Saturday, January 31, 2009

Information Dashboard Design

Information Dashboard Design: The Effective Visual Communication of Data

4 stars 
By Stephen Few
Publisher: O'Reilly
Pub Date: January 24, 2006
Print ISBN-10: 0-596-10016-7
Print ISBN-13: 0-596-10016-7
(Amazon US CA UK, Flipkart)
Stephen Few is a recognized expert in the area of the visual display of business intelligence information.

This book serves as an excellent basic primer on the good principles of displaying information on dashboards, typically business intelligence dashboards, such that information is presented in a clear, understandable manner.

Few expresses strong opinions, so it is likely that some will tend to disagree with his advice, more on emotional than on factual grounds. Especially so if you have been a practitioner in this area. But if you keep your personal/professional biases and dogmas aside, difficult though that may be, this book is an excellent primer on the topic. It is well written, amply illustrated with examples, and at 223 pages, short enough to be read in one go. To truly grasp the effective principles of information dashboard design you would and should go back to the book from time to time to read up on specific sections. And of course, read up on the ample literature that already exists in this area. There are some excellent sites that educate, and several books covering this topic.

Despite everything that's good with this book, there are two shortcomings in my opinion.
  1. The first is its silence on a fairly recent phenomenon, the increasing use of rich interactive visualizations, typically presented via the Flash technology. Good flash visualizations would need to combine not only the basics of information presentation, but also effective use of interactivity and animations, both of which are a common feature with Flash-based visualizations, and also, lamentably, much overused and abused by many vendors. The ability to animate is so often taken as license to go wild with hyperactive animations, with bar graphs that shoot up like sprouts from the x axis, with lines that do a manic jig before settling down, with pie charts that unfurl like some Japanese fan, and more.
  2. The second is the lack of coverage on good design principles for small displays, like the ones found on smartphones like the BlackBerry, iPhone, and others. What works on a traditional 1024x768 sized monitor would not work on a smartphone screen. This area is important for two reasons. The first is that the emergence of the iPhone and its increasing acceptability as a viable enterprise smartphone means that visually rich interfaces can now be rendered on smartphones. This has been possible with Windows Mobile and BlackBerry devices, but the iPhone brings a whole new dimension in usability. The second is that 3G speeds mean that more content, with more graphics can now be sent over the wire to smartphones, and not be restricted to only SMS-es, or text alerts, or plain vanilla emails.
Possibly, hopefully, these will be addressed in a subsequent edition. Consider that Flash broke into enterprises in a meaningful manner only after 2005/2006, and the iPhone released in 2007, and was opened up to third party developers only in early 2008. So these two lacunae are more sort of inevitable than deliberate.

While not explicitly so, the book is divided into basically three sections. The first section talks about the commonly made mistakes when designing dashboards. Thirteen mistakes are listed, some obvious and most painfully common like "Exceeding the boundaries of a single screen", "Introducing meaningless variety", "Arranging the data poorly", "Choosing a deficient measure", "Highlighting important data ineffectively or not at all", etc... Each listed mistake is described, with one or several illustrative examples from actual dashboards, and with frequent suggested improvements or redesigns of the charts and graphs.

The second part, for my money, is perhaps the most useful section. This section is however only a single chapter - "Tapping Into the Power of Visual Perception", where the cognitive underpinnings of how we perceive and process information are laid out, of how short term memory works, of pre-attentive processing of information and how information can be laid out and formatted such that pre-attentive processing kicks in, which is much faster than attentive, deliberate processing. "In Information Visualization: Perception for Design, Colin Ware suggests that the preattentive attributes of visual perception can be organized into four categories: color, form, spatial position, and motion."

