Dashboards, Scorecards & Visualization

Marketing Metrics and Key Performance Indicators (KPIs) are measurable values used by marketing teams to demonstrate the effectiveness of campaigns across all marketing channels. Social media is one of the many channels that marketing team widely use and keep track of. Here’s an example how to use KPIs to measure the performance on twitter.

The dashboard above first offer a glance of the total number of current followers and how far it is to the target number. Then it lists some key metrics in past 30 days and the trend of them (increased or decreased). The right side of the dashboard shows the trend of visits in past 30 days, which offers real-time monitor to marketing employees to see how things are going. Therefore, visualization is a powerful tool when you understand the business, pick the appropriate metrics and use the right way to present it.

In last exercise, we introduced interactivity in our visualization by creating parameters, applying filters, creating sets, calculated fields etc. We had different approaches to make it interactive in what sense. I found this article and would like to share some useful ways in which interactions can be used.

1. Highlighting and Details on demand: Making use of highlights helps user to focus on important part of the visualization. Instead of including all information at once you can allow audience to choose and get details of their interest.

2. User-driven Content Selection: With interactive visualization you give user an ability to change the content and drill down to grab relative information. Such a configurable visualization becomes the template through which different structurally similar data sets are displayed, and additional controls allow the user to change what data gets displayed. When used in such a manner, an interactive visualization can make a much larger data set accessible than a comparable static graphic.

3. Multiple Coordinated Visualization: When you use single graphical representation, it limits number of dimensions. For example, maps emphasize geographic location and timelines the flow of time. Those commonly used representations also often have well-known interactions such as pan and zoom for maps. By assembling multiple standard parts and coordinating them, you can show different aspects of the data set at the same time. Also using appropriate filters user can understand relationships among the data.

4. User-driven Visual Mapping Changes: You can improve the interactivity by showing data in different ways. Allowing the user to reconfigure the mappings from data to visual form (visual mappings) for a fixed visualization type is an alternative that can help in maximizing the visualization size.

5. Integrating User’s Viewpoint and Opinions: In interactive visualizations, you can allow users to enter their opinions and improve their satisfaction with the visual.

Please visit the following website and get better understanding with the given examples.

Source: http://www.scribblelive.com/blog/2012/08/06/interaction-design-for-data-visualizations/

To study importance of “right scale” let’s see the following graph which is from popular currency exchange website.

Source: www.xe.com

Now suppose you want to know the actual numeric value of right most point, we can see it is little less than half most point between 1.25 and 1.4  (i.e. little half of 1.4-1.25= 0.15), so about 0.6, now adding this to 1.25 it becomes 1.31.

The point I want to convey here is with wrong scaling techniques, it requires more of a mental work than one should actually perform to gain insights from the visualization.  One common source of this problem is algorithm used by common graph rendering software to create these scales. As a designer, one should be aware of this common problem and should consider the following points so that it is easy to perceive values from the graph.

1. All intervals should be equal: This means that the quantitative distance between 2 labels should be equal because if intervals are not equal, it becomes difficult to perceive the values in the graph.
2. Scale should be power of 10 or power of 10 multiplied by 2 or 5: Power of 10 include 10 itself, 10 multiplied by itself any number of times (10*10 or 10*10*10) or 10 divided by itself any number of time (10/10 = 1, 10/100 = 0.1 etc).
Also, it is important to note that 10 multiplied by 2 or 5 is not a constraint in cases where audience thinks of the measure as occurring in groups of any particular size. For example, months (3 or 12), RAM in Gigabytes (4 or 16) etc. A scale of month in form of (0, 5, 10, 15, 20..) is less cognitively fluent than the scale (0, 3, 6, 9, 12..)
3. Scale should be anchored to zero: This does not mean that scale should include zero, instead it means that if scale was to be extended to zero, it should have one of the labels as zero. For instance if we were supposed to extend the above graph the scales in decreasing order would be (0.80, 0.65……….0.20, 0.05, -0.10, -0.25) i.e. this scale has no place for ‘zero’ label hence it is an example of bad scaling.
4. Number of intervals: There is no general rule for this but the scale should provide as many intervals needed for the precision that audience requires but not so many that the scale gets cluttered.
5. Upper and lower bounds of the scale: The general rule is that the scale should extend as little as possible above the highest value and below the lowest value while still respecting the first 3 constraints defined above.
Exceptions to rule 5: a)When using bars, the scale must always include zero, even if it results to an extended scale. b)If zero is within 2 intervals in the data, the scale should include zero.

So next time, it is better to evaluate your scale on these five points before finalizing your graph.
Caution: Above rules apply to only linear quantitative scales.

References:
http://www.perceptualedge.com/blog/?p=2378
http://www.xe.com/currencycharts/?from=USD&to=CAD&view=1D

Organic Valley is one of the nation’s leading organic farm cooperatives, which not only provides milk to wholesale such as Wholefoods, Trader Joe’s, and Costco but also produces milk related product. It faces several challenges.

