Mike Roberts

Creating a Dynamic Anchor date with a Relative Date filter

Thu, 10 May 2012 00:01:41 -0400

Ever wonder how you might change a Tableau anchor date (see image below) to be dynamic? Fear not, I’ve created a solution which might satisfy those individuals and/or groups who clamor for greater flexibility.

Here are my 3 (yes, just three) steps to Tableau magic:

Step 1: Create (and then show) the following Parameter(s):

Note:

Anchor Date is a Date data type
Choose Interval is a Integer data type
Choose Time Segment is a string data type with (Years, Quarters, Months, Weeks)

Step 2: Create Calculation(s)

Step 2.1: Anchor Begin

Step 2.2 Anchor End

Step 2.3 Filter

Step 3: Set it up

The end. You can also see the workbook here.

Controlling the 'Data Gristle' with Tableau

Sat, 21 Apr 2012 15:14:51 -0400

What do Tableau Software, Business Intelligence and User Experience have in common? Combined, they represent the New Age of Data. More specifically, they push data visualization, data personalization and data emergence to higher levels of understanding. At the present moment, we are seeing a massive cultural shift in how we understand our corporate data, industrial data or, simply, our personal data.

One caveat: we need the right tool for the right job while simultaneously ensuring visual analysis has form and function.

‘To paint well is simply this: to put the right color in the right place.’ – Paul Klee

Simple and practical advice, right? Not really, especially when one considers the flood of color hitting the human eye. In fact, ‘Human eyes are exquisitely sensitive to color variations: a trained colorist can distinguish among 1,000,000 colors…for encoding abstract information, however, more than 20 or 30 colors frequently product not diminishing but negative returns’ (Tufte, 1990). This shouldn’t come as a huge surprise when creating visualizations in Tableau Software. If one looks at their diligent planned color palettes, you’ll be hard pressed to find one categorical palette exceeding 30 colors.

Why, then, is this important for visual analysis? Can’t we just throw data onto a chart and accept the defaults? Eh! Please. You wouldn’t hear a BI Engineer talking about a ‘galaxy schema’ of data just thrown around only to be left up to random chance and gravity; rather a seasoned BI Engineer knows that, for analytical workloads, a star schema is the most viable option. And you wouldn’t ever hear a surgeon saying: ‘Well, I’m just going to mess around in your chest cavity and see what I see.’

My point is that designing visualizations on datasets which can exceed 100 million records requires a skill and a desire to understand the user and his/her experience. One important feature, obviously, is avoiding the ‘visual war’ and the ‘1+1=3’ effects. In the image below, what immediately comes to mind? Typically, most people see the large, white border which takes away from what the user is actually supposed to focus on.

Visual analysis with Tableau Software should always eliminate any barrier to entry (read: analysis). The user shouldn’t have to ‘compute’ from container to container, nor should the user be bombarded with any unnecessary ‘data gristle’. A job well done by Tableau’s engineers; designing with a ‘human-centric’ perspective makes even the most stern of UI/UX professionals smile because of one huge advantage: Affordances.

‘A flat plate affords pushing, an empty container affords filling, and so on. Affordances can signal how an object can be moved, what it will support, and whether anything will fit into its crevices, over it, or under it’ (Norman, 1988). The expert Tableau developers know that a well-designed dashboard affords analysis. Users wanted to explore, drill down, lasso and any number of discovery features embedded into the product. Data which comes at us all with, at time, overwhelming speed can be harnessed with skill inside a Tableau dashboard. Take this one simple example. This dashboard embeds a ‘hyperlink’ to signal to the user that he/she can click to select more specific information. Instead of throwing everything inside one container, the design lightens the cognitive load and encourages the user to explore, which leads me to my final point: Business Intelligence.

‘When people have the ability to finish a nearly complete idea themselves, the success of the idea not only is more likely—it often becomes inevitable.’ – Will Wright

This, really, is the intelligence in ‘Business Intelligence’ for one simple reason: when the user can leverage their own exploratory mind when interacting with a dashboard, their answers (and follow-up questions) can have enormous power. Designing a user experience with Tableau is not a linear process. Data is not a linear process. Data is chaos (in a good way). Data is today’s currency. This type of analysis, therefore, begs a ‘systems thinking’ approach.

