The Back of the Napkin Part 8

Two combination frameworks appear often enough in problem- solving pictures that we're going to look at them in detail over the coming pages.

CHAPTER 8.

SHOWING AND THE VISUAL THINKING MBA.

Ladies and Gentlemen, Start Your Pens After we've identified our problem, selected the appropriate showing framework, and further focused our ideas using the SQVID, the next step is to put pen to paper (or napkin or whiteboard) and start drawing. There are two ways we can look at what we're about to do. If we're a Black Pen person, it's going to be the easiest thing in the world; if we're a Red Pen person, it's going to be impossible, and there's no way we'll produce anything worth showing anybody. Both views are wrong. Drawing our picture is going to be harder than expected for the artistically gifted among us (because we'll be forcing our brains into potentially unfamiliar a.n.a.lytic processes); and it's going to be easier than expected for the "I'm not visual" crowd (because we'll be taking unexpected advantage of a.n.a.lytic capabilities we use all the time). The important thing to keep in mind at this point is that we already know what to do. We looked well, we saw clearly, we imagined confidently-we've even got our starting framework selected.

Here's how this is going to work: Since each framework requires a different way of approaching a drawing, we're going to run through an example or two of each. That is plenty enough to cover everything we've talked about in the book so far, but nowhere near enough to cover every problem we might face in the world. But that's the real beauty of visual thinking. It doesn't take many pictures to see how just a few frameworks and rules make any problem easy to picture.

The Visual Thinking MBA: Putting It All to Work In business school, MBA students and executives rely on case studies to put into practice the theories of finance, operations, marketing, and management that they've learned in the cla.s.sroom. Whether based on actual companies facing historically accurate business challenges or hypothetical situations featuring fictional businesses, the case studies are the backbone of MBA programs because they make abstract ideas "real." In part III, we're going to take the same approach. By walking through a detailed case study, we're going to make the tools and rules of visual thinking come alive.

Using a fict.i.tious software company in crisis as a backdrop, we're going to put everything we've discussed into play: the visual thinking process, the SQVID, the <6><6> model, and the codex. To really show how effective visual thinking can be in understanding a complex business problem, we're going to use these tools to create pictures covering everything we would see in a business school seminar. Starting with customer research, we'll then move through marketing and product development, financial a.n.a.lysis, project planning, and finally strategic decision making. In short, there's going to be a lot to look at.

As with any rigorous case study, there are two ways to approach this: either as a top-level scan or as a detailed deep dive. To help readers who want to make a quick scan, this case study is broken into six chapters, one showcasing each of the six visual frameworks. If you're mainly interested in the frameworks themselves, read just the first two or three summary pages of each chapter-you'll still get a great sense of the overall business story.

If you're interested in following the entire line of reasoning in detail, start from the beginning. As you work your way through, you'll notice that each picture is created step by step over a series of frames-almost like a stop-action animation-to help you see exactly how each is composed. Either way-scan or deep dive-this is where solving business problems with pictures becomes real.

The Case Study Scenario Imagine that we work for an accounting software company called Super Accounting Exchange Incorporated, or SAX Inc. SAX has been designing and selling specialized accounting software for use by large organizations since 1996, and although SAX isn't a very big company, our flags.h.i.+p product has been an industry benchmark for nearly a decade. In our niche industry there are presently five main compet.i.tors, all with their own approaches to the business and all with their own strengths and weaknesses. The five are: SAX Inc. (That's us) SMSoft Inc.

Peridocs Incorporated Univerce LLC MoneyFree So here's the problem: For the past two years our sales have gone flat while sales at the other companies have continued to rise. Our latest product release a year ago introduced many new features, making our software the most feature-rich available, but our customers' reception has been lukewarm. Our sales reps complain that they're having an increasingly hard time selling our expensive software, given the rise of "open-source freeware" over the past year. Such freeware-typically created by loosely affiliated developers unenc.u.mbered by the overhead costs and shareholder demands seen in a bigger business like ours-is making increasing inroads into the technology industry everywhere. So far no open-source freeware comes close to our feature set, but that won't last forever. We don't know exactly what we need to do before we lose significant market share, but we know we have to do something. So let's move on to chapter 9 and start at the beginning, with our customers.

Starting with a basic who problem at SAX Inc., we're going to run through all six frameworks, creating several pictures that take us from defining the problem to arriving at a solution.