Those familiar with Few's writings will be well aware of his annoyance with pie charts and their uselessness in effectively conveying information. He writes, "The truth is, I never recommend the use of pie charts. The only thing they have going for them is the fact that everybody immediately knows when they see a pie chart that they are seeing parts of a whole (or ought to be). Beyond that, pie charts don't display quantitative data very effectively. As you'll see in Chapter 4, Tapping into the Power of Visual Perception, humans can't compare two-dimensional areas or angles very accurately--and these are the two means that pie charts use to encode quantitative data." ... "Humans tend to underestimate differences in 2-D areas, and hence you must be wary of using 2-D areas of different sizes to encode quantitative values--especially on a dashboard, where speed of interpretation is essential." He suggests that for most purposes bar charts are more effective.

The final three chapters deal with advice and examples of creating effective dashboards. His advice is that "The best way to condense a broad spectrum of information to fit onto a dashboard is in the form of summaries and exceptions", and to "eliminate all unnecessary non-data pixels", and to "de-emphasize and regularize the non-data pixels that remain."

Make no mistake - you are not going to become a pundit in effective dashboard and chart designing after reading this book. But this is an excellent start.

One note: this book lists for Rs 1800 or so on Landmark (, and does not seem to be widely available in India. So, try reading this book online from Safari (, or grab a limited preview of the book on Google Books.

Other links:
If this topic interests you, there are many, many more books to peruse:

Thursday, January 29, 2009

Mt St Helens

This photo is of Mt St Helens, taken in November or December 2000. It is one of my favourite photos, and was taken with a Canon A2 film SLR. The striking part of this photo is the snowline, which makes for a very stark contrast with the peak and the higher altitudes, bathed in pristine white snow, and the lower half, barren dried lava. One of the challenges in such photos is deciding on the exposure. Almost any camera's exposuring system is going to be challenged with such a composition. If you expose for the snow, the barren landscape in the bottom half is going to be too dark to make out any details. If you expose for the barren landscape, the snow is going to be too washed out white. If you try something of a compromise, well, nothing's going to be rightly exposed. But, there is NO right exposure. It's all about how you want the photo to come out. One obvious solution is to bracket. If you have a digital camera, you can see the results instantaneously, and adjust the exposure accordingly. Another solution would be to use a one or two-stop neutral density graduated filter such that the snow is darkened enough to retain details in both the snowed region as well as the landscape. Photoshop geeks would also argue that you simply shoot a RAW image, and then process it in Photoshop.

Saturday, January 17, 2009

When is the right time to do layoffs

It seems that the right time, if there ever is a right time, to do layoffs is during an economic boom, when your company and the industry as a whole is going great guns. Counter-intuitive, isn't it? Yes. Because the conventional thinking goes that it is during a downturn that you should do the layoffs, to reduce costs, to become leaner, and to focus. But I propose that layoffs when things are good is better. Think about it. Firstly, it is more humane, because the laid off employees stand a better chance of getting jobs when companies are hiring than when they are not. Second, and this is the delicious part, if you have conducted the layoffs based on some rational and logical process of weeding out the under-performers, then these people, when they do find a job, are likely to head to your competitors, right? That itself has two advantages. First, and this is the obvious one, your competitor with more subpar employees has to be good for you. Second, more subpar employees is going to tick off the super-smart people at your competitors, who will see this addition as proof that the company is heading towards trouble. So, they are more likely to want to look for alternatives. Knowing that **your** company laid off these under-perfomers, and that **their** company hired them is only going to make **your** company look smart. Eh?
But does it work this way? Nah. Of course not. Because boom times are a period of empire building, where quantity trumps quality, and where herd-mentality ensures that companies use the precious currency of an economic boom to squander the opportunity of focusing on and building world class competencies. Also, doing layoffs is considered a sign of trouble. Which hurts the stock price, which doesn't make executives very happy.
Doing layoffs, at any point, is a sure sign of having screwed up - you either hired the wrong people, or you hired more than you needed to - so by doing layoffs when everyone else is also doing so makes your culpability much mitigated, leaving the executives with a very well-preserved sense of self-esteem.

Reading - Jan 16 2009

Miscellaneous readings over the past week or so.