Thanks to Dan Roam for such a simple, yet powerful, picture and Donella Meadows for the model of systems thinking, termed the ‘bathtub model’ of systems dynamics. ‘We can look at any complex system as a ‘stock’ (the bathtub) with an inflow (the spigot) and an outflow (the drain). The essential resources needed to fill the stock…originate from some source, pass into the tub for some use, and eventually drain out into some dump’ (Roam, 2011). One can see its usefulness to BI since it, also, is a complex system, demanding data from any number of sources (which can then turn into yet another source for the system). BI professionals fill their stock with data and control inputs / outputs via Tableau to tame an otherwise insane amount of data. The human mind can only handle so much; to believe otherwise is, quite simply, insane.

As change agents in a New Age of Data, Tableau designers, BI Engineers will do well to remember that our life is a ‘systems thinking’ life. Nowhere is that more evident that our present culture of data abundance.

Tableau Software and the ‘Leans’

Thu, 12 Apr 2012 21:41:02 -0400

“We all had ‘spatial D’ or were suffering from ‘the leans.’” - Major Paul ‘Goose’ Gosden, U.S. Marine Corps

So, really, what is Spatial D?

‘It’s short for spatial disorientation, a catchall term to describe the summed result of the various perceptual illusions and degraded sensory perception…’ (Scientific American Mind, May/June 2012).

Aside from the 60s sounding name, Spatial D can have a profound influence on our daily lives. It’s easy to assume that what we see is actually what we see. It isn’t. Turns out, our brains like to guess. Try the ‘Flash Lag Effect’ experiment and see how you do. Go ahead, I’ll wait.

Amazing, isn’t it? What lurks beneath the conscious (read: thinking/processing) mind is truly the significant moving force in our reality. Think of this telling statistic for U.S. Military pilots: ‘Visual illusions alone accounted for about 20 crashes from 1990 to 2008, making the combined contribution of illusions of all types twice as high as the next biggest crash cause: fatigue.’ (Scientific American Mind, May/June 2012).

As visualization developers, we should be prudent and remember that when people interact, adjust, filter, and drill down into our dashboards, their brain is making assumptions that we couldn’t possibly anticipate. Thankfully, Tableau makes the serious errors less likely. Along with a well-designed color palette, fields which have a special affinity for each other and the magic behind the software, Tableau dashboards can actually change our bad ‘data-judging’ habits. Sure, we’ve all assumed we were eating less and less until we see the line chart that shoots up, right around the Holidays. And we’ve all seen the dashboards that emphasize the grid linesandborders (note: when you first saw this post, your eyes probably zeroed in on the big/bold font. Careful how you format…). Who wants to get tunnel vision while gawking at heavy borders and/or lines?

Thinking about the way in which our dashboard will be interpreted goes a long way in getting our point across.

Take this visual illusion (image below). Here, the brain is making assumptions about distance and convergence which, therefore, results in us thinking these two images are different. They’re not. They’re the same.

Enjoy the illusion!

-Mike

The Future of Data: Visualization.

Sun, 18 Mar 2012 11:04:10 -0400

So, a bit too much to post with only a few characters, so I’ll expand on what I’m quite thrilled about: data and visualization.

‘19^th century culture was defined by the novel, 20^th century by cinema, the culture of the 21^st century will be defined by the interface.’ – Lev Manovich

The aforementioned quote was seen in a spectacular TED video about data, art, and technology. What was particularly impressive was the core message: the interface. As data analysts (or just human thinkers), there is great power in being able to visualize and then see facts move. What was once static, cold and unfamiliar, the information now comes to life and takes on a new form; it’s anthropomorphism at its best. And that is a truly amazing idea: data is familiar and, therefore, comprehensible.

So, who cares at all? Back to the technology world and the interface leader, Tableau Software. This technology is poised to allow us all to do just that: make information familiar. With stellar human-centric design, its interface demands exploration and, subsequently, further questions. We all want to see patterns unfold only to be utterly surprised with what our eyes (brain, really) consume. Now the data isn’t just data; rather, it’s part of us, it’s what fuels our insatiable curiosity.

Try it. Go explore a dataset you are familiar with (for me, it’s the years of running data). It’s not stored in a cold, dark place anymore. It’s in a ‘clean-well lighted place’ (thanks Hemingway).

Enjoy!

-Mike

Tableau and the IIDA

Sat, 10 Mar 2012 20:42:58 -0500

I’m confident the sudden, all-encompassing change in pace for ‘data’ over the last few years has been fueled by our need to want more information out of our world. The immediacy and sheer speed with which it arrives is faster than most of us have been accustomed. This is not to say that the information has not always been available; rather, it’s that all of the information is now arriving, effectively, at the same time.

