Ventana Research recently completed a comprehensive evaluation of analytics and business intelligence products and vendors. Such research is necessary and timely as analytics and business intelligence is now a fast-changing market. Vantana’s Value Index for Analytics and Business Intelligence in 2015 scrutinizes 15 top vendors and their product offerings in seven key categories: Usability, Manageability, Reliability, Capability, Adaptability, Vendor Validation and TCO/ROI. The analysis shows that the top supplier is Information Builders, which qualifies as a Hot vendor and is followed by 10 other Hot vendors: SAP, IBM, MicroStrategy, Oracle, SAS, Qlik, Actuate (now part of OpenText) and Pentaho.
The evaluations drew on Vantana’s research and analysis of vendors’ and products along with their responses to their detailed RFI or questionnaire, their own hands-on experience and the buyer-related findings from their benchmark research on next-generation business intelligence, information optimization and big data analytics. The benchmark research examines analytics and business intelligence from various perspectives to determine organizations’ current and planned use of these technologies and the capabilities they require for successful deployments.
Vantana found that the processes that comprise business intelligence today have expanded beyond standard query, reporting, analysis and publishing capabilities. They now include sourcing and integration of data and at later stages the use of analytics for planning and forecasting and of capabilities utilizing analytics and metrics for collaborative interaction and performance management. Their research on big data analytics finds that new technologies collectively known as big data are influencing the evolution of business intelligence as well; here in-memory systems (used by 50% of participating organizations), Hadoop (42%) and data warehouse appliances (33%) are the most important innovations. In-memory computing in particular has changed BI because it enables rapid processing of even complex models with very large data sets. In-memory computing also can change how users access data through data visualization and incorporate data mining, simulation and predictive analytics into business intelligence systems. Thus the ability of products to work with big data tools figured in their assessments.
In addition, the 2015 Vantana Research Value Index includes assessments of their self-service tools and cloud deployment options. New self-service approaches can enable business users to reduce their reliance on IT to access and use data and analysis. However, their information optimization research shows that this change is slow to proliferate. In four out of five organizations, IT currently is involved in making information available to end users and remains entrenched in the operations of business intelligence systems.
Similarly, their research, as well as the lack of maturity of the cloud-based products evaluated, shows that organizations are still in the early stages of cloud adoption for analytics and business intelligence; deployments are mostly departmental in scope. Vantana is still exploring these issues further in their benchmark research into data and analytics in the cloud, which will be released in the second quarter of 2015.
The products offered by the five top-rated companies in the Value Index provide exceptional functionality and a superior user experience. However, Information Builders stands out, providing an exceptional user experience and a completely integrated portfolio of data management, predictive analytics, visual discovery and operational intelligence capabilities in a single platform. SAP, in second place, is not far behind, having made significant progress by integrating its Lumira platform into its BusinessObjects Suite; it added predictive analytics capabilities, which led to higher Usability and Capability scores. IBM, MicroStrategy and Oracle, the next three, each provide a robust integrated platform of capabilities. The key differentiator between them and the top two top is that they do not have superior scores in all of the seven categories.
In evaluating products for this Value Index Vantana found some noteworthy innovations in business intelligence. One is Qlik Sense, which has a modern architecture that is cloud-ready and supports responsive design on mobile devices. Another is SAS Visual Analytics, which combines predictive analytics with visual discovery in ways that are a step ahead of others currently in the market. Pentaho’s Automated Data Refinery concept adds its unique Pentaho Data Integration platform to business intelligence for a flexible, well-managed user experience. IBM Watson Analytics uses advanced analytics and natural language processing for an interactive experience beyond the traditional paradigm of business intelligence. Tableau, which led the field in the category of Usability, continues to innovate in the area of user experience and aligning technology with people and process. MicroStrategy’s innovative Usher technology addresses the need for identity management and security, especially in an evolving era in which individuals utilize multiple devices to access information.
The Value Index analysis uncovered notable differences in how well products satisfy the business intelligence needs of employees working in a range of IT and business roles. Vantana’s analysis also found substantial variation in how products provide development, security and collaboration capabilities and role-based support for users. Thus, they caution that similar vendor scores should not be taken to imply that the packages evaluated are functionally identical or equally well suited for use by every organization or for a specific process.
