This was a very interesting assignment. Less so because of its technical challenges but more because of how lazily it has been used as a screening process after the initial HR round for any applicant - be it in the operations domain, data science, or Product Management. I later found out on Glassdoor that even the yardstick for evaluation is the same and Revolut isn't the only one. Famed German counterpart in the Neobanking space, N26 has used the same (not similar, exactly same) case study challenge for their recruitment. I always thought they have similar product offerings, but hey, they have taken the imitation game to a whole new level. Sigh!

Anyhoo, we are here for the good stuff (as the great Agadmator says when presenting the Morphy Saga). You're getting double bang for the buck. Rejoice! 😃

Situation

As a financial institution regulated by the FCA, Revolut has the obligation to verify the identity of all customers who want to open a Revolut account. Each prospective customer has to go through a Know Your Customer (KYC) process by submitting a government-issued photo ID and a facial picture of themself to our partner, Veritas.

Veritas then would perform 2 checks:

The customer will ‘pass’ the KYC process and get aboard if the results of both Document and Facial Similarity checks are ‘clear’. If the result of any check is not ‘clear’, the customer has to submit all the photos again.

The pass rate is defined as the number of customers who pass both the KYC process divided by the number of customers who attempt the process. Each customer has up to 2 attempts. The pass rate has decreased substantially for the last few weeks.

Data

Document Check Dataset

Facial Similarity Check Dataset

Veritas API documentation

Task

Basis the data set provided, explore possible reasons for this dip and suggest solutions.

Solution

In hindsight, I could have structured the report even better. And dug a little more but to be honest, when I discovered the shenanigans around this challenge, my motivation had taken quite a hit.

https://s3-us-west-2.amazonaws.com/secure.notion-static.com/8ea5daf8-97dd-40c8-a465-3d144b327589/Revolut_Analytical_Task.pdf