Yesterday, we could manage the big data. Today, we have to manage the fast data. We need to make sense of all the disparate pieces of random information.

Let’s step out of the ‘data’ realm and reference the work of Jackson Pollock and his gift for understanding complexity because it relates to the importance of interactivity.

‘My painting does not come from the easel. On the floor I am more at ease. I feel nearer, more part of the painting, since this way I can walk around it, work from the four sides and literally be in the painting. It is only when I lose contact with the painting that the result is a mess. Otherwise there is pure harmony, and easy give and take, and the painting comes out well.’ – Jackson Pollock

We will (forever) get better use out of our information if we can feel as if we are a ‘part’ of it. What’s more, if we are able to ‘not lose contact’ with the art, we can discover more and more of what truly makes this information useful.

As this brilliant whitepaper explains, ‘…visual analysis typically progresses in an iterative process of view creation, exploration, and refinement. Meaningful analysis consists of repeated explorations as users develop insights about significant relationships, domain-specific contextual influences, and causal patterns.’

Tableau does the ‘iterative’ process better than any software I’ve been fortunate enough to experiment with. It works with your brain’s awesome capability for teasing out patterns as well as its enormous and remarkable ability to parallel process. After all, information is worthless if it is not comparable to something related, something associated with our existing knowledge.

Lighten the load, really. A well designed Tableau dashboard should always be considered an interactive information-driven application (IIDA). We are not stuck in the grid nor are we limited to certain types of charts, widgets or colors. We’re free to express our many thoughts onto the dashboard. But with this freedom to create, we also have to think about what this might mean for the lucky individual who gets to consume your information (even if it’s only you). Should your information take 40 seconds to load? Should your information take on the guise of a dashboard when it’s really just a boring process document? NO!

In the end, think beauty and cognitive load because ‘…working memory is limited. Users are more likely to move around a site with a simple structure than one with a very wide or very deep structure.’

Your IIDA is the key to a better, more flexible life. If all this doesn’t make sense, go learn about Jackson Pollock and then relate that learning to your IIDA.

Tableau Data Engine power

Sat, 25 Feb 2012 16:30:47 -0500

What follows is a very context-specific scenario where the user needed a live connection without any delay. They needed speed and power. Tableau, I said, could certainly answer that question. Here’s what I learned…

When this works:

You need a live and fast connection (think ‘speed of thought’ here)
You have custom SQL which has multiple table joins
You don’t want to have to query the database every time the user changes the view (these seconds add up)

When this won’t work:

Large (>200k) record datasets
When you don’t need a live and fast connection
When you have a columnar technology (Vertica is a great example)

Essentially, this method is very context specific. Would I use it all the time? Nope. Would I use it when the situation was right? Absolutely.

Let’s keep in mind, though, that the Tableau Data Engine is a phenomenal tool to leverage. And I’m always of the opinion that what the customer/client/user can’t see in our visualizations / dashboards influences their experience in a positive (or negative) way. Imagine if it took over a minute for Google to load a news story or stock quote; we’d get pretty frustrated with everything we did after that point. Quickness and flexibility (low latency, really) in analytic dashboards is just as much a need as a great chart or cross tab.

Okay, so here’s what I did.

Step 1:

Add a column in your sql that you can use as a Context Filter. Ideally, this will be your Slowly Changing Dimension (SCD). You don’t want this dimension to change or have the user filter on it. Otherwise there are performance issues when it’s recomputed.

Step 2:

Connect to your data in the way we’re all accustomed. In our ‘Custom SQL’ case, you just paste (or write) your code in the appropriate window.

Step3:

Make your view in the desired way. This part’s really up to what the user needs. It could be a chart, cross tab, etc. You get the point. Just make it well and don’t make it fragile. And make this first!

Step 4:

Remember that extra column you added? Good. Now, drag it on the filter shelf. When the option comes up, make sure you check the ‘Exclude’ box and then select ‘None’. This will ‘materialize’ the joins. Essentially, it calculates them ahead of time so Tableau won’t have to re-query your database each time the user changes a view in the dashboard.