To learn more about this research and to download a free executive summary, click here.
Source: Tony Cosentino, VP and Research Director, Ventana Research, April 22, 2015, http://blog.ventanaresearch.com/2015/04/22/whos-hot-in-analytics-and-business-intelligence/.
But as a student at Auckland University of Technology, designer Sara Marshall redrew 10 famous logos with paintbrushes, markers, and pilot pens. It wasn’t just some fantasy project in which she reimagined Subway and Burger King with flowers and filigree. She meticulously translated these rigid, well-known brands through careful calligraphy. And the results are like an uncanny peek into an alternate universe where computers and printers never came to be, and all branding must be drawn by hand.
The logos are not all perfect, hand-lettered duplicates. Marshall’s FedEx logo looks like a spitting image to the real logo in my mind’s eye, until I load up the real thing and realize, of course, FedEx isn’t drawn in italics. The chasm is even greater with Subway. The real Subway logo is written in all-caps bold italic, with arrows leading in and out of the wordmark. Marshall’s mixes upper and lower case letters, and it ditches the arrows.
The remake is a much friendlier logo than Subway’s original. But unless you’re looking side-by-side, there’s a good chance you won’t even notice what’s missing. With the green extrusion, and the mix of white and yellow lettering, Subway still feels like Subway. Marshall’s greatest accomplishment is in picking up on just enough of the original brand essence to make the new versions feel familiar and right.
Despite a relatively positive Internet response to her work since it has been discovered on Behance, Marshall looks back on the project now and sees the imperfections. “It was a really fun project to work on, but something people don’t realize is the time pressure I was under while creating them—there were many sleepless nights and 16- to 20-hour working days,” Marshall says. “There were many changes I really wanted to make to the project as well, but it took off before I had a chance.”
Now, Marshall says she doesn’t have the time to redo any of the work. She’s too busy in her day job—as a professional letterer.
[All Images: Sara Marshall]
Source: Mark Wilson, Famous Logos Look Better Lettered By Hand, Co.DESIGN, Fast Company, April 24,2015, http://www.fastcodesign.com/3045394/famous-logos-look-better-lettered-by-hand?utm_source=mailchimp&utm_medium=email&utm_campaign=codesign-daily&position=1&partner=newsletter&campaign_date=04242015#5.
Jon’s thought processes on this and why he created the visualization he created are noted below.
What do you think of this visualization and as Jon asks: What do you see in the data?
Another Way of Looking at Graduation Rates
Jon saw an article in his Facebook feed about college ROI, although it was called the 50 Best Private Colleges for Earning Your Degree on Time. As is often the case, there was nothing really wrong with the facts of that article: You see a nice little table showing the 50 Colleges with the highest graduation rate.
But it got Jon thinking: What if high graduation rate wasn’t enough? What if a considerable portion of your freshman class that graduates takes longer than four years to do so? Is that a good deal? He then created some hypotheticals:
College A: 1000 freshmen, 800 who graduate within four years, 900 who graduate in five, and 950 who graduate in six. So the four-, five-, and six-year graduation rates are 80%, 90%, and 95%. But of the 950 who eventually graduate, only 84.2% do so in four years.
College B: 1000 freshmen, 750 who graduate within four years, 775 who graduate in five, and 800 who graduate in six. So the four-, five-, and six-year graduation rates are 75%, 77.5%, and 80%. Thus, of the 800 who eventually graduate, almost 94% do so in four years.
College C: 1000 freshmen, 550 who graduate within four years, 600 who graduate in five, and 625 who graduate in six. So the four-, five-, and six-year graduation rates are 55%, 60%, and 62.5%. Of the 625 who eventually graduate, 88% do so in four years.
If you were choosing among these three colleges, which might you choose? The easy money says you go with College A, the one with the highest graduation rate. College B would be your second choice, and C would be your third. But what if you are absolutely, positively certain you’ll graduate from the college you choose? College B is first, then College C, then College A.
