Week 2 was a makeover of a bar chart showing iPhone units sold since the first iPhone was released in 2007. This week we received over 150 submissions.

In light of the criticism of the data for the week 1 challenge, I decided to state a few observations I made on the iPhone sales visualisations. I hope you will find them useful and they should give a good indication of how I arrived at this week’s list of favourites.

Thumbs up!

• I am very pleased to see so many people sticking with the original dataset and just trying to improve a viz. While it is perfectly okay to add further data to enhance your story, it is not expected and by keeping it simple, newbies especially have a great chance to learn new techniques and skills in Tableau.
• A lot of submissions came only a few hours after we released the data – I love the enthusiasm and I’m glad the simplicity of the dataset seemed to resonate with the community.

‘Problem areas’

• A lot of people didn’t notice that 2007 only included Q3 and Q4 and used the entire year 2007 as a baseline to compare to 2008. Subsequently it looks like in 2008 sales increase by over 700% which is wrong
• There are likely many reasons why sales have progressed the way they did – it is unlikely that we, with the limited data we had, have the single answer to why sales declined in 2016. Let’s be careful about our claims based on the data at hand
• Some data visualisations had great potential for being included in my favourites but let their authors down on details, e.g. not stating their data source, stating incorrect numbers, lots of typos (please note: English is not my first language either, so I feel qualified to be picky: if I can get it right, so can you 😉 ). A polished overall look will give your work much more credibility

Author: Michael Mixon
Link: Tableau Public

What I like:

• Excellent design – my eyes are drawn to the chart in the middle of the page; the design is nice and clean, and focuses on the essential message
• The visualisation reflects the emotional tone indicated in the heading
• It makes me think. This viz doesn’t just tell me the author’s opinion, rather by using a question it encourages me to engage with the data to form my own conclusion
• The line chart is great for showing the time series data
• The annotation supports the message of the headline
• The colours chosen for the line chart correspond almost perfectly with the images of Jobs and Cook. While I believe the colour changes according to the units sold, it also happens to line up closely with the periods of leadership of Jobs (white) and Cook (blue, check the collar of his shirt). Neat!
• The data source is stated, as is the author’s name with a link to his twitter account.

Author: Alex Barday
Link: Tableau Public

What I like:

• The design is clean and consistent across charts
• Using a question as a heading guides me to look for the answer in the data
• The heading works as a colour legend for the CEOs
• Using line charts twice, one focusing on quarters and one on years, helps to strengthen the overall message of the trend in iPhone sales as well as the focus on the different CEOs
• The call-outs add further context without cluttering up the viz
• The images are subtle enough not to distract from the message and fit nicely with the overall look of the viz
• The source and the author’s twitter handle are included

Author: Jay Lewis
Link: Tableau Public

What I like:

• Nice and clean look
• Simple colour scheme, which clearly indicates ‘growth’ versus ‘decline’ and is suitable for viewers with colour blindness
• Chart titles and subtitles add further context to guide the viewers
• Reference lines in the bottom chart reflect the release of new iPhone models and help see changes in the sales pattern more easily as they relate to those releases
• Mark labels in the bar chart provide information and remove the need for displaying an axis. This means the bar chart here can be lined up against the years in the line chart below
• The source and the author’s name are included

Author: Shawn Levin
Link: Tableau Public

What I like:

• Nice and clean look
• Unique design, which is easy to comprehend, as it resembles a stock quote so viewers are likely to know how to read the information
• Simple colour scheme with KPI indicators for ‘growth’ versus ‘decline’
• Clear and concise title
• Subtitle provides additional context
• The source and the author’s name are included