5 Tips to Deliver Analytical Solutions in an Unknown Industry

Image Source

Imagine going to a boutique store. You get a customized dress that fits you perfectly. Now, imagine going to a hospital and getting the exact same choice. Only here, the purchase would be for a cancer treatment that best fits your genes. Yes, a treatment that is modified to match your genetic pattern. The difference being, instead of 3 sizes — Small, Medium, and Large, there are 3 billion nucleotides in your body to choose from. Welcome to the world of precision medicine!

As a part of the UC Davis M.S. Business Analytics (MSBA) cohort, I was chosen to work with the reputed Sutter Health Organization on their latest Cancer Avatar Research Project. The student team is expected to work with scientists and physicians to apply data analytics to improve Sutter Health’s precision-medicine study and to develop a pharmacogenomics platform.

No, I haven’t misspelt any words and no, I did not have any previous experience or understanding of these terms. But that’s the reality. Working for a field that is not specific to any industry, you can’t be an expert or know all the tricks of the trade in each industry. Not having enough domain knowledge can seem crippling for any newcomer. However, that doesn’t mean you can’t learn as you go like I am doing (see what I did in the first line above?) If you are determined, know where to look, whom to talk to, and what sources to trust, you’ll find it much easier to become an expert in an industry you know nothing about.

Here are 5 tricks to help you become a real authority while providing analytical solutions in an unknown trade:

1. Asking the “Why”

“If I had an hour to solve a problem, I would spend the first 55 minutes determining the proper question to ask, for once I know the proper question, I could solve the problem in less than five minutes.” — Albert Einstein

Image Source

Though every member of any firm must know the reason for their work, asking the “why” is much more important when you are new to the industry. There are numerous cases of Data scientists spending time and effort to arrive at conclusions that were counterintuitive to the business. Asking why eliminates confusion caused by preconceived assumptions, due to lack of knowledge, or more dangerously, partial knowledge.

If you spend time understanding the reason for the problem statement, you can apply this knowledge to arrive at a solution that is more significant and empathetic to the consumers. Give users what they need, not just what they demand.

2. Collaboration

Collaborating with teammates is another integral part of succeeding as a data scientist. It is only when the analytics team merges their technical skills with business initiatives that they can arrive at meaningful solutions or insights. This requires collaborating with cross-functional stakeholders across the organization instead of working in a silo. Reaching out to people bridges gaps in your understanding of the industry and gaps in the executives' understanding of statistical outputs.

Image Source

1. Collaborate with the client: Once you read through the organization’s content, you must feel confident to ask them specific questions. You may be hired to solve one of their problems or enhance their efforts, but when it comes to the industry as a whole, they are the pros. Asking them critical questions allows for early clarifications and proves that you are spending time on research.
2. Know whom to talk to: Once you have identified the questions, you must also identify the people who are genuine and willing to help you. Experts enjoy sharing their knowledge and want people to use their advice to succeed. Having an insider is great to know the pulse of the organization and get you an ally during tough situations. It is a great practice to know who are the influencers or industry pros in any space. Reading their blogs or following their social handles will give you great, first-hand experiences that they’ve shared and could be used in your analysis.

3. Avoiding rookie mistakes

A few mistakes that most data scientists in new industries make could cause them valuable resources like time, energy as well as motivation.

1. Spending a lot of time on theory: It is a slow and daunting process to learn about any new industry and it could leave you overwhelmed if there's too much to learn. It is possible more often than not, that you may not even retain the entire information and may miss out on the important parts. To avoid such mistakes, learn to be comfortable with partial data and how to gradually fill in the gaps as you progress.
2. Coding from scratch: A common error data scientists make is to code everything from scratch. While it’s nice to implement new changes, the reality is that algorithms are fast becoming mature commodities. Most practitioners never code algorithms from scratch, and it is more important to understand how to apply the right algorithms in the right settings. This is developed by asking the why for every problem statement.
3. Wanting to jump into the deep end: As a newcomer, it may be natural to showcase your statistical competencies by wanting to use advanced techniques. However, it is important to master the fundamentals that could be applied to achieve low hanging fruits. A classical machine learning algorithm once verified, can be enhanced later.

4. Education and Research

Strong data scientists continue educating themselves to stay up to date on the latest trends and developments. Though a simple awareness of these skills doesn’t necessarily guarantee success, some of these top data science skills could help to ease your transition into a new industry:

1. SQL — Organizations use SQL to manage and store data. Knowing how to use this tool will allow you to gain access to client data which is the first step towards solving a problem.
2. Statistical Programming — Using R, Python or SAS would allow you to perform advanced analyses on the data much faster than using the plain old excel. They allow you to process vast amounts of data and come with inbuilt stats functions.
3. Data Visualization — If your findings can’t be easily and quickly identified, then you’re going to have a difficult time getting through to others. Tools like Tableau, Looker, and Power BI allow you to showcase your findings to a wider, non-technical audience.
4. Research — A quick Google search is helpful when you are trying to get an overview of topics or industry news, but in the business world it is important to know which source is reliable. Use your clients to conduct primary research — ask them for the resources they use to pull their data from and if possible, ask them to share access to their firm’s internal system so you can get the inside scoop. For secondary research — utilize industry accepted tools like Gartner (for company and industry-wise topic research and to boost your vocabulary), Factiva (Gives company data like annual reports and latest news), Statista (Provides data trends and outlooks across industries, countries, and years), Nexis Uni (Shows news related to industry and companies. It also uses sentiment analysis to display user reaction to new topics). Also, take advantage of industry blogs and trending topics.

Good data scientists develop their skills by learning and implementing. Once you have used courses, boot camps, or other mediums to learn, you must try and apply this knowledge to gain expertise. While technical knowledge is not the only thing data scientists should be focused on, these skills are intrinsic to the job position.

5. Don’t let perfect get in the way of good

Image Source

It is rare to come across a data set that can be used as-is to conduct analysis. Data is rapidly evolving and what you thought was meaningful last year could be obsolete in the next. A good analyst knows how to transform a data set to solve critical problems. Work to create a mindset that makes “sufficient” rather than “perfect,” acceptable. This will make you more nimble and better able to respond to new industries and trends in data.

To conclude…

Working with different industries and individuals is not the easiest task. It is earned over time through hard work, persistence, and a go-getter attitude. These tips have helped me navigate the hitherto unknown world of healthcare providers and cancer research. Hope this article benefits data analysts or anyone really who is planning to venture out into the unknown!