What will happen to data scientists as querying data gets easier? originally appeared on Quora – the place to gain and share knowledge, empowering people to learn from others and better understand the world.
Answer by William Chen, Data Scientist, on Quora:
I don’t think data scientists will be out of a job anytime soon—hopefully at your company data scientists do more than just querying data and doing data pulls for other people!
I welcome the days where querying data gets easier. This movement is already happening, as there are a ton of these tools available already for both data scientists and non-technical people at a company to pull up and visualize data easily in a friendly interface. Here are some examples:
- ImplyData’s Pivot (implydata/pivot) and Airbnb’s Caravel (implydata/pivot) are open source UIs built upon Druid (Interactive Analytics at Scale) or similar databases. These UIs make it easy for anyone at the company (especially non-technical people) to pull up and visualize data.
- There are a ton of paid enterprise software options that do this. Some of them brand themselves as “Next-gen Business Intelligence Tools” and are marketed towards business intelligence / business analyst / analytics teams. See Of the next gen BI tools (Domo, Looker, ThoughtSpot, Clearstory, etc) who has the best chance of gaining meaningful share and unseating legacy incumbents (Business Objects, Cognos, MicroStrategy, etc)?. Tableau is also a big early player in this space that makes visualization and querying easier.
There's still a ton of value that data scientists can provide to companies besides querying data, such as:
- Understanding and analyzing product launches
- Identifing and understanding potential shortcomings of the product
- Helping the company make strategic decisions
- Defining metrics that track company-level, team-level, and product-level goals
- Analyzing surprising changes in metrics
- Enforcing the company follows appropriate statistical procedures when making decisions with data
- Evaluating various machine learning approaches in a product (note that this is more than just comparing which models have better cross-validation performance—ideally this will also combine a deep knowledge of the product that can help the model have impact rather than just accuracy)
All of these tasks help strategic decision-making in the company and are far from being automated. As long as companies continue to want to make intelligent decisions with data, there will always be a need for data scientists. I very much welcome these new and better tools, and will feel more empowered than obsolete with them. I hope that other data scientists feel the same way.
This career will keep transforming in the upcoming years and decades, with data scientists focusing on more high-level tasks because of better tools. I don’t think we’ll be out of a job anytime soon.
This question originally appeared on Quora – the place to gain and share knowledge, empowering people to learn from others and better understand the world. You can follow Quora on Twitter, Facebook, and Google+. More questions: