top of page

3&4: Critique by Design

3&4: Critique by Design

3&4: Critique by Design

Winner - Philips Service Design Challenge 2023

3&4: Critique by Design

Aman Sinha, Telling Stories with Data

September 20, 2023

Step one: Finding a data visualization

For the purpose of this activity, I picked up a data visualization from the 2023 Q2 publication of the PitchBook Venture Monitor. I chose this visualization mainly because it offered me the data set to work with and some opportunities to redesign the visualization while conveying the same story as intended in the original version.


Step two: Critiquing the data visualization

My observations from the original visualization are based on the visualization effectiveness criteria highlighted by Stephen Few. The original visualization ranks best on truthfulness and does well on usefulness, completeness, and engagement criteria. However, the biggest setbacks were seen in perceptibility, intuitiveness, and aesthetics.



The original visualization is useful for the intended audience (VCs, investors, and startups) since it represents quite a broad set of data through just a single visualization. While the visualization presents useful information, it was critical to assess ‘how much information’ was ‘useful’ for the story that was being tried to be conveyed. There were some pockets identified which, while present, were not as useful as others.



Checking for the right information and the right amount of information. The visualization has more than what would have been required to convey the same story. The data sets up the context well - which was to suggest acceleration/deceleration of the VC deal values; however, it could be streamlined by ‘not ingesting data with empty calories’.



The original visualization forces the audience to compare the bar sizes of the ten categories for each year, which is somewhat demanding, difficult, and imprecise. Especially, with the small scale and absence of grid lines, it’s difficult to gauge the difference with similar heighted bars. Perceptibility has to be the major criterion where this visualization falls behind.



The visualization, in itself, is accurate. However, when placing the title/story against the actual numbers, the title seems to be misleading. The major portion of the chart shows acceleration/growth, and deceleration is mostly post-2021 - it might be helpful for the title to mention the exact timeframe it’s referring to as the deceleration period.



It is a familiar type of data visualization. But, the sheer amount of categories and scale of the chart makes it difficult to understand beyond a certain degree of clarity - hence, hampering intuitiveness. This type of chart would have worked well had there been five or fewer categories - resulting in a bigger scale and better distinction.



While the original visualization was not an eyesore at first glance - it definitely was a bit overwhelming. The sheer number of categories depicted makes it look cluttered and chaotic. However, the absence of grid lines makes it visually lighter and easy on the eye but compromises clarity. Maybe, using a very light shade of gray (HEX 0E0E0E) for the gridlines would have solved for both visual appeal and clarity of data. The use of Sans Serif typeface is again a good choice for this visually heavy chart. The typography balances the overall look and feel of the chart. The color palette is not the worst, but a few dark shades like those for Wealthtech and Regtech seem too dark against other shades and gather attention.



From the audience’s perspective, the chart doesn’t seem engaging or inviting. The first glance gives a very cluttered impression and wouldn’t invite a reader to dig deeper unless specifically looking for the information. As a designer, the visualization did gather my attention, but as a contender for a critique and redesign - mainly because of how heavy the chart felt at first glance.

If I were to pick one thing that stood out to me, I would say the extent of information conveyed through a single visualization. It has its downsides, but still - it’s a very good chart type to convey large sets of data involving comparisons and trends. What didn’t work well overall was the link between the title (story) and the chart. If the intent was just to highlight a trend (of deceleration) would you really need to put in the actual data points for each category for six years? Well, this is the question I tried to solve with my redesign.

Step three: Sketching out a solution

Based on the previous step of critiquing, the following are the two initial sketches I did to gather feedback on my thoughts and then build further on these. The idea of the redesign was to convey the same story as intended in the original visualization but in a more comprehensible way.

Feedback on the first sketch led me to the second version. Specific comments/feedback gathered from the people on these two initial sketches are documented in the next section.


Version 1

The following were the objectives for this initial redesign:

  • Since the story was around trends, and not specific numbers for the ten segments within FinTech, I replaced the bar chart with a line chart - as line charts are better comprehensible for trends.

