Hero – Starting Your Data Science Career

Starting Your Risk Strategy Consulting Career at Featurespace

by Andrew Cronie     8 min read  | July 3, 2023

At Featurespace, we strive to be the world’s best software company at protecting our clients and their customers from fraud and financial crime. Our Risk Strategy Consultant role at Featurespace places you at the very heart of this effort, where you will play a key role in designing and building solutions to detect and prevent fraudulent behaviour experienced by our customers.

What is a Risk Strategy Consultant?

The Risk Strategy Consultant (RSC) team consists of fraud analytics experts, with years of fraud prevention experience across various payment types. They regularly meet with customers to discuss fraud prevention strategies, capturing requirements to then translate into strategies within the ARIC Risk Hub platform. During a project delivery, an RSC will work directly with the customer to understand their process, dataflow and infrastructure setup so that they can deliver the optimal solution.


Andrew Cronie –

Senior Risk Strategy Consultant

An RSC will typically have had extensive, hands-on experience in working within a fraud strategy department. Often this will be in a Plastics, Payments or AML team.

An RSC will be an expert in the ARIC internal rules language AMDL, with a deep and technical understanding of how rules are built, and the efficiency of their strategy.

But more than anything else, an RSC is curious! Curious about:

  • What a customer’s fraud prevention strategy is at present
  • How it proposes to evolve
  • What gaps are present
  • How robust the solution is overall

This is a key quality which separates the best analytics consultants from the rest. The drive to find a solution that provides ongoing, long-term value rather than a solution which simply plugs a gap.

What You Need to Know

Interested in a career in fraud analytics? Well, the great news is that there are multiple ways to get started.

Many analysts begin their careers within the Fraud Operations department, handling calls with customers. This role trains you to recognise suspicious and non-suspicious patterns within data, keeping you aware of the latest trends affecting customers. This gives you front-line feedback on the success of the fraud analytics teams’ efforts. From here you build fantastic fraud subject matter expertise, which you can then build upon in a fraud analytics role.

Those with a numerate degree are often attractive candidates for fraud analytics roles, as are those who have previously worked as another type of analyst (Business, Credit Risk, etc.).

Additionally, if you have coding experience (e.g. SQL, SAS, Python) then being able to apply those skills to customer data to pick out patterns is a huge advantage, as these are very much the tools fraud analysts use in their roles.

Once working in a fraud analytics role, you’ll typically be analysing data and fraud rule performance with the aim of reducing losses for the company. You will be striving to strike the balance between preventing as much fraudulent activity as you can whilst at the same time, avoiding as much genuine activity in these strategies as possible. My recommendation here is to keep challenging yourself and your teammates to write the best strategies possible.

On this point, here are some thoughts I use when assessing a strategy:

  • What is the False Positive Ratio of this rule?
    • The lower the better- i.e. capturing as many fraud cases and as few genuine cases as possible
  • What is the estimated Prevented Fraud Value of this rule?
    • If this were in place last week/month, what fraud value would this have prevented?
    • Is this a strategy that is targeting the most pressing threat?
  • What is the Genuine Affected Rate of this rule?
    • What proportion of the genuine customer base will this be affected?
    • This is often more useful when considering a group of rules or an overall rulebase, so that you aren’t preventing too large a proportion of your customers from regular activity.
  • What is the Account Detection Rate of this rule?
    • When considering this fraud threat, what proportion of fraudulent accounts would this strategy have triggered against?
    • This is a useful metric to understand how successful this strategy is with regard to the overall fraud vector, on a per-account/customer basis. E.g. 40% of accounts were alerted on.
  • What is the Value Detection Rate of this rule?
    • When considering this fraud threat, what proportion of fraudulent value would this strategy have triggered against?
    • This is a useful metric to understand how successful this strategy is with regard to the overall fraud vector, from a monetary perspective. E.g. 65% of the total fraud losses would have alerted.

Using these metrics on a per-rule basis and on a full ruleset view arms you with the information you need to best assess where strengths and weaknesses are.

You will typically conduct some of your analysis in Excel and so proficiency in this is very much expected. When it comes to the larger pieces of work, Excel is not so efficient. And so being fluent in SQL, SAS, R, or Python allows you to tackle more complicated requests much more easily. As such, though you will likely receive training in one of these whilst performing an analytical role, taking some time to study this in your personal time will set you at an advantage. There are many resources, both paid and free, and at all levels of competency in these languages online.

A good way to challenge yourself is to combine this with another interest – perhaps you are interested in a certain sport, or politics, or something similar where you can find large datasets for free online. Having a personal project where you put these analytics skills to the test can have many benefits, but something of particular note is that it provides you with an opportunity to evidence your skills. Many people consider themselves proficient in something, but this is a way to share and prove to potential employers that you are competent in what you say on your application.

On that last point, though getting a GitHub is far from common – this is a perfect window to showcase your skills.

Writing Your CV

There are two key areas to comment on here which are:

  • The analytical tools you are comfortable and proficient in
  • The relevant fraud or analytics experience you have had in current/previous roles.

As mentioned earlier, much of the technical aspect of these roles utilises a coding language (SQL, etc.) and so highlighting your expertise here is important. Drawing out examples where you have used this to create X or built Y lets people know what you have accomplished previously, and sets up opportunities in the interview to expand on these and give the full details. There may be a technical stage to an interview for an RSC role, but considering Fraud Analytics roles more generally, there often is not a stage where you can really show this off to your potential employer. As a result, showing off the best examples where you have deployed these skills to their best effects in your current/previous roles on your CV is critical for employers to put you forward to the next stage.

Your industry experience is key to highlight here. Drawing out the key achievements at each stage gives you more opportunities to stand out.

Have you worked in teams that look after:

  • Cards?
    • Credit and/or Debit?
  • Payments?
  • AML?
  • Digital?
  • Cryptocurrency?

Have you had experience with industry events, such as UK Finance meetings or industry roundtables? These are excellent ways to show that you have a broad understanding of not just how fraud affects the company you work for, but across the industry too.

This is a very collaborative industry and a surprisingly small world at times. It is one of the few areas where banks and payment providers work together, instead of competing with one another. Your work makes a real difference, preventing money landing in the pockets of criminals and gangs. You protect genuine customers from having their savings taken from them, and all the stress and upset that can cause.