How to Approach a Time-Limited Project

by Louise Le

At the Data School, we are constantly challenged and pushed with limited time projects. This is to develop our time management capabilities, to teach us to set realistic goals and to give us the confidence and knowledge that we can do it.

So, what have I learned? In this post, I will go over my top tips on how to approach a time-limited project.

 

 

Step One – Finding your Story

You have been inspired to start a project on a certain topic or you have been assigned a project by your employer. Where do you start? You have to find your story.

If your data has already been given to you, you can begin exploring the data. Identify the interesting parts of your data or find an issue you would like to investigate. Be careful not to spend too long exploring the data and deliberating on a story. Make note of what’s interesting and try to settle quickly on the story you would like to tell. I recommend writing down your story as it will help you keep to the topic and avoid wasting time by straying.

If you are yet to find your data, you will need to think of a rough story and use this to find data that fits. You may need to be flexible with your story as the data available may not exactly suit the story you decided on. Once you find data you like, explore it as described earlier and quickly refine the story you would like to tell.

 

 

Step Two – Planning your Visuals

Once you’ve found your story and you’ve become familiar with your data, start planning your charts. Brainstorm the best ways to display your data and tell your story, maybe try a new technique! Don’t be afraid to look on Google Images or Tableau Public for inspiration. Search for infographics and visualisations on a similar topic to yours. Once you’ve got a bunch of ideas you’re happy with, go through each one and ask yourself:

 

Does this chart help tell my story or answer my question?

 

Pick the chart(s) that you would like to go ahead with. Now, on paper, make a sketch of your dashboard. This can be as rough as sectioning out your dashboard and writing what will go in each section. Personally, I try to put in the little details that I would like to include, like filters and parameter, so I don’t forget them (like in Figure 1).

 

Fig 1. On the left, my ideas and sketch of a dashboard. On the right, my final dashboard (Tableau Public link here)

 

Of course, you don’t have to stick to your sketch later on but it really helps to have a foundation on which to build your vizzes. Alternatively, you could use post-it notes to plan your dashboard. This has the advantage of being able to move post-its around to change your dashboard layout (Figure 2, also thank you Caroline Beavon for this technique!)

 

Fig 2. Sticky note technique demonstrated by DS11’s Andrew (photo courtesy of DS11’s Manuela)

 

 

Step Three – Make your Viz!

Now for the fun part, making your viz! When putting together your viz, if you realise a chart doesn’t work well, think about whether there are any easy fixes to make it clearer. If the chart is not easily salvageable, then scrap the chart and go back to your brainstorming sketches from earlier and pick another that could work better.

 

 

Step Four – Finish your Viz

Don’t get hung up on making your viz perfect because this is impossible. No viz is ever finished. No viz is ever perfect. You could spend forever improving your work and you may never be satisfied. As a perfectionist myself, I know I would spend hours or days on formatting if I were given the time. However, in a project limited by time, you won’t have the luxury of editing until you have reached viz-heaven. It’s okay to have a good viz. Be okay with good. Spending too much time sweating the small stuff with your viz could be the reason you don’t deliver on time. It is far better to meet your deadlines with a good viz than it is to deliver a week late with a slightly better product. Figure 2 is visual representation of what I mean.

 

Fig 3. It is better to meet your deadline than to be late with a slightly better product

 

 

This is a quote that is often said in the Data School:

Perfection is the enemy of good.

 

 

That’s all, folks!

 

Louise

Check me out my blog feedmedata and follow me on @FeedMeData_

4 mins read

Thu 31 Jan 2019