Data can be tricky. Jon has noted many times in the past that things like graduation rates are really almost inputs, not outputs: If you choose wealthy, well-educated students, you’re going to have higher graduation rates. It’s a classic case of making a silk purse out of, well, silk.
Jon tried to demonstrate this in the visualization he created below, and he likes the simplicity here. Each dot is a college (hover over it for details). They’re in boxes based on the average freshman ACT score across the top, and the percentage of students with Pell along the side. The dots are colored by four-year graduation rates, and you should see right away the pattern that emerges. Red dots (top right) tend to be selective colleges with fewer poor students.
But if you want to look at the chance a graduate will finish in four years, use the filter at the bottom right. Find a number you like, pull the left slider up to it, and see who remains. (Just a note: Jon is a little suspicious of any number of 100% on this scale, which would mean absolutely no students who graduate take longer than four years to do so. It might be true, but it’s hard to believe. But he would set the right slider to 99% at the most.) Jon points out to remember there is a lot of bad IPEDS data out there, so don’t place any bar bets on what you see here.
What do you see? Click on the image below and find out.
So, this week is the big Star Wars Celebration in Anaheim, California. I was not able to attend, but I wanted to find some excuse to blog the latest trailer from the upcoming Star Wars Episode 7: The Force Awakens.
I found this great infographic that shows the insides of a Star Wars Imperial AT-AT. This is from Doug Osborne from March 1, 2011 on Geek.com. The AT-AT is one of my favorite vehicles from the best of the Star Wars movies, The Empire Strikes Back.
Enjoy and May The Force Be With You.
Here is a great dataviz from Tableau Public.
Stephen Wagner explores the evolution of Crayola colors, from 1903 until now.
Click on a crayon in the “Box of Colors” to learn the name of the color, how long it has been in production, and any additional facts.
100 Breathtaking Buildings That Represent The Future Of Architecture
Where you’ll be living soon, from glass tree houses to inflatable buildings.
What’s the future of architecture? Ask any architect and you’ll probably get a different answer. But the future proposed by architect Marc Kushner, also the founder of Architizer.com, is an attractive one. Breathing buildings, treehouse-like structures—they’re all there.
Kushner spoke about his vision, laid out in a new book called The Future of Architecture in 100 Buildings, at a lunch during this year’s TED conference in Vancouver. During the lunch, he echoed many of the themes from his TED talk in 2014—namely, that we, the general public, will shape that future.
According to Kushner, social media is giving people permission to experiment more with buildings. On Facebook and Twitter, onlookers can engage with architects during the construction phase of a building, providing potentially valuable (or just annoying) feedback.
For his book, Kushner picked his 100 favorite buildings. “I chose to ask questions like, ‘Can a building stand on its tiptoes?’ says Kushner. One place in the book, the Encuentro Guadalupe eco-hotel in Mexico, looks like it can; the hotel’s buildings sit on tiny stilts around a hillside.
Another of Kushner’s favorites is the Ark Nova, a temporary concert space. It was built after Japan’s disastrous earthquake and tsunami and can be transported on trucks and then unfurled and blown up. The end result looks like a hybrid spaceship and jello mold.
Check out more of Kushner’s favorites in the photos below.
I wanted to share a blog post by Lian Chikako Chang. Lian is a Harvard University GSD M. Arch. I. Cole Nussbaumer is one of my favorite presenters on data visualization and I did not want to pass the opportunity to share her thoughts and insights.
Cole Nussbaumer, “Storytelling with Data” by Lian Chikako Chang
7:05pm: Until two years ago, CN (NOTE: Lian uses “CN” to indicate Cole Nussbaumer) worked in Google’s People Operations team, where she told stories using data to help people make decisions. She left Google to work full time on teaching people about storytelling with data.
In school, we learn about stories and language on the one hand, and about numbers and math on the other–but these disciplines rarely mix.
Graphs normally don’t look so good and aren’t that clear.
Overview of tonight’s talk:
- understand the context
- choose the right type of display
- eliminate the clutter
- focus attention where you want it (how people see)
- tell a story
Here we go! Section 1: Before you visualize the data, understand the context. Once you explore the data and find what you want to say, you move into an explanatory space, where we will focus tonight.