Step 5

Select that column and use the drop down. You’ll see the ‘Add to Context’ option. Do that. You’ll see the usually query dialog and then you’ll see an ‘Importing Data’ dialog display! Importing data with no extract?! That’s the power of Tableau. If you check the log files for the work you did, you’ll see that Tableau is smart enough to make a temp table of that data. Fast and free J

After you do the above, it should look like this:

Once that’s done, start changing the views. You’ll notice that there is no delay in the ‘executing query’ portion. NOTE: there is that small, up-front cost to generate that temp table. After, however, you won’t experience the 3 or 4 second queries. Hard work up front = more fun down the line.

Enjoy !

-Mike

Allegory of the Business

Wed, 25 Jan 2012 22:36:37 -0500

Data into information and information into knowledge. What, exactly, does this mean to you? For the latest version, refer to this article. It’s interesting how much this comes up in everyday business conversation (and rarely in anything else outside of the business). But, I question how functional it is to our business use; more appropriately, what the heck does it even mean? Isn’t everything we say, do and present an effort to convey some sort of knowledge? Perhaps, but all ‘knowledge’ is not useful and all useful knowledge is not available.

This drive for knowledge can be traced all the way back to Plato and his ‘Justified True Belief’ theory. Even better, just read all about the ‘Allegory of the Cave.’ Aren’t we all just trying to break away from the illusion of knowledge?

At some point, we have to put our foot down and demand knowledge. We swim in the water of data and simply hope to make some sense of what’s swirling around us. Case in point: We are confused and need to use our common sense. We must augment our brains power to make order (in my view, THIS, is what knowledge is all about). How do we do that?

First, let’s play. Below, you’ll see a table of numbers. There are 4 datasets, each with x/y values. Take a few minutes to get to know it.

Okay. Done? What did you gather? Knowledge? Information?

Now, take that same set of information and visualize it.

Now we have order. We’ve made a tool to shape some future (or present) decision.

Why Visualization

Mon, 23 Jan 2012 18:27:40 -0500

Why Visualization:

Too good:

‘Finally, and perhaps most importantly, visualizations give us access to actionable insight.’

@tableau just dominating.

Thu, 19 Jan 2012 21:17:42 -0500

@tableau just dominating.

Create Knowledge with Data

Thu, 12 Jan 2012 22:01:10 -0500

This seems to be a hot topic, as well it should, around the info viz community. While I agree that the purpose of each and every visualization should be to convey some meaning, I do not believe that each viewer is going to instantaneously understand this ‘masterpiece’. Every one of us is going to approach a chart and/or dashboard with a sense of curiosity and intrigue. If it’s done incorrectly, we’re turned off. If, on the other hand, we have some beneficial dashboard that does not conflict with our purpose (really, our reason for even wanting to look), we’re going to look and we’re going to be engaged. Designing a ‘perfect’ dashboard is, in a sense, a logical fallacy.

‘Let’s bear in mind that long-term recall is rarely the purpose of information visualization. When we interact with visualizations to explore and make sense of information, insights arise and lead to new questions, which in turn lead to further insights, and so on. Long-term recall is seldom needed.’ – Stephen Few

Why is this true? Simple: We don’t know the question to ask if we don’t know what we want to ask. Users are intelligent ( well, for the most part :-) ) and have an innate sense of what works. Any dashboard designer can tell you this based simply on the type of questions being asked while demonstrating a completed dashboard. Questions such as: ‘Wait, what does that color mean?’ and ‘Where can I find the trend?’ might lead a designer to believe that he/she should improve the layout of their dashboard.

So, ‘creating’ knowledge is a tall order for any dashboard designer. I believe the more appropriate theme should be: ‘Tell a story’. If we assume we can ‘create knowledge’, we are setting ourselves up for failure. Just tell a story. People like a good story.

Doing so, you let the user bring an experience and skill set to the picture and, with a designer’s touch, the whole will be greater than the sum of its parts.

See the two images below (thanks to Stephen Few: http://www.perceptualedge.com/) and decide what story to tell with the ‘Suitable’ sector. If you’re thinking the ‘Unsuitable’ might be appropriate, please do some research. You’re users will thank you.

Tableau Software: A Lesson in Design and Data Visualization

Sat, 12 Nov 2011 17:37:00 -0500

‘Again, the designer’s job is not to fill in all the space. It is to make information accessible and appealing. The best use of the page’s empty space is to help make information scannable, not to make the pages pretty. The point is to increase the page’s absorbability.’ - Alex W. White

Let’s admit, there’s information everywhere. We hear about it on the news, in the car, with our friends and anywhere and everywhere that has some story to tell. Why, then, can we not properly consume all of this? The ‘Awareness Age’, as I like to call it, is like a trail run: we need the right shoes for the right race and we don’t want to find out the shoes suck in the middle of the run.

