Customization at scale

The first five pieces were about getting the risk right — replaying real market crises, repricing every instrument in full, modelling how assets truly move together, and gathering and cleaning the data behind it all. None of it cut a corner. This piece is about what comes next: once those analytics exist, how do we make them easy to use — running for you in minutes, configured to your own rules, and open to build on?

Every bank, every desk, every client wants something slightly different from a risk engine — a different question answered, a different rule enforced, a different system to plug into, a different region to satisfy. One way to meet all that is to build something bespoke each time. We do the opposite. We compute a single set of analytics, to the standard the past five pieces have described, then make that one foundation easy to reach, shape and build on — so it fits thousands of different needs without being rebuilt for any of them. That is customisation at scale: one accurate engine, made your own.

Quality is non-negotiable

We will not cut a corner on the quality of the analytics. That principle has run through every article in this series, and it holds firm at the end: the ease we are about to describe is never bought by lowering the standard of the numbers underneath. The hard work simply sits with us instead of with you.

Sourcing data from multiple top-tier providers and standardising it. Modelling how assets move together in the tails, not just on average. Keeping the feeds honest and flagging uncertainty rather than hiding it. We carry all of that. What reaches you is a full suite of risk analytics, ready to use — Expected Shortfall and value-at-risk are only the most familiar of them, sitting alongside risk contributions, sensitivities, scenario impacts, suitability grades and more, computed at the level of each individual asset and of the whole portfolio.

Two things make that suite distinctive. The first is coverage. We span the entire universe of assets a wealth portfolio actually holds — not the convenient subset, but the real one: equities and funds, every flavour of bond, options and other derivatives, and the complex structured products that are the hardest of all to price. A single risk number means the same thing across all of them, so nothing has to be left out of the picture because it was too awkward to model.

The second is how we get there. Every instrument is priced by full repricing — modelled as it is actually written, in full, rather than approximated by a linear shortcut. It is the most accurate method there is, and for anything with optionality — a barrier, a coupon that switches on under conditions, a payoff that bends — it is the only method that tells you the truth. We do the version that is hard to compute so that the answer reflects how the product really behaves.

That overnight work is not small. After the market closes we run more than 1.5 billion calculations, repricing every instrument in your universe — over 600,000 of them, driven by more than 260,000 underlying risk factors — and we finish in under six hours, before the first desk opens. By the time you ask a question, the answer already exists. It comes back in about half a second.

Ready in minutes

Because we have absorbed that complexity, getting started is not a project. There is no data to license, no pricing library to build, no overnight infrastructure to stand up. You connect, and you are running — in minutes, not the months a foundation like this usually takes. Plug in wherever you already work.

Build on top

The analytics are a foundation, and a foundation is meant to be built on. We have designed them to be used, not just read.

Technically, that means meeting you where you are. Pull the numbers into Excel and work in a spreadsheet. Plug them straight into the software your teams already use, through a small set of interfaces — a Risk API for the core measures, a Universe Asset API for instrument-level data, a Scenarios API for stress tests and what-ifs. Or build an interface of your own on top of them, exactly the way you want your people to see risk.

It goes further than the technical. A foundation this solid is something you can build a service on — even a business. We provide the engine and the analytics; what you construct above them is yours. We have done the part that is hard to do well and tedious to maintain, so that your effort goes into the part only you can do: serving your clients.

One set of analytics, many uses

The same computed analytics answer very different questions. Where is the risk actually coming from in this portfolio? Will a client's plan still reach its goal through a poor decade? What goes into the monthly report? Does this position keep the account within the rules? Risk contribution, goal-based planning, reporting, compliance — these are not separate engines or separate calculations. They are different uses of one consistent set of numbers, which is exactly why they agree with one another.

Configuration on top

On top of all that, you set the parameters that make the analytics reflect your house and your clients. The confidence level behind a measure. The regions and regulatory regimes you operate in. The thresholds that define what is acceptable and what is a breach. The methodology underneath stays consistent and faithful; the choices layered over it are yours to make.

Suitability is a good example of where this leads. It is a set of analytics built for compliance — a Product Risk Classification for each instrument and a Portfolio Risk Grade for the whole portfolio, produced by one coherent methodology across every instrument type, so a rating of "3" means the same thing whether it sits on a government bond, an equity fund or a complex structured product. That consistency is what lets you compare and aggregate with confidence. Around it, you configure: the regions you serve, the regulatory frameworks that apply, your own house limits and a particular client's tolerance for loss. Same analytics underneath, your rules on top.

One engine, made your own

Six articles have made a single argument: do the hard thing first, and never quietly undo it. We replay real crises rather than inventing tidy ones, reprice every instrument in full rather than approximating it, model how assets truly move together, source the data ourselves, and say so plainly when it breaks. The work is heavy, and it stays with us.

What reaches you is one accurate set of analytics, and the freedom to make it your own. You can reach it however you work, build whatever you need on top of it, and shape it to your house, your clients and your regions — one engine, put to every use. That is what we mean by customisation at scale.

The complexity stays with us; the clarity is yours. A clearer picture guarantees nothing — markets do not deal in guarantees — but seeing risk as it really is, and being ready to act before you are forced to, greatly improves the odds.