Data Artist & Influencer, Human - Machine mediator, Commercial connect-the-dotter, ML Trainer
There is a great demand for data scientists presenting market dynamics that are favourable for the community. More so than your peers in other professions, you will be able to evaluate a company for what it is able to offer you, rather than solely being the one that is being evaluated. So what should you look for when comparing and evaluating data science roles? Here is a list of some commonly known factors plus some less discussed ones that will help you in your evaluation.
Makes sense right? This is the thing that you will be working with.
This is the input for your increased productivity
and valuable insights output.
Evaluate organisations on the quality of the data they have. How well is it organised and how easily can it be accessed? How trustworthy is it and how much effort is required to prepare, clean and correct it from its current form? Is everything that the organisation wants to measure, currently quantifiable?
Essentially, in your evaluation, you are trying to determine two things. What is the opportunity that exists — which is presented by what can be achieved with the data asset as presently constructed? And, how easily can this be achieved?
So the data is the ingredient, but what can be done with it? How does it translate to actionable opportunities in the real world? What are the problems that can be solved in the space that the organisation competes in given their data assets? Is there a breadth of problems or is the opportunity narrow and deep (or just narrow)?
Would you be working on customer or product related data insights, or can data be used to influence decision making across a gamut of areas within the organisation. Consider what problems you will be working on and have the opportunity to solve and whether they fit into your career goals.
Is the organisation investing in data just because everybody else is or do they truly believe that they can get a competitive advantage through the use of data analytics? How does this look like on the ground? Have they made a commitment to be a data driven organisation and have a certain % of key decisions be made via leveraging data?
Are non data professionals also being challenged to be data literate, so that the conversation can flow freely between technical and non-technical teams — and everyone is incentivised to work towards a common goal (which partly involves the use of data)? Is there a data strategy, a transformation program, or even better C-suite representation in the form of a CDO (or similar)?
Assess whether data is something that the organisation is just doing around the fringes or whether it is a core part of the success that they want to achieve.
This you should be considering regardless of what your professional experience is (i.e. whether you are working in the data space or not), and it will largely be visible when you consider the level of commitment.
You want a culture that is iterative and fast moving (but not too fast) and focused on outcomes. Culture should just hit you in the face and be ubiquitous, and you will quickly and instinctively know whether they work for you or not.
However, one underrated thing that you can consider, as you are trying to get under the hood of an organisation’s culture (as well as it’s commitment to data), surprisingly is…..
One underrated thing that I have come to realise is important to determining how ripe the culture is for data scientists to excel — is to look for how many professionals there are who also work in an iterative manner, but aren’t necessarily data scientists. Most of these roles will be in new professions, tied closely to lean start-up methodology (or similar).
These are a good proxy indicator for how experienced the organisation is in working with teams whose workflows are somewhat similar to yours (i.e. interview end users — gather business requirements — shape expectations — get data — build — test — learn — repeat).
This is important because it means the organisation is well versed in pursuing results without a belief that the outcome or process is predetermined — as may be the case in more deterministic teams.
For you, this means that you will reduce the friction spent organising and explaining the work that you are doing — allowing you to focus on achieving results. This can be a really useful consideration especially when weighing up small or lean teams who may not necessarily have a large amount of data scientists but are still committed to them.
What stack does the organisation use and how does that blend in with your skill set? Will you be able to pick up the workload in your new organisation as if you never left your old role? Better yet, is the organisation tool agnostic, and then if so how much freedom do you have to make decisions and manage your own workflow? If there are considerable gaps in what you currently know, and what you will need to know in your new role, what level of training and onboarding will be provided for you to get up to speed? Additionally, is there a training budget and how is it intertwined with the skills that you will need to acquire to progress.
Best to visualise what success looks like in the medium and long term.
After all, you are looking for a career and not just a role (as much as is possible in 2019). What will your place in the team be, as you continue to progress? What does leadership entail in the data space under the model that the organisation deploys? Are there opportunities for advancement on a purely technical path, or do you need to move into a position of leadership to progress? And if so, what does leadership look like and which skills will you need to retain and which will you need to acquire?
Managing your career is finding the ‘right’ roles and not just ‘good’ roles. Luckily the breadth and number of data science roles is likely to increase, especially with further specialisation in the space and growing investment across industry. So, consider all these things as you do your research, consult your peers, participate in interviews in order to find the best data science role for your situation. Best of luck.
This story could originally be found here