FIRST PERSON

If the CAP fits …

By Michael Mortenson, CAP
M.J.Mortenson@lboro.ac.uk

Personal experience of taking the INFORMS Certified Analytics Professional exam.

Michael Mortenson

The OR Society (Operational Research Society, based in the United Kingdom) has a longstanding interest in analytics and in helping its members engage with the field. As part of this, the Society has been investigating the possibility of providing INFORMS’ Certified Analytics Professional certification to Society members and the wider U.K. operational research community. It was decided that it would be useful to have someone with first-hand experience of the exam, i.e., a guinea pig. This was where I came in.

Having agreed to take the exam, I did the usual thing that any conscientious student does: next to nothing for ages and then panicked at the last minute. In many ways the panic was unnecessary; this is not a “book-smarts” exam but one where prior experience is really the only way to pass. You do need to have significant knowledge of the concepts and methods of analytics, but I would say it’s one-part theory to four-parts practice. Many questions specifically focus on practice-based challenges (stakeholder meetings, problem definition, lack of data and so on), and even where the more technical questions arise, predominantly it is on the basis of selecting the right method to fit the problem or identifying the steps to take when model performance dips. In short, and to INFORMS’ credit, this is not an exam that can be bluffed – you have to walk the walk as well as talk the talk.

It is, however, an exam I would recommend, both to analytics/O.R. professionals seeking to “prove” their practical expertise, and to employers looking for recruits who can genuinely hit the ground running. For the benefit of anyone considering the exam I will give you my perspective on assessing your readiness and how to best prepare.

Many CAP exam questions specifically focus on practice-based challenges.

Many CAP exam questions specifically focus on practice-based challenges.

For recent graduates my honest recommendation would be taking the exam after working on some real-life projects. However, you may be able to lean on a mentor or a more experienced colleague. The handbook gives a reasonably thorough definition of the analytics project lifecycle, so it will give you a good idea of the topics to cover.

Specialist consultants will likely have the necessary depth to cover many of the questions. However, the bigger issue is breadth. There are many debates about what “analytics” actually is, and definitions range from “another name for O.R.” to “a branch of computer science.” This certificate is definitely more at the O.R. end of the spectrum (no bad thing to my mind), and for specialists with a predominantly IT-type background there will be the need to familiarize yourself with the O.R. canon to a fairly extensive level. However, for those from an O.R. background, this alone is unlikely to be enough. You will need a reasonable knowledge of a range of topics, including data warehouses, project management and machine learning.

Ultimately, I think the exam is best suited to analytics professionals working in larger companies or those working on “full-stack”-type projects. Whatever view you have on what analytics is, it is a wide-ranging field and the exam reflects this. For this group, working through the handbook and taking a refresher on anything that looks less familiar is likely to be enough.

Overall, I found the experience nerve wracking (my first exam in more than 15 years) but rewarding, suitably challenging and definitely a source of personal achievement. As to whether it makes a massive difference to my career, well, only time will tell.

Michael Mortenson, CAP, is a doctorate candidate at Loughborough University researching the development of analytics education in U.K. universities. Prior to returning to academia, he worked for several years in marketing and ecommerce in the travel industry and in recent years has worked as a consultant providing digital marketing and analytics services to a variety of companies.