As an IT professional, I can recall conducting Google searches on various technological problems 10 years ago and finding nearly instant answers. Today it seems the process is much harder; too many non sequiturs, amateur hypotheses, "me too" tales of woe and dead ends have made it difficult to pinpoint some solutions. The answers can generally still be found, but I find myself doing a lot more digging (and frankly, eye-rolling). Simply put, there's a ton of data out there and making sense of it is becoming a bigger and bigger challenge.
Of course, if you follow this site you'll know that one answer to data complexity is business intelligence. The concept can be broad or focused. One example which fits the latter category is a data processing engine called Tableau, which interprets and presents data in some fun and unique ways.
Before getting into how it works I want to present some immediate examples of how data is literally visualized by Tableau. If you head to their public demo site at public.tableau.com and click Gallery you can see some very specific illustrations as to how they display data, and I'll present some here as well.
Tableau works with data visualizations (each of which is also known as a "Viz") to display information. Their Gallery contains a Greatest Hits section and a "Viz of the Day" area where some more noteworthy Vizzes are kept.
One greatest hit which immediately caught my eye was Rolling Stone's 500 Greatest Albums of All Time:
And hitting close to home in my perspective as a mobility writer was a Viz demonstrating component cost for various Apple and Samsung smartphone models:
Another interesting Viz depicted the film locations of James Bond movies around the world:
Other examples included:
- A world map showing where a certain traveler had been and how many miles he covered along with the modes of transportation involved.
- A map of London listing where the presenter had walked in eight hours along with video clips.
- An analysis of Edward Snowden's first 24 hours on Twitter, detailing his retweets by followers.
- Parental sleeping patterns before and after having a baby.
I had the opportunity to discuss Tableau and conduct a personalized test drive of Tableau with Dave Story, VP of Mobile at Tableau. Story discovered Tableau as a customer working for Realtor.com while engaged in a 6-month Business Intelligence implementation which brought him exposure to the company. He liked it so much he signed up with them.
First we chatted a bit about the background behind Tableau.
"Data is crucial now, and the first question is 'what can you do with that data?' Story said. "Tableau started from a core situation based on a shortage of people having data and knowing how to analyze it. Our goal was to make it easy enough so the person who has the data can understand it rather than sending it off to someone else for analysis. You can understand what it means, so now it's a process of self-service data discovery, exploration and analysis."
In the 12-year history of the company they have released five products to handle data:
Tableau Desktop, a Windows/Mac application for inhaling data and providing meaningful results. It allows the user to connect to hundreds of data sources including web data (such as to Twitter or Google Analytics), Excel, MySQL, financial/custom databases and Hadoop to name just a few. The cost is either $1,000 or $2,000 per license and includes one year of maintenance.
Tableau Server, which is like "Sharepoint for data visualization," Story said. Instead of mailing around a workbook people can collaborate around data by working with it on the server - this allows them to work with live interconnections, see trending, refresh displays with the latest data, schedule actions, and proxy security. It acts as a data hub and can access curated data which is important to the enterprise.
Tableau Public, the free offering outlined above, which lets the world share interactive data visualizations. It's somewhat different from the server product, where users can work with interactive graphics but outsiders can't do so. Users can download a Tableau workbook, then use the desktop app to connect to data sources and upload the results to Tableau Public. "It's like the YouTube of data visualization," Story said, adding that one Viz has 20 million views.
Tableau Online, a cloud-based offering. It can operate the server product and integrate security/data sources. It's "like Salesforce for data visualization." Cost is $500/user/year.
Vizable application, which is free and launched in October 2015 on the iPad. It's geared towards mobile devices and addresses the core situation of a tablet with a spreadsheet - for instance, point of sale/cash register information or fitness tracker data. It differs from Tableau Desktop; it has more power, lots of menus and drag and drop functionality similar to Adobe apps.
"We went to the Apple store and told the manager we were working on app to understand a table of data on a tablet. There are 1.5 million apps out there, so we asked if they had one which could to do that. They told us to 'Open it in Excel' which is a great tragedy. The situation is much more common than people would expect - nobody has solved this problem yet. This is part of our mission with Vizable; to fill the gap that's been a limiting factor in terms of working with data in a meaningful and interactive way," Story said.
Minority Report embodied the concept of 'data visualizations' - I don't mean using hand gestures to manipulate the data; you'd be exhausted in 15 minutes. We wanted to answer the question of 'how do we do something that is amazing for touch, on a magic tablet?' If you tap and hold a spreadsheet you can open in Vizable. Our aim is to push the limit on touch and fill that gap."
Story told me he had an 1842-row Excel spreadsheet containing data about various popular American movies and that it would be the basis for our Vizable demonstration. He accessed his email and showed me the standard Excel file attachment:
Opening the file showed the raw Excel data:
Scrolling to the right displayed more columns indicating that while the spreadsheet certainly had lots of information, it wasn't particular user-friendly in terms of gleaning insight into trends or statistics:
Firing up the Vizable app, Dave proceeded to show me the data based on Global Box Office receipts filtered by Genre:
Right off the bat it was clear to me that the interactive method of accessing and presenting data was livelier than that of a stodgy old spreadsheet.