• Where do them put the milk?
• What do they do with the excesses of milk?
• How fresh is it?

Organic Valley applies SAP Lumira data visualization software to get insight of their business performance and unveil hidden areas of opportunity. The profits visualization tool brings are:

• Milk comes out of cow at 4% butterfat and skimming yields skim (1% and 2% milk). Most of the profit is in whipping cream, half and half, and butter. Organic Valley got the idea how to better use of their raw elements.
• Define most profitable customers, satisfy their need, and prevent their purchase products shortage. Visualization reveals hidden data and gives them a big picture, which shows various dimensions and highlights the group of customers they want to reach.

Visualization tools give Organic Valley a new way of thinking. First, visualization tools facilitate communication. Veterinary farm experts and dairy supply experts can give an effective showcase to the senior leadership and help them executives make data-driven decision. Moreover, Organic Valley also applied BI visualization in its IT department to examine and allocate its spending such as telecommunications and digital data communications overtime.

Reference:

http://searchsap.techtarget.com/feature/Organic-Valley-milks-insights-with-SAP-data-visualization-tool

http://searchsap.techtarget.com/feature/Give-SAP-Lumira-data-visualization-software-a-good-look-says-expert

 

Many Silicon valley companies opposed Trump’s immigration order immediately a week after it was implemented because the US innovation economy relies heavily on foreign talent. Also, it make it more difficult for US companies to continue with their day to day business operations and to recruit, hire and retain the world’s best employees.

Main Intention:-   So far since Trump was elected, four bills related to reforming the visa programs have been presented to the Congress. This has caused a huge concern among the Tech community since they feel these bills could be a stepping stone to a huge change in the entire legal immigration system. The four bills are:

Understanding the viz :- The below visualization is first broken by Visa holders and Green card holders and further partitioned by the type of the worker visa. Let’s direct our attention to the worker visa section which has a total of 807,212 visas. Workers with these types of visas take a lower salary in comparison with American workers which thereby implies drop in American worker wages.

One of the above bills proposes to raise the salary cap for obtaining a H1-B visa to 100k per year. Another bill proposes to curb the replacement of American skilled workers with cheap H1-B or L-1 workers.

As we can clearly see from the below plot on green cards, only 14% is employment based whereas a huge 33% belongs to spouses or children of green card holders working in US. These green card holders could potentially compete for US citizen jobs.

Critiquing the visualization:-  

• For the bigger box chart on Visas, a better representation could be to use percentage instead of total number of visa holders of each type.
• Also, when the main focus is on employment visas it can be highlighted better with a different color that catches the attention of the audience (“Eye beats memory”).
• The segmentation of the different types under employment visa again can be shown more clearly using a better tool tip.

KPI :–

• Number of visa issued in 2015
• Number of green card holders in 2015

Argument:-

Why does US need these foreign workers?

They are talented, skilled and top class competitive workers. Their expertise and labor is beneficial for the economy which raises the standard of living for Americans. This helps US compete more effectively on a global scale.

Reference: https://www.washingtonpost.com/graphics/national/visas-impact/

 

While creating visualizations we need to understand what is the purpose we are looking to get from the data. Usually the visualizations are based on the data that is being considered. The types of viz. are:

• Conceptual visualizations focus on ideas and the goal is to simplify or teach
• Data Driven visualizations focus on statistics and the goal is to inform or enlighten.

When I was working on my redesign project I understood the need to approach the data in a specific way and question if the information is conceptual or data driven. When I started out the question was whether I am declaring something or exploring something? With the declaring part, we usually focus on designing and documenting information to affirm something. With the exploratory part, we must focus on prototyping, iterating and interacting to confirm something.

But with today’s technology development it has become very easy to make visualizations that look good. The only drawback is that is boosts our impulse to “click and visualize” without thinking about the purpose or goals. This leads to charts that are inadequate or ineffective. And the path usually leads to visual discovery with no idea that what you are showing in a chart.

Visual Discovery is a complicated and as bigger and more complex the data gets, and the less you know going in, the more open-ended the work. And this was my journey to get to the visualization in my redesign project.

 

https://hbr.org/2016/06/visualizations-that-really-work

The typical old-school measures of writing cocktail recipes such as stating different base alcohol in ml and oz, can be teasing at the best.

This data visualization I found online is super interesting and creative. First, every recipe is presented in the forms of image, from the percentage to every different base alcohol you may use. Compare to the old way of listing different stuff in accurate amount, this kind of graph visualizations makes the recipes more obvious and intuitive.

http://www.informationisbeautiful.net/visualizations/cocktails-interactive/

The most important thing is that the interactivity:

As you can see, if you click the different base alcohol tab above, such as Champagne or Gin, drinks using such alcohol will be displayed, which means that you can use it as you guidebook of making your own cocktail parties!

reference: http://www.informationisbeautiful.net/visualizations/cocktails-interactive/

Eye beats memory. Visual Analysis. Visual Exploration.