A NOTE ON THE PICTURES WE'LL BE CREATING

Before we get started, it's worth revisiting an earlier comment about the images in this book. Everything we're about to create is intended to be drawn by hand: on a whiteboard, on a yellow pad, on the back of a napkin, on whatever drawing surface you might have in front of you. In the introduction I said that Daphne's strategy chart was the first and last picture in the book to be created on a computer, and that remains true. While computers are insanely wonderful tools for countless applications, I can't think of anything that they add to visual thinking at this level-while I can think of several things they take away. In fact, because using a computer seems to mask a number of our basic cognitive tasks-especially the unexpected ideas that emerge when we put pen to paper-relying on computers at this stage is more likely to undermine our visual thinking abilities than to advance them.

On the plus side, it's also true that computers make the composition and finis.h.i.+ng of the more advanced pictures infinitely easier than anything we can do by hand, are essential for creating accurate quant.i.tative images, and are irreplaceable presentation and communication tools. Those points are all not trivial. That's why appendix C is included: It addresses which software I find most useful for further developing each framework, and introduces a few simple software tricks that will be helpful if you decide to go the entirely digital (and I don't mean fingers) route.

But for now, let's stay with pens and napkins: It's good practice for the next time we meet someone interesting at an airport bar.

CHAPTER 9.

WHO ARE OUR CUSTOMERS?.

PICTURES THAT SOLVE A WHO/WHAT PROBLEM.

The Customer Crisis We all agree: We don't know our customers as well as we should anymore, and in order to figure out which customers to go out and talk to, we need to create a portrait of who we think they are. Let's pick large client company and use what we know about it to create a sample baseline customer profile. We know that our baseline will contain a lot of information, that we'll want to be able to look at it from many different angles, and that we'll share it inside and outside our company, so it makes sense to create a picture.

We already know how to pick the right framework: Look it up on the Visual Thinking Codex. In this case our problem is about people (who our customers are), so the codex tells us to start with a portrait, or qualitative representation.

Recall that the first way of seeing was who and what, meaning that we saw objects that we recognized because of distinct visual qualities: their components, shape, proportion, size, color, texture, etc. To show to others what we saw, we create a portrait (or qualitative representation) that represents the most evident of those qualities, emphasizing especially those that made our object visually distinct from others. While portraits don't show how many of something there are, where they are, or when and how they interact-all of which are addressed by the other specific frameworks-they do provide the starting point by helping us identify and keep track of who is who and what is what.

Portraits: General Rules of Thumb Think simple. The goal isn't to be Rembrandt. In fact, an overly elaborate or cute picture inevitably draws too much attention to itself and distracts from the essence of the idea to be conveyed. The simpler, the better: Think visually telegraphing an idea rather than painting the whole picture .

Illuminate lists. The purpose of creating a business portrait is to trigger the unexpected qualitative ideas that emerge when the hands and the mind's eye work together. Visually representing someone or something (regardless of actual likeness or detail) always triggers insights that writing a list alone cannot achieve.

Visually describe. When time is limited (and in business, time is always limited), pictures always make for better comparisons than verbal descriptions. Comparative portraits can be as simple as a series of smiley faces. Adding even that thin a visual aspect brings objects to life and makes them memorable.

Renderings, profiles, plans, elevations, diagrams: There are lots of kinds of portraits, but all show the same things-the recognizable qualities that differentiate objects.

Creating even the simplest of portraits engages the mind's eye.

Even the sparest of portraits make comparisons come alive.

With these ideas in mind, let's go back to our customer portrait. With our framework selected, we then look across the SQVID, answering its five questions as we go.

Simple or elaborate? Given that this is our first effort at visually portraying our customers, we'd be better off with something simple. Qualitative or quant.i.tative? For now, this is just a portrait, not a numeric representation, so by default it will be qualitative. Vision or execution? As a baseline, we're not yet talking about where we'd like to go or how to get there, so that question doesn't matter for this picture; let's skip it. Individual or comparison? Since we'll be looking across the whole range of customers, this will be a comparison. Change or as is? Since we're hoping to see the baseline, our picture will be as is for now, although depending on what we find, we may want to show change at some point. Summing up, this is a pretty simple starting framework-a simple, qualitative portrait of a few customer types, something like this: ???. Now we're finally ready to draw.

What to start with? Before thinking too hard, it's helpful to know that although the first mark on the napkin is the most difficult to make, it is also among the least important. We'll be adding to it, altering it, and possibly erasing it entirely. It's more important that we get something down on paper than worry too much about what it is. A good way to start any picture is to draw a circle and give it a name. Since we've already agreed that we don't know our customers as well as we should, let's start with something we do know-us.

Let's start with a simple circle and then give it a name.

