Dashboard Week Day 4

by Vikash Bhardwaj

I’m nearing the end of Dashboard Week at the Data School.

For those who don’t know what this is; it’s an intense week with daily project which needs to be ready to present by 3.30pm. During this week time NEEDS to be managed properly. 

After a couple days producing dashboards I wanted to take a more hands-on approach to understand how I manage my time and decided today would be the day to do it.

I have another post about time management here – where I discuss how I came up with the framework to work from.

Today’s project is about looking at the data within a research document, assessing what the data says and comparing it to the conclusions of the paper. 

After familiarizing myself with the task and a quick call from Robbin checking in on how I was doing I set out a plan for the day (it was 9.30 at this point)

  • 1h; 9.30-10.30 – understand the project, research into insights etc
  • 1.5h 10.30-12.00 – build out all chart types in Tableau (without being tempted to create a dashboard)
  • 30mins leeway
  • 12.00-12.30 Start planning the dashboard – pull together.
  • 30 mins 12.30-1.00 LUNCH
  • 1.00 – 2.00 continue to refine the dashboard 

My plan is to leave 1.5h as leeway (as I’m sure I’ll run into problems and need to spend time elsewhere. 

In addition ’m setting a timer for every hour and plan to write an update answering the following 3 questions. Or whenever I felt it would be useful.

  • What worked well (abbreviated in the day to WWW)
  • What did not work so well (abbreviated in the day to WDWSW)
    • Why?

So, let’s go into it and see how the day went.

This is the unedited version of the notes I took on the day (welcome to my mind)

9.45 Making a start

Going through the journal 

  • The main focus of the article is to highlight the change of number of bugs with climate change

It may be worth looking at climate change related data too – plotting it against the decline in bugs

  • Local insect availability increases with temperature and decreases with wind speed, cloud cover, and precipitation.

What the article is essentially about;

“Flying insects were sampled as the number of insects killed on the windscreen of a car when driven at a fixed speed of 60 km/hr on a specific road (transect) of 1.2 km at Kraghede, Denmark and a second road of 25 km across the study area at Pandrup, Denmark (Møller, 2013; Appendix S1).”

Moller (2019) Parallel declines in abundance of insects and insectivorous birds in Denmark over 22 years

10.00

– have read through all of the journals and made notes accordingly. Starting to think about which direction I’d like to take the analysis in. 

  • WWW – reading through calmly and making notes has really helped with context.
  • WDWSW – Nothing as of yet. I’m still in the very early stages of the project.

Plan update – between 10am – 10.30am 

Research into insights and put the data into Tableau to see if it all makes sense. At first glance a few of the columns (Date and time) are not in the best format for analysis.

We have three sheets of data;

  • S1 – Shows car make and model (the article says the car doesn’t have any statistical significance on the results so shall ignore this)
  • S2 – Date, time and number of insects on screen – key for this project
  • S3 – Abundance of birds over time (may be useful)

I’ve found the location of the study from the study; Kraghede (57°12′N, 10°00′E) and added it to the dataset. I’ve done this as I love maps.

Also found pictures of the three birds and added them to the shapes folder in Tableau. Thinking back I’m not sure if it is relevant / useful but have done it now and the birds all look cool.

I’ve started to look for additional supplementary data but think it may be a waste of time (as precipitation, cloud cover and wind is included in the data set) 

I am thinking of other factors however – and shall look for the next 5 mins for data.

10.30

WWW- Feel like I have a good grasp of the task and data. Feels unusual that I haven’t opened Tableau yet but I feel ok about it. This is the most I have ever stuck to a schedule and I feel keeping an eye of the time is working wonders for managing personal expectations. 

Still tempted to add new data BUT realise it won’t change the analysis a great deal or add more value to the project than the hassle is worth. 

I have an idea of what the dashboard will look like (long and scrolley with a fair amount of text and large numbers for context. 

WWW – keeping an eye on time and avoiding going down unnecessary rabbit holes

WDWSW – nothing as yet. We may be entering uncharted productivity territories 

POA (plan of action) – Sketch out the dashboard to 11.00 / 11.15 (I’m running behind my planned schedule but it all seems under control (…for now)

10.40

Its 10.40 now, will re-read Andys brief and skim the Guardian article and get back to work at 10.45. 

10.43

Re-read the briefs comment;

“It would be great if the team can verify and visualize the findings of the research paper and even find their own insights/correlations if they exist. And please make the dashboard visually appealing.”

Andy Kriebel (2020) Dashboard Week Day 4: Car ‘splatometer’
  • I feel like it was an excellent decision to re-read the brief before planning.
  • Note to self: I may not be the best at following instructions.
  • The two highlighted lines give focus to what should be looked at and the sort of questions that need to be answered.