Wednesday, January 14, 2009

The Self-Destructive Habits of Good Companies

The Self-Destructive Habits of Good Companies, by Jagdish Sheth

You have certainly heard of the 7 habits of highly effective people, the bestseller by Covey, which spawned lots and lots of similar books like the 7 habits of highly effective teens, happy kids, families, and what not. So when you see a cover of a book titled 'The Self-Destructive Habits of Good Companies', it is sure to catch your attention. And guess what? The habits are seven in number. Just like the seven deadly sins. The author, Jagadish Sheth, lists seven habits that can afflict good companies. If left unchecked, these habits can destroy companies.

Dozens of companies are used as case studies here to highlight these seven 'self-destructing habits'.

Monday, January 5, 2009

The Corporation That Changed The World

The Corporation that Changed the World: How the East India Company Shaped the Modern Multinational

 (Amazon, Amazon India, Kindle IndiaFlipkart, my review on Amazon)

Orient Longman, the publisher, has this to say about the book:
The Corporation that Changed the World is a popular history of one of the world’s most famous companies.

Founded in 1600, the East India Company was the forerunner of the modern multinational. Starting life as a trader in Asian Spices, the Company ended its days running Britain’s Indian empire. In the process, it shocked its contemporaries with the scale of its violence, corruption and speculation.

This is the first-ever book to expose the truth about the Company’s social record. Robins reveals a hidden story of tragedy and intrigue. War, famine, stock-market bubbles and even duels between rival executives are all to be found in this new account. For Robins, the Company’s legacy provides compelling lessons on how to ensure the accountability of today’s global business.
The book and the author's contention is that the East India Company had a lot in common with the corporates of today, especially with the likes of Enron and Worldcom, and that a lack of appropriate corporate governance and weak oversight on the part of the government contributed to excesses therein. Specifically, "the drive for monopoly control, the speculative temptations of executives and investors, and the absence of automatic remedy for corporate abuse." [page 35]

The title may seem like hyperbole, but when you consider the impact that the Company Bahadur (which was the honorific bestowed on the company by the natives of the land it ruled - that would be us Indians) had on much of the world, including India and China, mostly for the worse, and mostly with tragic consequences, hyperbole does not seem like an exaggeration. This book documents the social impact of the Company's rule in India.

I as an Indians am not particularly inclined to be neutral towards the company, so to some this book comes as a validation of all that was worst about the company, its brutality, venality, avarice, machinations, and all. Sure, one can possibly find some good that the Company, and then the British rule, did for India, but when weighed against the costs that it extracted from India, the balance is certainly not in favor of the Company.

A must read.

In 1700, the GDP of Britain was $10.7 billion, representing 2.88% of world GDP. The respective figures for China were 82.8b, 22.3%, and for India 90.7b, 24.4%. By 1870, these had changed to $100b (9.1%) for Britain, $189b (17.2%) for China, and $134b (12.2%) for India. [page 7]

Disconnect and denial abounds in some quarters of the British aristocracy. The chief executive of the Standard Chartered Bank remarked that the challenge is now (in 2002) to "build upon the courageous, creative, and truly international legacy of the East India Company." [page 14]. "Rod Eddington, one time chief executive of British Airways" in a similar vein saw it "as a case study in how corporations succeed 'by dint of hard work, shrewdness and charm.' " [page 14, 15]. Yes, charm indeed. The charm of the likes of Robert Clive, a petty thief, who found himself dealing with another thief, Mir Jafar, and managed to come out ahead.

The author points out, correctly, that these "romantic interpretations ... fail to confront the costs associated with th Company's business practices." [page 15]

What contributed in no small part of the venality and the machinations of the Company employees in India was the fact that "... the Company's overseas staff received a minimal salary and the right to conduct private trade on their own account within Asia." [page 33].