Since a portrait is intended to help us identify one object from another, let's add something visual to make "us" more distinctly us-our building, for example.

Remember this is a portrait, so let's add our building to make us more recognizable.

Does seeing ourselves portrayed this way trigger any ideas about how to show our sample client? How about we add them in the same way?

We add in our client and already we've got a good picture going.

Even this spare picture starts to show us something about the relations.h.i.+p between us and our client, and helps our mind's eye begin imagining ways to create a portrait of our customers.

So if we're going to be showing people, why don't we again start with our own? That won't tell us anything about our customers, but drawing us (who we know so well) will get us in the right frame of mind for thinking about them.

We draw in the people of our own company: the boss, the account reps, team leads, and software developers.

That's us. All those smiley faces we talked about are starting to appear. Loosened up by drawing ourselves in, we're finally ready to sketch in our customers.

We draw in the customers that our people sell to: our client's execs, sales teams, accountants, and technical folks.

There they are: our customers. Interesting. There are more types than we might have initially thought. Just creating a portrait like this has already started us thinking about customers in different ways. So far we've spent just a couple minutes with this picture, yet we've already created a baseline portrait of who's who in our business and have triggered many new ideas simply because we drew it. There's only one more thing we've got to do before we start making copies: Label everything.

We've instinctively been giving names to the shapes as we've drawn them in. In fact, right from the beginning our task was to give a name to our first circle. As we added more people we kept labeling them, too. For good reason: While our brain's visual centers are happy to have pictures to look at, other mental processing areas demand names, and if they're not written there, we're going to make the names up ourselves. It's always better to be proactive about labeling and leave no doubt about what we're showing.

We also always need to give our pictures a t.i.tle. While it should be completely clear to us what we just drew, it always pays to a.s.sume that someone else is going to approach our picture from a different perspective, perhaps completely missing the point we intended to make. So as a rule, spell it out right on the top, every time.

By adding a t.i.tle we know that we're being clear about what we're showing to anyone who sees our picture.

Simple as it is, this picture is useful as a backbone for mapping in other qualitative traits about our customers. We know from previous market studies, for example, that each of these customer types wants something different from accounting software. Client executives are ultimately responsible for anything good (or bad) that happens whenever our software gets used, so they want a product that is easily accessible to their own people and impenetrable to anyone else. Above all, execs want security. Sales teams want a product that makes it easy for them to sell their company's services, so they want software with a good reputation-they want a salable brand. Accountants want accuracy and stability; they want reliability. And technicians want software that is easy to connect to other systems and easy to update, they want flexibility. That's along list of wants, just the kind of thing more easily digested in a picture.

Adding what they want.

Now we have two portraits of our customers, one showing who they are, and another showing what they want. These are just two of many versions we can make. In different businesses and different contexts, similar pictures might be called renderings, plans, diagrams, or elevations, but all do essentially the same thing: They provide a visual record of what something looks like, the who and what that we see.

CHAPTER 10.

HOW MANY ARE BUYING?.

PICTURES THAT SOLVE A HOW MUCH PROBLEM.

The Customer Crisis, Now with Numbers We've seen our customers, noted some of their distinctions, and even begun thinking about what they might want from our company's software. That's good information that will be useful for helping us get sales moving again, but it's only a start. To be meaningful, we're going to need to know how many of each of those customers we have, quantify how much they're willing to spend on products like ours, and even try to numerically measure how they feel about us and our products.

We're not talking about who and what anymore. Now we need to see how much. The Visual Thinking Codex tells us that we're going to s.h.i.+ft to charts now-pictures that show quant.i.ties, ill.u.s.trate measurable criteria, and represent numeric comparisons. Unlike portraits, which we could create without any specific quant.i.tative information, charts demand numbers, measures, and data.

REVIEW: A CHART SHOWS HOW MUCH.

Pie charts, bar charts, numeric comparisons, histograms: There are countless ways of representing how much, but they are all variations on the same theme-providing a visual measurement of quant.i.ty.

After who and what, we next saw how much or how many objects there were. For smallish numbers, our minds did a quick count; for slightly larger quant.i.ties, we made rough estimates; for large quant.i.ties we just said to ourselves, "A lot." To show these numbers to others, we use a chart (or quant.i.tative representation) in which we turn abstract numbers into visually concrete pictures of amounts.