  • Since the title highlighted the two segments that are not decelerating, I highlighted them using vibrant colors and used muted grays for the others since they are just placed for comparison.

  • Tried to highlight the time frame for which the story is being told, i.e. 2022-2023


Version 2

Based on feedback received from three participants, the following changes were made to the initial sketch to get to version 2:


  • One of the common questions was - Why do we even need the rest 8 categories? Can we combine them to just show a generalized combined trend? To solve for this I replaced individual categories with common average values for them and placed them against the two categories highlighted in the story.

  • Rephrased the Title with a more intentional inclusion of '2023 Q2' and actively mentioned the two categories seeing growth.

Step four: Testing the solution

Feedback Session One: The initial sketches were tested with three participants - M(32), F(27), F(24)

Feedback Session Two: Feedback on the final Flourish charts (later in the process)

Script for the feedback session One:

  1. Show the original data visualization and get first impressions. 

  2. Ask questions - for the graphic, what do you see first? What's the first thought you develop as you look at the graphic? What do you like?  What do you dislike? What do you wish you saw? What would you change in this graphic?

  3. Show the solution sketch and get first impressions.

  4. Ask questions - for the graphic, what do you see first? What's the first thought you develop as you look at the graphic? What do you like?  What do you dislike? 

  5. In what ways do you think the sketch solves problems with the original graphic?

  6. What are some changes you would still want in the solution sketch?

Participant One
Student, Asian



  • Don't include individual data for each segment other than Payments and Capital Markets.

  • Rephrase the title to something like - 'Post COVID-19, only these two sectors within FinTech are growing'.

  • The line chart is a good change for the kind of story that is being told.

Participant Two
Student, American



  • The original graphic is super cluttered, with no order.

  • It's good that you have still kept all the data points as in the original graphic (contradicting the feedback from participant one).

  • Introduced to the Flourish's version of a line+bar chart combination.

  • Liked how the original graph conveyed good details of the information.

Participant Three
Student, Asian



  • 'Please get rid of all the gray lines' - either replace them with a single average or choose a very light shade of gray to depict these secondary data points.

  • Good idea to replace a bar graph with a line graph - it's clear why you would do that. 

  • The larger portion of the original graph shows growth - it's not clear which time frame's 'deceleration' is pointed out in the title.

Key takeaways from Feedback Session One:

  1. A line graph is a good choice for the kind of story being told.

  2. Again, for this kind of story, individual data points for all segments are not required - Just highlight the two accelerating segments and club others to form one category (to show a trend).

  3. The title has to be rephrased - make it more specific to the time frame and segments being talked about in the story.

  4. Find a way to highlight the time frame in the Title or chart.

Step five: Building the solution

Based on the feedback from the three participants, the sketches were then translated into final visualizations using Flourish. Flourish, to my excitement offered a version of a line chart that came with a live bar chart as well. This version helped with a real-time variation of VC deal value with time. The first version below was again tested with the same set of participants and feedback was gathered on the overall look and feel.

Based on the final feedback, the revised, final version of the chart was created on Flourish. The final edits were around the Title, color palette, and the duration of the animation.

To sum up

I found this method of evaluating the visualization quite effective. It helped me look at the same visualization through different lenses. Usually, as designers, we get stuck with the ‘aesthetics’ part of the visualization. However, the seven effectiveness criteria allowed me to look beyond just aesthetics piece. I feel the need to include another criterion to the list - Accessibility. Having worked extensively in the realm of digital accessibility and studying WCAG guidelines, accessibility is something that goes unnoticed for data visualizations. Data visualization should be compatible with screen readers and accommodate various forms of color blindness.

This critique and redesign process followed five steps involving two sets of feedback sessions. While each section shows what exactly went into each step, the overall process was a good learning experience which helped me in developing a new approach to critiquing existing data visuals and taking constructive feedback while redesigning the visuals to include multiple perspectives.

bottom of page