I personally love ETL and agree with Andy, it can be a full-time role that someone who has the right mentality, should work solely as the ETL developer for organizations. In my organization, which is where I support a full team of data scientist, it certainly is a full-time role to help them with ETL process that can be automated.

When I worked in the video game industry for many years, my opinion on thinkers versus doers has become skewed to match how they view them. This is where thinkers, people who do nothing but want to be the idea person for a game, was a dime-a-dozen. The industry really had no room for people who just wanted to do nothing but spit out ideas. Everyone had ideas, but very few people could actually act on those ideas and make them into reality.

The reason it's different for data is because you don't have a lot of people wanting to develop data in comparison to people who want to develop the hot new video game. You also don't have a lot of data people who want to just be the thinker, most of them want to be the doers. Where in the video game industry, you could totally sway most of the doers to just sit and talk about video games all day without actually making a video game for the business.

But seeing that, I think it's important to get the doers to be the thinkers too. That means, hiring people who work with the data to be excited about thinking up ideas for the data too. Not just hiring a person who is working in data for the money or to do just one specific thing with one specific tech.