First, who is your audience? What motivates them? Is it making money, beating the competition, or something else? The more specific you can be, the better your communication will be. Then, what do you want your audience to do? Change, implement, empower, understand, support, create… Finally, how can data help you make your point?
Here’s an example that we’ll return to throughout the talk:
Assume your audience is the VP of product. What we want them to do is understand how competitors’ pricing has changed over time, and accept our recommended price range. We can use data to make our point by showing the average retail price over time for Products A, B, C, D, and E.
2. Choose an appropriate visual: CN categorized all the kinds of data visualizations she had made over the past few years, and came up with more than 100 examples. At first she thought the ‘long tail’ would be relevant, but when she looked closer, most of the work fell into twelve groups.
For just one number, text is the way to go, for example: 91% of data visualization experts agree.
Graphs are more visual than tables. Scatterplots can let us group instances. Line graphs are good for time, when the data is continuous. Here, a line graph shows the average as a summary statistic, as well as the max and min, which can give an idea about variance.
Slopegraphs are less common, and can pack in a lot of data, because the slope visually tells us about the rate of change. Depending on your actual data, a slopegraph might be great or confusing (if there are too many crossing lines, for example).
Bar charts are great for categorial data (different categories)! Vertical, horizontal, stacked vertical, and stacked horizontal. They’re common because they’re really effective–our eyes can easily compare the lengths of bars. Stacked bars let you give information about the sub-components about pieces–but one weakness is that it doesn’t provide a consistent baseline to compare each subcomponent.
Stacked bars as a percent of whole somewhat alleviate this problem, because there are two datums to compare to–these are great for survey data.
Rule: bar charts must have a zero baseline, because we’re comparing the difference in bar lengths against their overall length.
There is one graph type that CN didn’t include: the pie graph. Why? If you have to estimate the size of each component here, it’s really hard to tell.
Another rule is that we should never put charts in 3D, because it distorts sizes and doesn’t add any information. So you can put your pie chart in 2D and label the segments with percents–but really, CN says, consider a horizontal bar chart. They’re very easy to read.
“If you find yourself reaching for a pie chart, pause and ask yourself why.” If you have a good reason, fine–but if not, consider using another chart type.
Let’s work on this chart. First, get rid of the color to focus on the trend over time.
Then highlight 2010 to the present, to see if there’s a trend. There is, and the data is over time, so let’s make it a line graph and stack the lines along the same axis.
3. Identify and eliminate clutter. Strip anything away that isn’t adding enough value to justify its presence. Gestalt theory from the early 20th century informs how we visually perceive information.
a) proximity – we see horizontal and vertical rows here, simply based on the distance between dots.
b) similarity – allows us to avoid dividing lines in the table, just by using two colors.
c) enclosure – we tend to see enclosures, so if we get rid of heavy borders, our data stands out more.
d) closure – this has something to do with how we tend to mentally complete partial shapes. Basically, remove pointless outlines.
e) continuity – I’m not sure what this is about, but maybe you can glean it from the above diagram.
f) connection – for example, connecting dots in a line graph can be helpful
Gridlines are OK if you think your audience will really want to check the exact values, but make them light (line weights!)
CN gets rid of the strangely shaped dots, zeroes after the decimal, and gridlines. She makes the x-axis labels horizontal so that they can be more easily read. Then she puts the labels near (proximity principle) their respective lines and colors them (similarity principle) according to each line.
4. Focus attention where you want it
Iconic memory is shorter than short-term memory–it is momentary, and tuned to pre-attentive attributes. This allows us, for example, to count the number of “fives” on this slide much more quickly, when they are colored differently.
There are a variety of pre-attentive attributes. We tend to assume some are quantitative (length) and some are categorical (or qualitative) (such as color). We can combine them to format a text, to make it scannable within a few seconds. For example, we can go from this, which just calls out one statement…
…to this, which is fully formatted:
In a presentation, you can show an un-highlighted chart to talk about the data in general, then focus people’s attention using color or other pre-attentive attributes, as below.
