Now three weeks into my time at the Data School (with two of those weeks working heavily with Alteryx), I find myself increasingly enjoying Alteryx as I become (slowly) more comfortable with the software. As someone who has enjoyed puzzles and logic for as long as I can remember, the genuine satisfaction I've been getting from finishing a workflow with the desired output has been a surprising positive to starting a job in such a new field.
My favourite thing we've done during Alteryx sessions is tackling the Alteryx Weekly Challenges set on the community website. To anyone reading who has never done a challenge before, they essentially boil down to being given an input and an output, and being tasked to find out how to get from one to the other. See this example below from Weekly Challenge #189 about the Billboard Top 100 :
Some challenges can feel overwhelming when reading over them for the first time, but I think the above example feels relatively simple to understand. However we don't have some of the fields we need, so what do we do? My best advice would always be to start with a plan. It's so tempting to just dive into something and hope you find something that works straight away or by playing around, but it's easy to get lost along the way. We've all probably done the same thing at some point in a different facet of life, right?
So here's what my plan might look like:
1) Start with the most basic prep. Do my fields have the right data types? And do we have any values in the data that aren't OK or null?
2) We need a decade field. We have a year field so we just need to find a way to to group the relevant years together.
3) We need not only a count of number of top 100 hits by each artist, but the max values of those counts for each decade. How can we get those counts, and what do we need to group to achieve this?
4) Some decades have multiple artists with the same number of top 100 hits in a decade, stopping us from just sampling the top 1 for each decade. How do we fix this? This should be our final step.
There will be multiple ways to achieve all of the above steps, and the best part is none are technically 'more right' than any other. Some might be more efficient, but if you get the output, you get the output! Job done.
And that's all you're getting. I can't just give you the answer to a challenge! If you ever get stuck on a challenge, there's no point in dwelling on it, and there are plenty of others to tackle in the meantime. You'll often find that when you come back to it with a fresh set of eyes, the answer was staring you in the face the whole time!