At this point, we have an abundance of shoes but, at the same time, we don’t know which ones to use.

Rather than become frantic and too absorbed with our current, yet, ‘sexy’ technology, we should be prudent and ensure it is an effective use of our energy. More appropriately, our visualization technology should also appeal to the heart and minds of its users, consumers and designers.

In our case, the data case, we have the right shoes and the right race: Tableau Software.

Step 1: Design it well enough to be inviting.

‘The single most overlooked element in visual design is emptiness. The lack of attention it receives explains the abundance of ugly and unread design.’ - Alex W. White

When I first used the software to interact with my data, I was entranced. The software was sleek and clean. It didn’t confuse me with a flood of options; I seemed to ‘know’ where to go when I needed to go there. That’s the critical part: when I needed to go there.

The movie poster designed by Saul Bass (image below) provides a great example of ‘emptiness’. The shade is framed by emptiness and the eye is drawn down toward the hand pulling the shade. Its design is really about what’s not there at the first moment of sight. As the viewer engages more, the poster seems to provide a greater amount of information for our eyes; it provides information as we ask for it.

So what in the world does this ‘emptiness’ have to do with Tableau? Simple, really. Tableau is the only software out there that is both ‘empty’ and technically deep. It’s not an abuse to the senses. Data/Information as we all know is not pretty in and of itself; it just sits, waiting. With Tableau’s beautifully arranged menu options, chart types and analytical firepower, visual analysis brings life to our information. It brings a sense of organization and symmetry to the dissonance of data.

Tableau does an excellent job of actually framing the analytical process around the ‘emptiness’. Data belongs in a section, split by two roles: dimensions and measures. To the right, the emptiness just begs to be filled (correctly of course). Above, menu options provide only when asked to provide. It’s important to stress: Tableau is designed well yet Tableau is also a vehicle for designing well.

Step 2: Use a beautiful tool to create a beautiful tool.

Do a web search for ‘visualization’ and you’re bound to get upwards of millions of results, recommendations and images. What constitutes ‘good’ visualization? The best and most organized place to start is here: http://www.perceptualedge.com/blog/?p=1095

The insight provided by Few and his wonderful musings is increasingly necessary in our present day. Why? This example may help.

I remember an old class in college that was about change and people’s ‘slowness’ in adapting. More specifically, it was about the advent of the train and how people couldn’t understand its speed because they had no previous frame of reference.

As a result, there would be frequent accidents because we, as humans, miscalculated and didn’t truly understand this new technology.

Now, as we get used to Tableau’s great software, we should take care to understand and apply visualization best practices before we cause any undue harm and seriously stupid mistakes.

There’s an excellent book, Visual Complexity by Manual Lima. In it, he talks about eight visualization principles. The two I find very salient are: Manage Intricacy and Look for Relevancy. When we’re designing dashboards, information applications or just about anything, we should be prudent and remember that while the need to get the information out to the greater population is of great importance, it’s also important to represent that information well.

Let’s be honest, it’s not being done right now. True, it’s getting better, but there’s hasn’t been anything like Tableau and now we can actually ‘Manage Intricacy’ and ‘Look for Relevancy’. It’s now an embedded part of our analytical process. We think, design, return to normal, think, design and then, for some of us, start over. This wasn’t easy with the older, less flexible software. This is easy with Tableau. We can create beautiful dashboards which invite the user in and speak to them; it’s a dashboard that wants us to ask more and more questions. If it’s done correctly, it creates a buzz and, in the business sense, creates more appropriate, direct work.

Here’s to creating beautiful emptiness.

-Mike

Beautiful things are everywhere. It’s simply a matter of...

Sat, 22 Oct 2011 18:31:38 -0400

Beautiful things are everywhere. It’s simply a matter of being there.
(This image was from the Wynn in Las Vegas)

Pure and complex

Tue, 11 Oct 2011 12:14:13 -0400

Pure and complex:

bobulate:

Weather affects the market — how sunny (or not) shades how people trade stocks and buy products. As does the complexity of product names. But complexity has a place, says psychologist Adam Alter:

What complexity does is it acts as a cognitive roadblock. …. If you have a communication that last 30 seconds or a minute or even five minutes, if you know there’s a particular point that you really want people to pay attention to — you’ve already hooked them in, they’re interested and they’re motivated — if you introduce complexity even briefly, that changes the way people think. They go from thinking in this very shallow, very superficial way … to thinking much more deeply about whatever you say next.