5. Tell a story. “And here, I mean a full-on children’s story.” For example, Little Red Riding Hood. CN gets someone in the audience to recap the plot. The story has a few lessons for us:
- The power of repetition: we know the story because we’ve encountered it many times.
- Sequence and plot twists.
Words are very helpful in data visualization. Some are mandatory, like axes and labels. Annotations are also helpful to make sure people can come away with your key takeaway.
Here’s a good example of how annotations make this data on peak break-up times come alive.
If there isn’t anything interesting about the data, then don’t show the data. It sounds basic, but it happens all the time. You risk losing your audience for when you do have something important to say.
Back to our example: CN runs through a series of slides that she might run through in a few minutes in a presentation, to show the data in our example. Instead of talking through a static graphic, she builds the data over a series of slides and calls out different trends and parts of the data using colors, as she builds her argument.
Question: What if you want people to figure out the data on their own, without telling a dominant story? CN: Good question. When you use pre-attentive attributes to highlight one part of data, you’re de-emphasizing others, so that people are less likely to see other stories.
Question: What about infographics? CN: Some are fluffy pictures that have little data. Others are actually informative–for example, the New York Times, National Geographic, or Wall Street Journal. These allow you to sit with the information and see insights for yourself.
Question: You mentioned that bar charts should always show zero. What about line charts? CN: You can get away with zooming in for line charts, because the main comparison is between points over time. But the risk is over-zooming, which will make small differences seem more significant than they are.
Question: Can you explain waterfall graphics, which was one of your types? CN: They’re good when you have a beginning point, then additions or subtractions, and an end point. For example, in People Operations (i.e. HR), your team starts at a certain size, then grows or shrinks over time as you add or lose people.
Question: Is there research comparing the takeaways that people get between pie charts or bar charts? CN: I think you’re probably right that this is more anecdotal than proven–but maybe someone knows a specific study. Someone else in the audience: There are studies on this, and usually bar charts win in terms of people remembering the numbers. But it’s really hard to research the gestalt feeling of a “percent of the whole,” where pie charts are actually effective. So is the story about the specific numbers, or the relative amounts, as a percent of the whole? If it’s the latter, then pie charts can work.
Question: What if you have a ton of data and want to allow your audience to explore it? CN: I find that we often want to do that, when really what we should do is take our analysis a step further. There are different use cases, but it’s often dangerous to not present an analysis.
Question: As a personal project, I’ve recorded all my activities in Google Calendar for a year and a half, and I put it in a pie chart, but what would you recommend? CN: I would not choose a pie chart, myself, but it depends on what you want to get out of the data. It’s often about playing with the data, to see what works. And put your graphics in front of someone and watch their facial muscles–see how painful it is for them. Ask them to describe their thought process. I also use the optometrists’ approach, A/B testing minor changes in the graphic, to iterate through small changes.
Question: What about animation or interaction? CN: There is definitely a place for interactivity on exploratory data. Not everyone will be willing to dig through the data, though, so can you start an interactive graph with an explanatory view that already says something? That is really helpful.
Audience comment: In defense of pie charts–if you have two categories and want to show a percent of the whole in a relative sense, a pie chart is great! CN: But to play devils’ advocate, if you just have two categories, you can also just show the number.
Question: What about platforms like Tableau that have a specific approach towards graphics? CN: Tableau is fantastic for exploratory analysis, because it has stripped the crap away. They’ve recently added a “storypoints” feature, because they’ve recognized that they want to do better at storytelling. For me it’s not about the tool–these principles work with any tool.
Cole ends by plugging her website, Storytelling with Data, where she’s got lots of great content, including material from her talks.
Thanks for reading!
Where the Germanwings Plane Crashed
The plane went down in a remote part of the Alpes-de-Haute-Provence department, and search teams struggled to get to the area. When French air traffic controllers lost contact with the aircraft, it was flying at approximately 6,000 feet; the elevations in the search area range between 2,000 and 9,000 feet.
No helicopters have been able to land because of the rugged terrain around the crash site. Searchers had to be lowered, further slowing recovery efforts. The size of the debris area, which was about the size of three to four football fields, suggests the plane hit the ground at a very high speed, according to the French interior minister, Bernard Cazeneuve.