.
Why "who is the client" is worth asking in an interview
A fund manager talking about "the client" in a segregated context usually means the asset owner itself: the pension scheme, the insurer, the sovereign investor. That's not always who the manager's day-to-day contact is. An investment consultant often sits between the two, advising the asset owner on which manager to hire and how to monitor it, paid directly by the asset owner rather than out of the mandate's fee.
Mixing the consultant up with the client is a quiet tell that a candidate hasn't thought about the structure underneath the job. Asking, in an interview, exactly who a seat's client is, the asset owner, the consultant, or both, does more work than any rehearsed definition, because it shows you already understand that the answer changes what the job is.
Test yourself
Partner levelRunning a segregated account costs a manager more per pound than adding that client to an existing pooled fund. Why can a large segregated mandate still end up paying a lower fee?
The mistakes that give away a candidate who's only read the definitions
- Treating "segregated" as a synonym for "safer." It changes ownership and control, not investment skill or risk.
- Confusing the consultant with the client. The asset owner pays the bills and owns the assets; the consultant only advises them.
- Assuming every breach is the manager's fault. A passive breach, caused purely by market movement, is treated differently for a reason.
- Reciting a fixed £ threshold for "big enough to go segregated." No such number exists; the real answer is a cost-versus-benefit comparison that keeps moving.
- Assuming a lower fee means cheaper to run. Usually the opposite — the account costs more to administer and can still be priced lower, because the fee is negotiated, not cost-plus.
How to prepare, in order
- Learn the one-sentence version of the core distinction: one client's portfolio against many clients sharing a vehicle, and who bears the cost when someone leaves.
- Have the Woodford example ready, not as trivia but as the mechanism: illiquid holdings, sustained redemptions, and a cost that fell on the investors who stayed.
- Know what an IMA contains — guidelines, benchmark, fees and termination — rather than treating it as a vague legal formality.
- Explain a passive breach against an active one, in your own words, without needing to recite a regulation.
- Have one line ready on liability-driven benchmarks, especially near LGIM or another large LDI manager.
- Ask who the client is whenever a segregated mandate or consultant comes up. It's a stronger signal than any recited definition.
Where to go from here
The vehicle-level vocabulary built on top of all this, UCITS funds and ETFs, is covered in asset management fund structures. Position sizing and concentration limits once money is inside a mandate are covered in portfolio construction interview questions, and LGIM's own liability-driven business is covered in the LGIM interview guide.
One client, one contract
Most of what gets written about asset management assumes the job is picking assets inside a fund. A large share of the industry runs on a different relationship: one client, one contract, one set of rules nobody else gets a vote on.
A segregated mandate and a pooled fund can hold identical securities and still be different jobs — somebody else's redemption is genuinely your problem in one and structurally can't be in the other, and the document governing a segregated account is a real, negotiated contract, not a formality.
A candidate who can trace that through to fees, benchmarks and the daily discipline of watching a limit has understood something about this industry that "the manager beats a benchmark" never will.