Story showed me the other filters we could apply to the information, using simple swipes and pinches to control the app.
Possible filters included "Production Company," "Director," "Movie Title," "Actors," "Ratings" and more.
The Production Company filter showed us that Columbia Pictures had earned the most revenue:
Next we looked at the highest grossing directors:
We then drilled down to see which directors had the highest number of movies on this particular list (keep in mind these are based on popular films, not ALL films):
"Apps that give you a blank canvas are intimidating," Story said. "We wanted to make a hard problem simple and data analysis is really hard. Our goal is to take our wisdom and put it at your fingertips - literally. With Vizable you're always getting a useful view, answering a question whether you ask one or not."
Next we introduced a bit more zing by analyzing revenue by Movie Title:
Now, these figures constitute the earnings at the time the film came out, not taking inflation into consideration. So, we took a look at the "Adjusted Global Box Office" data.
Now we begin to see some interesting elements. Older films like "The Godfather," "E.T.", "The Exorcist" and "The Sound of Music" are on the list, and "Gone with the Wind" turns out to be the highest grossing film of all time, with figures adjusted for 2015 values!
Next we looked at the big money makers among the acting set.
It was also worth looking at which movie ratings brought in the highest revenues:
Still working with "Adjusted Global Box Office" data, we returned to look at how much money the highest-grossing directors made.
Story then showed me how we could break the data down further to look at highest-grossing films by director.
We could then bring the actor data into the mix by showing the top actor for these films:
To gain additional insights we looked at the budgets and profits involved with these films:
Here we can see that films like "Jaws" and "E.T." reaped tremendous payoffs compared to what investors had to provide. In contrast, "War Horse," while still a profitable movie, only made about 18 million dollars over what they spent filming it, perhaps due to being a period piece appealing to a certain group of people.
We dropped the director data to look solely at the most popular films:
It seems like "Gone with the Wind" seems to be taking the top spot, so we excluded that and sorted the data to see which movies made the biggest profits:
We then flipped the concept to look at films with the largest budgets:
We looked at further statistics during the presentation, including line graphs showing adjusted global box office revenue by year:
Other data included movie lengths and ratings vs/ profits. Story showed how it was easy to attach certain graphs to email to communicate results to others. He also demonstrating logging into his Amazon account and using the Vizable app to present similar graphics showing his purchase history, the books he preferred, and an interesting analysis of which menu items in a theoretical restaurant menu were the big winners:
"It's really fun to be able to see, understand and explore your data in a friendly and powerful application," Story said. "Vizable is being made free to the public in the iTunes app store. It's just for the iPad at the moment, since we're focusing on highest usage platform.
You know, when we set out to build this we had no idea how hard it would be. It took 10 years of code to move up from the Desktop app - that was to read the information, apply machine learning to understand what kind of data it represented, then the guys who built the animation engine made it fast so it could run on a tablet, split categories, do an animation. It can split a thousand orders by product ID, price, etc."
I wanted some specific examples of using Vizable out in the field, so I talked to Marination, a taco food truck and restaurant based in Seattle.
Scott Matteson: "What made you aware of the problem which drove you to seek Tableau as a solution?"
Marination: "In our case we were invited by Tableau to participate in a beta test of their new iPad app - Vizable. When presented with this opportunity it was fairly simple to envision how having this tool might help my business, and those of us who lead it, to become more self-aware. Although there was no specific problem there was absolutely a broad horizon of "what if we knew XYZ" opportunities that made using Tableau's Vizable a very attractive opportunity."
SM: "How has Tableau helped your business?"
Marination: "To start it has heightened our awareness of the need for good clean data. This has pushed us to work with some of our vendors to get this data, and also challenged us to create some internal systems to track data so that we can use it in the app. The app itself has offered us a clean, new way to review large amounts of data and quickly look for meaningful 'anomalies' that we can look into."
SM: What are some examples of day-to-day use of the product?
Marination: Honestly we are not set up yet to have a daily stream of good data. But when we do have data and use the app, we typically review our cost of goods and major financial metrics.
SM: "What gaps does the product fill for you and your business?"
Marination: "Vizable is like having another set of eyes at the managers table. It provides another perspective that sometimes validates our discussions, and sometimes gives us pause or reason to look into something more deeply."
SM: "How easy is it to use Tableau/interpret the results?"
Marination: "The app is very simple to use once you learn the basic functionality. The results are clearly presented but like all statistics you must question and validate the source data."
SM: "What specifically about Tableau works on a mobility basis for you (accessing real-time data on the go, etc.)?"
Marination: "Since we work with the app on a tablet it is very easy to engage with Vizable wherever I go."
And there you have it. I like reading frontier history from the 1700's and 1800's, and one theme that resonates throughout is the need for an experience guide. Tableau can be that guide as you navigate through the immense - and unforgiving - forests of data which surround us on a daily basis. Having the proverbial trail of bread crumbs to follow means better business decisions and insights, a win for both businesses and customers.
Scott Matteson is a senior systems administrator and freelance technical writer who also performs consulting work for small organizations. He resides in the Greater Boston area with his wife and three children.