11.00

I’m hungry. I think it’s time for a quick break before planning.

11.20

Nutrition confirmed.

Well I guess break’s take time.

A simple snack and sending off some work messages took 20 mins. Now back to the grind after hearing about a problem with the data down the whatsapp grapevine..

Also, Need to clean my desktop (the tabs and open folders are becoming overpowering)

11.30 

It took me 10 minutes to close all the random tabs and windows that have accumulated on my desktop.

However, organized desktop. Confirmed.

I plan on putting the data into tableau and having a little play around for 15 mins (then will plan the dashboard).

12.00 

Upon loading into tableau a few issues arose with the data. Currently in alteryx cleaning up the data times to create a format which will work in Tableau. Feel like I may have underestimated the data somewhat. 3.5h to deadline

WWW – not too much since the last update. I’m happy I’ve caught the data issues now though – would be a whirlwind panic otherwise methinks

12.25

Managed to quickly sort things out in Alteryx and joined the file I forgot to do beforehand. Glad I added extra time in as that took nearly half an hour to sort out. Now to put it into tableau and see if it worked.

12.27

Completing the calculations in Tableau. Ahh Tableau

12.52

Time fields seem good – seems like the amount of time spent was a bit ridiculous. 

  • However would like to have a look a the variance over time of day
  • Continue to explore the data..

13.00

I’m hungry. It’s time for lunch.

I’ll create a list of what needs to be done when I get back to my desk 

  • Re-read Andys brief (only takes 30s)
  • Have a look at the chart types in the document
  • Scim through Kents document to get ideas
  • Plan and action the story
    • This will be exploratory as opposed to anything else so NO THINKING ABOUT SET / PARAMETER actions for now – you know what happened yesterday.. 

Its exactly 1pm now so will be taking lunch and should have 2.5 hours left.

13.45

Guess whose back. Lunch was cool. I ate some chicken and went for a quick walk around my flat. Back to work starting with the list I made before lunch.

Chart types in the document that i’d like to cover;

  • Residual insects over time
  • Number of insects against
    • Time of day
    • Temp
    • (could do precipitation)
  • Bird type against number of insects (one for each bird type)

Looking at one of Kent’s wildlife trust’s “Insect declines and why they matter” document I really like these;

I’m going to use the orange pages for ban inspiration (#ban-inspo)

13.55

List I wanted to look at is reviewed and I’m moving onto paper to draw out my dashboard (in the knowledge that hopefully the data is fit for use)

14.00

Im going to look ask questions and create charts in according to answer those questions

I’ll start by replicating the charts in the journal. 

Ergo think Im going to split the dashboard into 2 sections.

  • Reviewing the windscreen insects
  • Looking at the bird information 

14.15

– struggling to plot the data for even the simple charts. This is not ideal. My issue is to standardize the analysis with the date-stamped values I have

14.30

Time is the enemy. I miss Power BI

14.40

dashboard planned with stuff and things to build

14.45

Created the bans – time to focus on the graphs. Note to future self – go over table calcs

15.15

Not looking great – however everything on the dashboard. Filling in the text needs to be done (or does it…). Anyway I’m running seriously low on time

15.30

Just about put something together to present + managed to include the small pictures of the birds. My hope that the cute lil birds distract from the rest of the dashboard.

Feedback:

The dashboard was well received. Seems like somewhat of a successful day.

So;

I finished in the nick of time with something I didn’t really plan out at the start which wasn’t my plan.

Ultimatley what happened?

  • Issues that arose took way longer than expected (especially the date/time conversion)
  • I started on Tableau later in the day than expected
  • I still plan and expect too much in the timeframe

However, I’m happy with my final output and proud that even with the setbacks and even though I was unable to showcase the functionality I would have liked to I completed and presented on time (I’ve definitely streamlined my last minute decision making procedure).

I didn’t look at my original plan when I was in flow or even answer the three questions which I wanted to answer every hour.. I think the time pressure came in post lunch and, to be fair, I didn’t really have time to sit back and be reflective. 

On a positive note, I feel like the tasks this week have been a test of managing stress and pressure (or creating these conditions to get used to in case any projects have a schedule as such) ; and I’m getting much better at doing so.

Time tracking today was aslo kind of fun as I can see what I was thinking throughout the day (apart from towards the end – when you can see the posts become much more brief) . I’m sure I’ll look back at this post, and my boggled state,  and find it amusing (or more likely embarrassing). 

So, from after receiving the brief at 9am, going through the research paper, assessing the data, this is what I came up for the 3.30pm deadline:

Here’s the link to the dashboard on Tableau Public;

If you have any comments, queries or questions please drop me a line on Twitter @vikb03 or Linkedin.

Peace out you cool cats and kittens!

Thanks for reading!