In Futures Thinking: The Basics , I offered up an overview of how to engage in a foresight exercise. In Futures Thinking: Asking the Question , I explored in more detail the process of setting up a futures exercise, and how to figure out what you're trying to figure out. In Futures Thinking: Scanning the World , I took a look at gathering useful data. In Futures Thinking: Mapping the Possibilities (Part 1) , I gave a broad overview of creating alternative scenarios. In Futures Thinking: Mapping the Possibilities (Part 2) , I moved to the nuts & bolts of creating scenarios.
But that didn't really tell you what scenarios actually look like. So I'm going to rectify that.
In 2008, the San Francisco-based user experience design firm Adaptive Path  was asked to create some prototype designs of what the Firefox Web browser of the year 2020 might look like. Adaptive Path, in turn, asked me to help them  think through what the Internet and the world of 2020 might look like, so that they would have a better sense of how a future Firefox might be used. This is exactly the kind of task that scenario work is well-suited for, so I suggested that rather than give them a single Vision of Tomorrow, I'd help them see a small set of alternatives. They agreed.
I brought together some folks from Adaptive Path and from my own network, and had a two-day brainstorming and scenario-design session. Three scenarios resulted from the workshop--that is, three overarching scenario concepts, supported by lots of bullet points and sticky notes, all in roughly chronological order, resulted. I then took these results and turned them into narrative scenarios. Adaptive Path used these narrative scenarios as inspiration and "future reality" checks for their own design scenarios, presented on video. (See the Adaptive Path "Aurora" videos  here.)
But in creating my three scenarios, I took an unusual turn: I decided that I'd write each of the three scenarios in a different scenario style. That made it harder to compare the three, but it meant that each would speak to audiences in differing ways, so that readers who found one style unpalatable might find another style much more to their liking.
You can download a PDF of all three scenarios here ; it's actually done as a Creative Commons licensed-work (Non-Commercial/Attribution/Share-Alike), so feel free to play with these scenarios if you so desire.
The three styles I used for these scenarios can be categorized as "Scenario-as-Story," "Scenario-as-Recollection," and "Scenario-as-History."
In Scenario-as-Story, the presentation is similar to that of a work of fiction. Named characters operate in a lightweight plot, but in doing so engage in behaviors that display key aspects of the scenario. Here's a sample:
< p>The narrative continues from there. The advantage of the Scenario-as-Story approach is that fiction is a familiar presentation language for readers, and they can more readily grasp the changes to one's life that emerge from the scenario. A story model lets you describe some of the more nuanced aspects of a scenaric future. The disadvantage is that, generally speaking, scenarios are lousy fiction. Even the best-written scenario stories generally wouldn't pass muster with a fiction editor. A more difficult problem stems from differing views on human behavior--if the character in the scenario story does something off-putting or inexplicable, the reader will find it harder to accept the rest of the scenario.
Luisa Martinez stretches, trying to wake up her muscles. It's cold, it's dark, and it's too damn early to be doing this. Grabbing her run pack--her specs, her tablet, and a Nalgene bottle of water--she heads out into the San Francisco morning.
On the sidewalk, she turns on the Web tablet and puts on the specs. They're new; she got them as a gift from her brother, who loves the latest gadgets. They're supposed to work with the new networks that the City is installing (yet another Google project, she recalls), as well as provide a new way of controlling her tablet (a pocket-size device replacing an old iPod she recycled a month or three ago).
She sees a slowly opening flower off to the side of her vision as the system comes online, a rose. She smiles; Diego must have done some customization before he sent it. A text crawl, just below her eye-line, asks her to confirm if she wants to use her "Jogging" profile from the tablet. She pauses--was this a voice interface?--and remembers that the specs have one of those accelawhatchamacallits in it to respond to movement. Hesitantly, she nods, the display clears, and she sets off.
In Scenario-as-Recollection, the scenario narrative remains personal (usually done as a first-person perspective), but the structure is more linear and straightforward, with no pretense of a plot. A sample:
The biggest user-side advance of the early 2010s came in hardware. The ZCard was the first in the line of MMDs--multiuse mobile devices--and it caused quite a stir when it hit. A terabyte of built-in storage, camera, multitouch, 4G wireless, in an ID card-size device. It worked like a phone or Web pad on its own, or could serve as the core of a laptop for people needing a bigger display and better interface. It really changed everything for a lot of us; you probably have one or two on you right now.
The funny thing is, it wasn't as much of a paradigm shift as its enthusiasts claimed, at least not initially. It combined a few devices, and the modular system was cool, but it was otherwise just the latest version of a technology path that had been going on for a decade. It was only when decent wireless broadband had become commonplace--2012 or 2013 or so--that we got a chance to try something new. A few vendors started putting out ZCards that offloaded processing and deep storage to the Internet, making the cards little more than extremely smart (and small) network terminals.
Some of us old-timers shook our heads--how many companies would refuse to learn the lessons that network computers never succeed?-- only to discover that this time the model worked.
The advantage of this approach is that you can easily add a bit of subjectivity to the scenario without making it all about the speaker. The reader can come away from the piece understanding that opinions may actually vary about some aspects of this world, just like in the real world. The disadvantage of this approach is that it's harder to talk about big-picture changes to the world, with references to underlying drivers and inflection points and such, without the scenario sounding like a lecture.
One useful variation of this model is to write in the style of a newsmagazine (The Economist is what I usually shoot for) analysis piece. Still subjective, still descriptive, but less personal.
Finally, in Scenario-as-History, the tone is more dry, not at all personal, and focused on the chronology. A sample:
2014: Language is a Virus
As the reputation network advocates slowly begin to rebuild their systems--using the identity management tools as a core--a new technology gives the reputation model a shock from an entirely unexpected direction.
Samsung releases software allowing for real-time, highly accurate (over 99%), context-sensitive language translation. Initially embedded in their phones, Samsung offers to license the code at extremely generous rates (sometimes even for free), in the name of promoting international harmony. The result isn't quite what they expect; it turns out that what was needed along with language translation was some kind of cultural translation.
This hits the reputation network folks hard, because reputation is a highly culturally-situated concept. Behaviors that would be neutral or even commendable in some locations can be disdained or reprehensible in others.
This approach allows for a much higher-level perspective on the scenario, potentially giving the reader a deeper understanding of how things connect. The narrator can make explicit the chain of events leading to the final scenario year. Conversely, the scenario-as-history approach tends to focus more on the "how we got there" than on the "what is it now like." While this can be useful for seeing the connection between the present and the scenaric future, it does lessen the potential for showing just how disruptive and transformed the world of the scenario may be.
I use all of these approaches, at different times, and won't say that one is intrinsically better than the others. That said, the second one--Scenario-as-Recollection--probably offers the best balance of subjectivity and chronology, and makes it easy to do a basic overview of the world (including how people react). This approach, typically in the variant model I mentioned, is probably my go-to method for an audience I don't know well.
Above all, though, the goal of a scenario narrative is to get across a set of ideas about a changed world that allows the reader a new perspective on change and on choices yet to be made. Your goal, as a scenario writer, is to make sure that what you think they need to know is what comes across in the scenario.