The floor, the roles, and how it actually works

Sales and Trading is the part of a bank that sits between the world's largest financial institutions and the markets where securities are bought and sold. That is the simplest true description of it. The bank is the intermediary. S&T is how that intermediation happens.

The clients — pension funds, hedge funds, asset managers, sovereign wealth funds, corporates — need to buy bonds, sell equities, hedge currency risk, execute rate swaps, and do hundreds of other things in size. They cannot do this directly. Markets are not organised as neat exchanges where you type in an order and it fills. For most products — credit, rates, FX, structured products — you pick up the phone, or you send a Bloomberg message, and you deal with a bank. The bank takes the other side of your trade, finds you a price, and manages whatever risk it is left holding.

S&T is that operation. It is a service business, in the same way a law firm or a consulting firm is a service business — except the service is liquidity provision, and the product is a price.

The term "sales and trading" is also slightly misleading, because the division contains more than just salespeople and traders. It includes sales-traders (a hybrid role), structurers (who design bespoke products), quants (who build pricing models), and research analysts (who provide market views to clients). All of these functions sit within or adjacent to the S&T floor.

What S&T is not: it is not asset management. Asset managers invest money on behalf of clients, with discretion over what to buy and sell. S&T does not manage portfolios. It facilitates transactions. When a client wants to execute, S&T provides the price and the execution. When the client wants advice on what to buy, that is research. The distinction matters — and interviewers will notice if you blur it.

Sell-side vs buy-side is the most fundamental distinction in institutional finance. The sell-side (banks doing S&T) sells financial services — primarily liquidity and distribution. The buy-side (asset managers, hedge funds, pension funds, insurers) buys those services. The sell-side is in the business of serving the buy-side. Understanding this dynamic explains almost everything about how desks behave — why certain clients get better prices, why relationships matter, why market share is tracked obsessively.

A trading floor at a major bank is a large open-plan space — typically a full floor of a building, sometimes more — organised by product group. The rates desk sits near rates research. FX sits near macro. Credit sits near credit research and syndicate. The physical layout is not random. It follows information flow.

Within any product group you will generally find three categories of people sitting close together: traders, who make prices and manage risk; salespeople, who talk to clients; and sales-traders, who execute client orders in the market. Research analysts sit nearby but are technically a separate function — they are covered by information barriers and cannot share certain information freely with the trading desk.

The floor is divided first by asset class, then by product within that asset class. At a large bank like Barclays, Deutsche Bank, or Goldman Sachs, the major groupings are broadly: Fixed Income, Currencies and Commodities (FICC), and Equities. These are the two main S&T divisions. FICC covers rates, credit, FX, commodities, and structured products. Equities covers cash equities, equity derivatives, and prime brokerage.

Research is worth addressing separately because it confuses a lot of applicants. Research analysts produce written reports, models, and recommendations on markets, companies, and economic conditions. Their output is distributed to buy-side clients as a service — it is how banks justify asking for order flow in return. Research sits formally outside the trading floor due to information barrier regulations, but in practice the two functions are deeply intertwined. A rates trader will read rates research every morning. A salesperson will often relay a research view to a client and invite them to trade off it.

At European banks operating in London — Barclays, Deutsche, HSBC, BNP — the floor structure also reflects the firm's geographic franchise. EM desks are usually larger. Sterling rates will sit close to the desk that trades gilts at auction. The specific layout tells you something about where the bank's strength lies.

Knowing the product landscape matters because interviews will often lead to a question about where you see yourself. The answer needs to reflect genuine understanding of what the desks do — not a list of names from a Wikipedia article.

This is where most candidates have a significant gap in their understanding — and where being able to speak clearly puts you ahead of the majority of the room.

The naive answer is: "the bank takes the spread." That is partially true, but it is the beginning of the answer, not the whole thing. There are three distinct sources of revenue in S&T.

1. The bid-offer spread. When a client asks for a price on £50m of 5-year gilts, the trader quotes a bid (what the bank will pay) and an offer (what the bank will sell at). The difference — say, half a basis point — accrues to the desk on every trade. On a product as liquid as gilts, this spread is tiny. On an illiquid EM corporate bond or a bespoke structured note, it can be tens of basis points. Spread income is reliable, low-risk, and directly tied to volume of client activity.

2. Warehousing risk. When a client sells the bank £500m of 10-year credit, the bank does not instantly find another buyer. It warehouses that risk — meaning it holds the position on its book, often for hours, days, or longer — and manages the exposure while it works to distribute it. During that period, the market can move. If it moves in the bank's favour, the desk makes money beyond the initial spread. If it moves against them, they lose. This warehousing function is why traders exist: someone has to decide how much risk to hold, when to hedge, and when to run the position.

3. Discretionary positioning. Post-Volcker and post-Basel III, banks cannot take purely proprietary positions. But they can and do have views. A rates trader who is running a £2bn book of gilts will have that book positioned in a way that reflects their view of the curve, even if every individual trade was initiated by a client. This grey area — between market making and having a position — is where most of the interesting risk management happens. Desks that consistently position well over a cycle generate meaningfully more revenue than those that do not.

How a desk's P&L is actually built. A desk's revenue for a given day comes from: net spread income on trades executed; mark-to-market gains or losses on the book as prices move; and any hedging costs incurred. At the end of the day, every position in the book is marked to market — meaning the current market price is used to value it. This creates daily P&L visibility, which is both a discipline and a source of pressure. A trader who runs a large position that moves against them at 3pm will know about it immediately.