These are a few keywords and jargon that we have been using and dealing with in our course. This week I came across a very interesting topic, the use of visual imagery in charts. The debate was, is visual imagery treated as junk or is it useful in charts?

The main question is, why do we visualize data? Why do we create visualizations? To convey a few facts, or to represent what is happening? To depict the data visually. But another question is, whom are we visualizing the specific data-set for? Is it for a niche audience, for an article that targets the general market or for some sales team to make some decisions?

When it comes to making visualizations for the general people, the concept of visual imagery kicks in. Consider for an example, a chart depicting the type of drinks consumed by Manhattan on a Saturday night. There are 5 drinks, Beer,Martini,Cocktails, Champagne and Wine.

If shown to the mass, are they going to remember simple bars and the figure depicting the number of drinks, or a chart which is representing the same information more creatively yet simple. For instance, icons of beers , wine glasses and martini glasses for each category instead of the normal labels?

From personal experience, some information is better remembered when associated with pictures. The below linked paper also mentions that people were able to recall the visual imagery charts in a better way after a longer period as compared to the normal conventional charts.

When it comes to visual embellishments , I feel  the audience of the visualization is very important. A marketing executive or a sales executive wouldn’t want to see images for monsters in monstrous costs or pictures of beer in the sales trends, but the general audience will be able to remember it in a better fashion.

References – View page 9 , figure 10 for the chart I am taking about here

http://hci.usask.ca/uploads/173-pap0297-bateman.pdf

 

This interactive network visualization sheds light on the complex relationship between the different countries in the Middle East. Anyone willing to understand the relationship between these countries can have a good overview of the political situation in the Middle East. This visualization focuses on four different relationship factors: ‘love’, ‘hate’ ,’good’, ’strained’. At a first glance, the visualization seems to be overcrowded and clumsy, but it also provides a lot of factors for drilling down and filtering.

Things I liked:

• Depicting the relationship with an interactive network graph is excellent.
• Filtering on the basis or Country or group which helps the audience to have a better understanding of the position of each country. Hovering or clicking on each country provides the filtering.
• Filtering on the basis of the relationship types.
• A short description pops up providing details of the relation between two countries when one country is selected and other is hovered upon.
• The use of different color and line type in depicting different relationships helps identify them easily when a particular country is selected.

Room for improvement:

• The visualization also distinguishes the entity based on whether it is a country or group. But the colors used for ‘Group’ and ‘Non-Muslim’ has little difference. Also, there are no means of filtering on basis of this. So, it becomes ambiguous.

Reference: http://www.informationisbeautiful.net/visualizations/the-middle-east-key-players-notable-relationships/

We all are facing different issues while completing our projects. I experienced myself losing an argument with professor because either the veracity of the data, where it came from, or how it was collected was called into question. Hence my data or my conclusions were not trusted. I learned that if I ever present arguments, back it up with data not story.

Hence after professor suggestions I made sure to have done my homework. That way, I can address any questions about data with confidence. Trust on correct data to consistently deliver meaningful, relevant results based on evidence and fact. Here are the tips in how to present data in era of alternate facts:

1. Be impartial: Try not to have preconceived notions about what the data should show or how it should be interpreted in advance. If you go into an analysis without an agenda and present your results as objectively as possible, it won’t seem like your analysis takes a side or pushes a particular point of view.
2. Provide Context: No analysis is done in a vacuum. There’s always a reason for conducting it, as well as a plethora of factors that go into what data is used, where the data comes from and the methodology you choose to approach it.
3. Obsess over accuracy: Put yourself in the shoes of your audience and try to question your numbers the way they would question them. Does everything add up? Does everything make sense? Yes? Good. Now bounce your analysis off someone else for one final review before you take it to present.
4. Admit your mistakes:Honesty is always the best policy, with no exceptions. If being accurate helps build trust, admitting it when you’re not reaps similar rewards. You will get far more respect for owning up when you are wrong than if you cover it up and are caught.
5. Be Thoughtful About How, What and When To Communicate : How, what and when you communicate can have a major impact on how trustworthy you are perceived to be, too.  On what you communicate, it is important to know your audience and explain yourself clearly in terms they will understand. Talking too much or being long-winded can turn people off and be a sign that you don’t listen.

Source: https://www.linkedin.com/pulse/how-present-data-executives-era-alternate-facts-hint-aaron-maass?trk=v-feed&lipi=urn%3Ali%3Apage%3Ad_flagship3_search_srp_content%3BK45zLsFN7bdjFYCdvXf6pw%3D%3D