Charts: General Rules of Thumb 1. It's the data that matters, so let it show. Many people find numbers boring, so we jazz up our charts with visual bells and whistles hoping to make the pictures look more interesting. Three thoughts: First, insightful data is never boring. If what we're showing resonates with our audience (either because it shows exactly what they hoped for or surprises the daylights out of them), they won't fall asleep. Second, we should always show the fewest possible pictures to make our point. Either limit the number of one-point pictures we show or combine as many data points as possible into one or two multiple-variable plots (more on those later). And third, the addition of low-key anthropomorphic elements ??? where appropriate does add cognitive engagement. In other words, if you're counting people, go ahead and show the people.

2. Pick the simplest model to make your point. This year's version of the most popular spreadsheet software* includes ninety-nine different charting choices right out of the box. No wonder we're confused about which chart to use. The fact is, it only looks like it has ninety-nine. In actuality, it has four-bars, lines, pies, and bubbles. Everything else is a jazzed-up version of one of those. If we think of those four types like this, we shouldn't have any trouble picking the right one.

Bars: For comparing absolute quant.i.ties of something (1,000 apples versus 800 oranges versus 120 pears).

Lines and areas: For comparing absolute quant.i.ties between two different criteria or times (pies have 1,000 apples, 0 oranges, and 60 pears while tarts have 0 apples, 800 oranges, and 60 pears). (We'll look at times series charts in when frameworks, chapter 12.) Pies: For comparing relative quant.i.ties of something (52 percent apples, 42 percent oranges, and 6 percent pears).

Bubbles: For comparing more than two variables (which we'll look at when we come to why frameworks, chapter 14).

3. If you start with one model, stay with one model. If our chart has the right coordinate system to convey our data and is built with precognitive attributes, our audiences should get it in no time. Nevertheless, once they've "learned" to read our first chart, don't jar their "seeing settings" by suddenly flopping an axis, changing the chart type, or introducing a wildly divergent way of thinking. Think of showing a series of charts as a drive through a beautiful landscape: Gentle or expected transitions are pleasant; suddenly flying out over a cliff is not.

Back to SAX Inc. While making our customer portrait, we collected the who data; now we need some how many numbers. As we look at our company's sales records, it turns out that we have those numbers after all. Since "job t.i.tle" is one of the fields in our software registration questionnaire, we have a record of how many customers of each type we have. If we were to create a picture showing both customers and quant.i.ties, it might look something like this: Our sales data tells us exactly how many customers we have.

Numerically speaking, this picture couldn't be any more accurate: It's as if we had all our customers stand in their parking lot and took a photo. But accuracy aside, there are major problems here: First, although we can pick out individual types, we can't see the groups (since they're all mixed together). Second, it's almost impossible to count. We can see quant.i.ty, but we can't be precise or do any math with it. So let's straighten out the coordinates and add summary numbers.

Same image, now with numbers and coordinates added.

Much better. Shown this way, we can rank and compare each customer type instantly. We immediately see that there are a lot more accountants than salespeople, about half as many technicians as salespeople, and only a few execs. Still, it's a hard picture to draw. What we really need is a simpler way of showing those quant.i.ties without having to draw every one of the people. Let's try something: How about we get rid of the picture altogether and just show the numbers?

That also gives us numeric accuracy, but loses all the pictorial immediacy-it now takes our mind a few seconds to dance back and forth between the rows and columns in order to see how the customer numbers compare. A table also doesn't provide any hooks to catch our visual memory. If we can't remember the precise numbers, we've got no larger context to fall back upon. What we need is a hybrid, something that combines the best of both pictures. What about a bar chart?

We could get rid of the picture altogether and subst.i.tute a table.

A bar chart helps us see the pictures and the numbers.

We use a pie chart to show quant.i.ties relative to the whole.

There we go. Easy to see who we're talking about and how many of each, plus we've got the numbers right there-we've even got precognitive quant.i.ty bars for our eyes to read immediately, compare, and viscerally recall long after we've forgotten the numbers: "I don't remember exactly how many, but I know there were a lot more accountants than salespeople." Perfect. If we need to see precisely how many in total there are of something, a simple bar chart is the way to go.

Seeing exactly how many customers we have is only part of the equation. What we really need to know is how many execs we sell to relative to accountants relative to salespeople. That's how we'll figure out who is most important to focus on given our fixed marketing budgets. If we've got only one pie's worth of marketing budget available, for example, we need to know who should get the biggest slice.

That's why we use a pie chart when we need to see percentages relative to the whole.

We don't see the total numbers anymore; instead we see how many of one customer type we have relative to others. If all customers were equally likely to buy our software, we'd want to divide up our marketing budget according to these same percentages. That way we'd know we were spreading out our marketing dollars evenly among all customers.