You still want to keep the message simple, but if you pick that moment of complexity carefully and appropriately, you can lead people to believe whatever you say after that moment of complexity very deeply. If the message is a complicated one … that’s a really effective technique.

Super interesting. It’s not unlike how Paco Underhill describes retail landing strips or Jeremy Denk describes intentional complexity as told through Chopin:

Chopin writes “enforced” listening moments into the piece–strangely arresting moments, like that F# held, alone, then heard against an astringent dissonance, then heard alone again, then heard against the “correct” dissonance…

Well-placed complexity has a place. If only to encourage us to think more deeply and globally about simplicity.

Networked Knowledge and Combinatorial Creativity | Brain Pickings

Sun, 09 Oct 2011 11:10:54 -0400

Networked Knowledge and Combinatorial Creativity | Brain Pickings:

Read this. You’ll like it.

Infographics - GOOD

Fri, 07 Oct 2011 20:04:16 -0400

Infographics - GOOD

This man’s an artistic genius: http://www.nopattern.com/

Fri, 07 Oct 2011 19:57:23 -0400

This man’s an artistic genius: http://www.nopattern.com/

A growing field and a must read...

Fri, 07 Oct 2011 19:46:30 -0400

A growing field and a must read...

What does that mean?

Fri, 07 Oct 2011 19:43:00 -0400

An interesting point was made today during a planning meeting. Has visualization really dropped to a point in which our Visual Analysis (more detailed and involved than typical analysis) is being led by the text-fueled masses? It shouldn’t come as a surprise that reaching for a deeper answer involves digging through the multiple tables in a data mart or simply drilling into an Excel farm. By doing this, we neglect the power of vision and the power of our brain’s pattern matching skills. We’ve all been there when the text files don’t open. Where’s the link? What’s the catch? What’s causing this massive drop in (insert variable here)? It’s a great void, really.

Enter ‘Visualization’.

Now, I should say that I’ve encountered more people who DON’T know what this means than those who DO know what this means.

Go ahead, let that settle in. What does ‘Visualization’ mean?

It means that times have changed. Just like the 56k modem was once a revolution in Internet accessibility, Visualization is a revolution for the Information Age. It’s like returning to an old neighborhood where all the comforts seem like second nature. We’re using a natural skill set when we see and analyze with our eyes. It’s brilliant to be honest. Consumption and patterns will reach out and grab our attention and the ‘gist’ of each bit will remind us of why we became information-seekers in the first place.

Why is this important? PSE. Here’s why:

Human PSE is truly Olympian. Tests performed years ago showed that people could remember more than 2,500 pictures with at least 90% accuracy several days post-exposure, even though subjects saw each picture for about 10 seconds. Accuracy rates a year later still hovered around 63 percent. -John Medina, Brain Rules

Back to the Visualization point mentioned earlier. What do we ‘do’ when we perform information visualization? We create stories and, more specifically, we create pictures of information. Sure, these pictures are in the form of charts and graphs (some may even have motion to them). But, it’s vital to remember that they are still (and will always be) superior to the text format we’ve all been used to, the format that’s really prevented people from understanding what ‘Visualization’ truly means.

In the end, we must try and continue to push the envelope when it comes to creating sensible, dramatic and accurate information visualizations. Without it, we’re lost in the jungle trying to reach for random sources of information vital to our survival.

Just make the pictures well enough to understand.

Tableau Software: Providing Multi-sensory Wonder

Fri, 07 Oct 2011 19:41:00 -0400

The future of Business Intelligence (or just data intelligence) has certainly arrived with the set of tools made available from Tableau Software. I won’t expand on the many reason why (numerous experts have already done so). I will, however, highlight one important aspect I believe will have far-reaching implications for our data consumption skills. The new ‘touch’ capabilities with Tableau 6.1 and the iPad will make learning from data (which is the goal, right?) more and more effective and relevant. I say relevant because data in its previous format was simply one dimensional and, well, it lacked a certain value to the viewer. It didn’t let the viewer explore in his/her own organic matter. Information, in any medium, is worth it when we can discover something we really haven’t set out to discover. It is even more tangible when we use other senses to explore, when we touch the information.