As we know, Columbus was on the search for a route to India, to trade in spices. ("... so essential was pepper as a way of making preserved meat edible... " [page 41]) This demand was such that "In the two centuries after 1600, about one-third of the silver produce in American found its way to Asia to pay for Europe's imports." [page 41]

One of the earliest and principal architects of the Company's vision was Sir Josiah Child. "Throughout the 1680s. he was either governor (chairman) or deputy-governor." of the Company. [page 47]. "... he fervently believed profit and power must go together." [page 48] "On 9 June 1686, Child underlined the imperative for the Company to transform itself from 'a parcel of mere trading merchants' into a 'formidable martial government in India'." [page 49]

It is often remarked by modern 'experts' that India cannot become a manufacturing power. The lesson these self-styles experts take great pains to point out is that China is 'the' place for manufacturing, and that India should be thankful for whatever advantage it has in the area of software services and diamond polishing, and leave manufacturing to the Chinese. It is therefore with some amusement that one reads that "The Indian subcontinent was then the workshop of the world, accounting for almost a quarter of global manufacturing output in 1750." and even more so that "Even in the first century AD, the Roman historian Pliny was complaining that the extensive importing of cotton fabrics from India was draining Rome of gold." [page 61]

Such a leadership position in manufacturing should unsurprisingly lead to the assertion in the book that "Indeed, trading houses, such as those headed by Jagat Seth and Amir Chand, were often far richer and better connected than the Company." [page 65]

History textbooks tell us that the battles of Plassey and Buxar marked the beginning of the end of the supremacy of Indian trade, and the rise of the English loot of the subcontinent. "Plassey allowed the Company 'to carry on the whole trade of India (China excepted) for three years, without sending out one ounce of bullion'. The reversal of global economic eminence had begun." [page 74]

At the time, "there is compelling evidence that India's weavers had 'highher earnings than their British counterparts and lived lives of greater financial security.'" [page 77]. When one reads about the abject poverty that Indian artisans live and die in today, one can only weep. Here I would like to add that people should make it a point to read P Sainath's excellent book, "Everyone Loves A Good Drought", which is the anti-thesis of the pro-liberalization enthusiasts, and serves as a balanced counterpoint.

Exploitation is time invariant, as the author documents. "One practice that was particularly resented was the classification of perfectly good cloth as sub-standard (ferreted). ... According to William Bolt's celebrated account, 'various and innumerable' were the 'methods of oppressing the poor weavers, such as fines, imprisonments, floggings, forcing bonds on them, etc.' ... the Company's practices led to a shocking form of self-mutilation, stating that 'instances have been known of their cutting off their thumbs to prevent their being forced to wind silk.' " [pages 77, 78]

The corrupt and decadent lifestyles of the Company's leaders was such that "A new catchphrase entered the language - 'a lass and a lakh a day'." [page 83]

Expectedly, food shortages and famines were virtually unknown in India, prior to the Company's rule. "... Cornelius Wallard calculated that in the 120 years of British rule there had been 34 famines in India, compared with only 17 recorded famines in the entire previous two millennia." [page 90]. Do the math and you can see that before the advent of British rule, India saw roughly a famine every 115 years, while after it was a famine every three and a half years! Even accounting for famines that may not have been recorded in the previous two thousand years, the difference is still staggering.

However, famine led to riches for Company officers. "One junior executive accumulated over 60,000 pounds, as rice prices soared from 120 seers per rupee at the beginning of the famine to just three seers a rupee in June 1770." [page 91]
"Not only did the Company continue to collect its land revenue throughout the famine - instead of introducing some form of relief in the Mughal fashion - it actually increased the rate." [page 92]
"In 1772, Warren Hastings estimated that perhaps 10 million Bengalis had starved to death, equating to perhaps a third of the population." [page 92]. Mortality rates in some districts in Bengal, like Malda, reached 50%.
"... in December 1770, when the Gentleman's Magazine reported that 'provisions were so scarce in the Company's new acquisitions that parents brought their children to sell them for a loaf of bread.' " [page 95]

However, it's not as if life for the Company employees was a bed of roses. "... for the Company a round trip from London to India and back could take up to two years. ... Over half its employees posted to Asia died while in service." [page 27] Yet such was the lure of India's riches that there was a steady stream of Englishmen making their way to India to make their fortunes.
The book lists for Rs 330, and you can pick this book from most boostores in India. Some may also offer a 10-20% discount.