The desk's overall economics are tracked on a weekly, monthly, and annual basis. Revenue per client is monitored. Desks with strong client franchises — meaning clients who consistently direct flow to them — generate more stable revenue than desks that rely on one-off trades. This is why the salesperson's relationship with the client is worth real money to the bank.

Walk through a real example. A large UK asset manager wants to sell £200m of 10-year investment grade sterling corporate bonds from its portfolio. Here is what happens.

This is where interviews get specific. You need to be able to describe each role in a way that reflects genuine understanding — not a list of job description bullet points.

A desk does not just exist. It justifies its existence every day through a set of metrics that senior management tracks closely. Understanding these is important because it tells you what people on the desk actually care about — and what they will care about when you are sitting with them during an internship.

P&L. The primary measure. Revenue is tracked daily, weekly, monthly, and annually. A desk that is consistently unprofitable will be restructured, downsized, or shut. P&L comes from spread income, warehousing gains and losses, and hedging costs. Senior traders and desk heads have a sense of what their desk's "run rate" revenue should be — and they are acutely aware of how actual performance tracks against it.

Risk metrics. The risk team monitors DV01 (duration sensitivity for rates products), delta (equity sensitivity for equity derivatives), credit spread DV01 (credit exposure), and Value at Risk (VaR) across all positions. Exceeding risk limits is a serious matter — it requires immediate escalation and can result in forced position reduction. One of the most important things a junior trader learns is how to manage a book within limits rather than purely for P&L.

Client franchise metrics. Salespeople are evaluated on wallet share — the proportion of a client's total trading activity that is directed to this bank versus competitors. A client who does 20% of their credit flow with your bank is a different relationship to one who does 80%. Wallet share is measured systematically via client surveys (the Greenwich Associates survey is the most widely referenced in fixed income) and through internal revenue attribution. A salesperson who is growing wallet share with key clients is building long-term value even if current-year revenue is modest.

Market share and league tables. Banks track their position in league tables — rankings of which bank did the most volume in a given product or region. Being number one in sterling IG credit or European rates matters because it signals to clients that you have the deepest franchise, best pricing capability, and most reliable execution. Losing league table position is a reputational concern that gets escalated quickly.

The S&T career path has a clear structure. Understanding it matters because it shapes what you will be asked to do, who your role models are, and what progression actually looks like.

Analyst (years 1–3). The learning phase. In sales, you spend the early years building product knowledge, learning how to communicate with clients without embarrassing yourself or the bank, and making yourself genuinely useful on the desk — running reports, maintaining spreadsheets, preparing morning notes, and gradually taking on smaller client relationships. In trading, analysts shadow senior traders, learn to manage small pieces of the book, and develop the market intuition that cannot be learned from reading alone. The expectation is not that you are immediately valuable — it is that you are absorbing as much as possible and not creating problems.

Associate (years 3–6). Responsibility increases meaningfully. In sales, associates are typically running a defined set of client relationships independently, attending client meetings, and contributing to revenue in a measurable way. In trading, associates are managing portions of the book with real risk limits attached. The transition from analyst to associate is where many people either accelerate or stall — it requires moving from "doing what you are told" to "initiating."

VP (years 6–10). By this point, a salesperson is expected to be an expert on their client base and product set, and to be generating meaningful revenue independently. A trader at VP level is managing a significant position with real accountability for P&L. VPs also begin to have a management dimension — mentoring junior staff, running the desk when the MD is absent.

Director and MD. MDs are the most senior client-facing people in S&T. A managing director in credit sales will have relationships with the heads of fixed income at the largest asset managers. They set the strategic direction for their client franchise, manage the most sensitive client situations, and are responsible for retaining key clients. At this level, compensation is heavily driven by the revenue you can be attributed with generating.

A summer analyst programme in S&T typically runs for 8–10 weeks. You will rotate across desks — usually 2–3 rotations — and at the end of the summer, offers are extended based on a combination of desk performance, relationships built, and fit with the division's needs for that year.

What you will actually do. The work varies enormously by desk. On a rates or credit desk you might be asked to produce a daily morning note summarising overnight market moves, maintain a spreadsheet tracking client flow or positions, research a specific trade idea, or help the sales team prepare client presentations. On an equity derivatives desk you might run Greeks reports, update a vol surface, or work on a structured product pricing model. The quality of your experience depends heavily on the desk and on your own initiative — there is no guaranteed curriculum.

What actually gets you an offer. Analysts who convert internships to full-time roles do several things consistently. They learn the product fast — not just the vocabulary, but the mechanics. They make themselves useful without needing to be told what to do each morning. They are present on the floor — visible, engaged, asking intelligent questions at appropriate moments. They treat everyone with respect: the operations person booking the trades, the analyst two years senior, the MD who walks by once a week. And they do not overclaim — pretending to know something you do not is far more damaging than being honest about a gap in your knowledge.

The rotation structure matters. If you are at a bank with a rotational programme, your first rotation determines your second, and your second has a large influence on where you end up. Take the first rotation seriously even if it is not your first choice product. The people you impress there will often speak to the people running your second rotation before you arrive. Reputation travels faster on a trading floor than you expect.

Networking during the internship. Most large banks have structured intern networking events, but the informal networking is more important. Ask for a coffee with an analyst two years ahead of you on a desk you are interested in. Ask the salesperson you sit next to whether you can listen to a client call. These asks, done at the right moment and in the right way, are almost always welcomed — it shows initiative without being intrusive.