JohnI’ve been dealing a lot with the ETL work and building visualizations lately. Regardless of the data source and the destination, there are three things that I make sure I do: define the goal and question everything about it, ensuring I have everything I need to get the work done, and then asking 100 more questions. I hope these tips help you!
Define The Goal and Then Question The Crap Out Of It
With any visualization, it’s obviously important to solidify the end goal and the requirements that surround it. What is equally important is to re-clarify everything about that initial end goal. I say this because there have been numerous times where the first requirements that were set weren’t what was needed at all; half the time it’s a shot in the dark to “get stuff done”. The unsureness can be avoided by questioning the crap out of the end goal. A few questions that come to mind are:
- Do we need to do this?
- How do you want your visualizations to look like?
- What are these visualizations calculating?
- Is this actually important?
- Are there other ways you want to aggregate the data?
- Why do you want to do it this way?
- Have you thought about doing it that way?
Questioning everything and providing different perspectives can be helpful to avoid group-think, encourage new ideas, and helps everyone understand the plan. I’m sure you’ve been there before when people “go with the flow”. It creates boring, redundant processes that hinder innovation because either “this is how we’ve always done it” or “if it ain’t broke, don’t fix it”. A quick reminder that in order to survive in competitive spaces, new ideas are required to pioneer the future. And these new ideas need to be understood by all so there is no confusion.
Another thing – imagine going off the first requirements, spend time completing it all, and you get the “that’s cool, but it’s not what we’re really looking for” statement. Just a real waste of time. So why not clear everything up in the beginning, right?
Ensure You Have or Will Have Everything You Need
Now that you’ve defined the goal and questioned the crap out of it to ensure everything is good to go, you need to take stock of everything you will need to get this done. This could involve data, the human resource(s), the technical tools and software, and the time. This isn’t a comprehensive list, but you get the point. Make sure you have or will have everything you need to get the visualization built. Just iterating over my list: without clean data, it’s garbage in garbage out. If you don’t have the technical and business people to help answer questions or do stuff for you that you can’t do, you’ll have obstacles every step of the way. If you don’t even have the software or tools to ETL the data or pipe it into a program, what are you even doing? And last but not least, if you don’t have the time to do it, how are you going to get it done?
Like I said, I have 4 items on my list as an example, but this can extend to incorporate other things such as understanding the underlying elements of what is in your visualization or even having enough money to get it all done. It’s the little details that can really derail a project.
Ask 100 More Questions
I’ll admit, 100 is a little ambiguous; the point I’m trying to drive here is to keep asking questions about anything and everything so everyone stays aligned. During the process, there can be many obstacles or issues that arise from the work. People can be reassigned, tools can stop working, and priorities are constantly changing. Even if you don’t have a lot of questions, I’d recommend asking the following three:
- Are there any changes to anything since we last met?
- Does anyone need help with anything?
- Are we still on track from a time perspective?
The three tips I wrote about seem obvious, but when timeframes shrink and resources are stretched thin, it’s important to have a clear goal and have everyone on the same page.