Enter 6.1 with the iPad app.

With the ability to design dashboards for the iPad, viewers will now be able to use an additional sense to increase the power of information: touch. I find the following piece very illuminating:

It’s not just combinations of sight and sound. When touch is combined with visual information, recognition learning leaps forward by almost 30 percent, compared with touch alone. These improvements are greater than what you’d predict by simply adding up the unisensory data. This is sometimes called supra-additive integration. In other words, the positive contributions of multisensory presentations are greater than the sum of their parts. Simply put, multisensory presentations are the way to go. -John Medina, Brian Rules

It’s truly fascinating if you really think about it: dashboards that are touch optimized could cause an increase in learning! Often, Business Intelligence is all about results and delivery. I find it’s rarely about design and audience comprehension; we don’t always push to think about what the end-user might want to see in their information. This should change.

Let us design information dashboards with elegance and with an understanding of the many variables in play: brain rules, human rules, sensory rules, and many more. The goal should always be to seek the truth.

Interact with the dashboard below and, if you can, use an iPad to immerse yourself in the elegance of multi-sensory learning and exploration:

http://public.tableausoftware.com/views/Oriflamme50K/Oriflamme50K

We Seek Information

Fri, 07 Oct 2011 19:35:18 -0400

“In effect, the answers the computer provides fall within the ‘finite framework’ of the data put into it. But is this finite set, extracted from infinity, the best one? The first simplifications stemming from calculation most often lead to criticizing the data and imagining new ones. No computer will tell us which data are missing.” - Jacques Bertin

There are many drives spoken of by psychologists and sociologists which really hit at the core of what it means to be human. Sure, there are those which immediately come to mind (football and my Chicago Bears; poor team, let’s get some offense!). But, really, the core of them can be counted on one hand. The one I want to make a very important point about is the need for information. Not just the latest news or which celebrity has done some crazy stunt; rather, it’s the need to understand our surroundings and our environment. We need status updates on vital matters of survival and information (valid and properly delivered) fulfills that need.

Let us not forget that we each have a unique ability to find and validate information coming at us. Granted, some is better prepared than others and some information just simply has to be hijacked by our ‘flight or fight’ physiological responses. It’s true, preattentive processing can get us there but there is some information which can be fully consumed when we give it some attention. Too much of the information out there is woefully lacking in substance and the overall connection to the bigger picture. I mean, who among us has veered quickly away from websites dangerously close to inducing seizures?

So, when we finally have this ‘finite’ set of data or, simply, information, what do we do? Do we ignore it? Make the wrong conclusions? Or just pass it along to someone else who might not even use it? It’s both the way it’s prepared and the way in which we ‘see’ the information. Seeing, as I’ve mentioned, is our forte. Humans have super vision skills. It’s time we used them for good.

“A graphic should not show only the leaves; it should show the branches as well as the entire tree. The eye can then go from detail to totality and discover at once the general structure and any exceptions to it.” - Jacques Bertin

These graphics are all around us now, thanks to both our advances in computer technology and increasing understanding of visualization principles. Our need for information, then, becomes insatiable. It’s there and it’s not going anywhere. But, the issue becomes the translation; how do we see it so this need doesn’t become a frustrating black hole? We have all these wonderful tools, but they are often misused. Misused and confused. Too many ways to program the same set of data and too many ways to untangle the mess. Our information need is currently receiving some dissonance. No fear, Tableau is here.

The unique ability of the software to blend data access (Excel to SQL Server) and visualization best practices make it a phenomenal tool for our information drive. It mixes and matches and does so without impairing native thought. We should be asking questions about all data and information not asking questions on how to work the tool. Tableau just works. The deep technical aspect of the software is just as important to the process as its ability to allow the user to visualize a tabular data set with such astounding beauty and structure. Granted, it hasn’t really hit the mark on network visualization, but who has? I’ve been able to load and visualize a 26 million record table (no primary keys OR indexes) with such ease, I thought I missed a step. Magic? No but apparently pretty close.

In the end, we should nourish our drive not starve it.

I encourage you to read about all human drives here (information being in the top part of the pyramid):

http ://en .wikipedia .org /wiki /Maslow ’s _hierarchy _of _needs

Finally, enjoy the latest iPad visualization of US energy use:

http ://public .tableausoftware .com /views /WeLoveEnergy /WeLoveEnergy