It really depends on the project! Some are based in the office and others are based on client site (this can sometimes be abroad too). Regardless of where your project is based, within Data Analytics, you will always be doing some sort of analytics!
Yeah. What kind of data play with often, transactional from ERP or emails, documents?
Also forgot to mention that within Electronic Discovery we also have a forensic lab where we store and work with evidence.
I can help out on point 2. We really analyse data from a huge variety of formats. From more structured data types such as transactional/payment datasets to unstructured types such as mobile and social media data to emails.
Beyond SQL and tableau, what kind of skillset do you want to see, Python or R?
Good question Mengxiao. We do use a wide variety of tools. In Data analytics, we do use quite a lot of Python for scripting and data munging as well as other tools such as Graph databases.
With regards to the application however, there is no requirement to have prior knowledge of such tools. You will pick up a lot of this on the job and there will be training provided as well
Great! Does forensic data analytics every year have quotas for colleagues from other Deloitte member firms? For secondment or transfer?
No we don’t have a quota for employees from other firms. The positions are open to anyone!
I mean, currently I'm at a China member firm, in audit innovation and analytics area, would there be "global mobility positions' of forensic data analytics in the UK?
Hi Mengxiao. You would be eligible to apply for the grad scheme regardless of being in Deloitte China. One of our new joiners this year joined us after spending a year in audit in another Deloitte firm overseas.
We are very open to candidates with other experience looking for a change in career to join us.
Is any machine learning, e.g. SVM, decision tree, etc. done in FA? On what kind of data?
Well, the majority of Data Analytics work does not really involve Machine Learning. However there are innovation projects in the department which revolve around this. For example, I've worked on a document classification and topic modelling task - where yes you get to implement said learning models like SVMs, trees etc.
Great! So, as I understand it, forensic investigations usually get data at a more detailed level compared to audit, is that right? Because clients are more willing to share data (existence of fraud or FCPA violation signal) ?
The kind of data that we get would really just be a function of the type of investigation at hand.
2 readers find this useful!
Thank you for your feedback!