Coming Up Short

Due to the diminutive size of the taxi, my three-person party needed to split between the front and back seats. I had the address of our destination—a spa somewhere outside walking distance—and was volunteered for the front.

The driver was older and nodded at the screen of my phone. Looking at the three of us he quickly determined we were Americans and pulled out his cellphone.

He spoke rapidly in Korean into the phone and then held it up as I waited for it to provide an English translation. After a moment of waiting, the phone began to speak in what sounded more like a riddle. A string of words that were English, to the credit of the app, but made little sense. Something about water and magic, but in broken English—missing all articles and missing any emotion that might suggest context. I smiled, nodding, as the driver laughed and threw the taxi into drive.

In Seoul the traffic is something unlike anything I’d seen in other cities more famed for their traffic like Toronto or Boston. As our group slowly crept across the dense city, the driver and I continued attempts at conversation.

The app’s English was as good as my Korean. And the only true outcome was entertaining my fellow travelers as they documented the encounter.

A day or two later, my friend and I laughed at the story as we finished our drinks and decided to head back to our hotel.

The bartender, the only other person in the subterranean bar, handed us the tab and we quickly handed over 20,000 won. The man walked over to the till and then returned without using the machine at all.

“No.”

His English was better than most in a city where I was lucky to find anyone who could speak it. But explaining what he meant by no was beyond his grasp of the language.

We attempted to resolve the unknown problem by indicating I was trying to pay for both of us, unsure if the tab I’d been given was for only half. I suggested I would be willing to tip with the change I should get. I offered a different note and that’s when the bartender pointed excitedly at my wallet.

“Yes.”

Holding all three bills together, I quickly realized my mistake. Instead of the two 10,000-won notes I intended to hand the bartender, I had given him a 10,000 and a 1,000 note—trying to stiff him 9,000 won on the drinks while happily telling him to keep the change.

Embarrassed, I handed over all of the money as a tip. This is when his command of English returned.

“No tips.”

As a compromise we stayed for a few more drinks as he told us about his favorite bars in the area and laughed at my attempts to pronounce the names of Korean neighborhoods we wanted to visit.

Sitting across from the small marketing team, I thumbed through the almost phonebook-sized stack of documents about atypical neurodevelopment in children I had been given upon our arrival. The people sitting across from me could answer almost any question about brain development, social learning and all other topics covered in these pieces of paper. However, a close-up or a dolly shot would elicit blank looks.

What they knew about atypical neurodevelopment could literally fill a book. What they knew about production was they had been given a grant to produce a video series and my company had won the bid.

Anything technical I could say would sound just like the riddles I tried to decipher on the way to the Korean spa.

Design challenges and suggested uses of lessons from tutorials gave me no shortage of ideas to use the new Storyline software I’d downloaded and just finished installing. Colorful examples that taught everything from making a perfect cup of coffee to performing technical calculations.

This is my favorite part of any creative project: the beginning. That open-ended, anything-goes, let’s see what sticks time known as the brainstorm. The first ideas were the obvious ones: an onboarding module or a how-to lesson. However, I wouldn’t have any employees to onboard anytime soon, so that idea made little sense as a practical, first project. Also, the exhausted landscape of YouTube tutorials and how-to videos was so bloated the second didn’t even slow down and attempt to take root.

A few days of this thinking started me asking another question—what about a tool that asked questions about a creative project?

But not technical questions.

Rather, questions that got to the idea of a client’s project. The ‘what do you see when you close your eyes?’ questions that don’t tell me one technical thing. But, reveal multitudes of more important information.

The rough draft of the idea began to take shape in my head: a user would answer a few quick questions and then be shown a few suggested types of content those answers described in the form of a simple graphic with a name and short description.

How many content types? How many questions? How would I come up with the weighting to make sure one answer that applied to multiple content types was tallied as such? I won’t say there aren’t tutorials on Articulate’s website that could answer all of those questions and probably several more along the way. But what I needed was someone to converse with about the idea and help with the heavy lifting when it came up.

That’s where Claude came in. How I use the system—bringing ideas and asking for help disassembling them and stress-testing them to make things better—offers that feeling of conversational brainstorming. I’m not starting the ideation with Claude and having him tell me what I’m building. I’m improving my designs by involving a collaborative partner who knows all the things I didn’t know I didn’t know.

I worked on an extensive document that described everything I saw when I closed my eyes and thought about the finished tool. No more than three or four responses per question; nothing technical; no more than five or six questions; and a graphic-centric layout for the results.

Starting with the questions, I spent time replaying numerous discovery meetings I’d run as a video producer, the questions I would ask that would help produce a script or an outline for the direction of a project. Every client had specific desires they could name, but I needed them to spend time in the abstract to make sure what was imagined was what was delivered.

Was I going to hand them what had been asked for or was I going to come up short and confused?

Claude not only helped with my needlessly extensive copywriting but also identified themes within the questions that would help with the score weighting and other needs down the road. Claude helped identify a possible issue with the number of questions I’d planned—someone could retake the survey, answer differently and still land on the exact same result.

I had initially come up with a paltry list of five content types. With the issue identified, we approached the problem not by inventing from whole cloth new content types but by examining how much work each name was doing. The updated list grew to nine types, each with names more playful than my initial technical names. Thankfully my partner is literally math pretending to be something else, so Claude was able to perform the calculations needed to validate the scoring model and provide a spreadsheet that contained all of the score weighting I would need to build the triggers in Storyline.

The triggers would be adjusting variables tied to each content type and adding anywhere from one to three points to any of those variables each time a user selected a response. In the end those content type variables are checked, rechecked and rechecked one final time to display the top three content types, in order.

I went through the questions and put myself into the shoes of a client describing a video. I wouldn’t say ‘two-shot interviews’ or ‘cinematic motion graphics,’ I’d probably go with ‘people talking to the audience’ or ‘an uplifting intro that gets people excited.’ With the starting point of the questions decided and the finishing point of the content types finalized, I needed to figure out how a user got from point A to point B.

The idea of sitting and coming up with a list that covered everything, omitted nothing and contained as few repeats via synonyms as possible, reminded me of scenes from Mad Men when Don Draper would have Peggy write up fifty ideas by lunch. The show would’ve been horrible, but I think Peggy would’ve had better things to say about Sterling Cooper if she had access to someone who could combine a thesaurus, calculator and Draper-strict copywriter into one writing partner.

I call him Claude, and he turned writing the responses from a character arc to an afternoon task.

I was able to ask for a list of answers to the questions without needing to repeat or further explain the ideas that needed to be present, restate the connections between question and result or specify a word count.

What I got back had a sense of understanding the ideas that needed to be connected from the questions through the user responses leading to meaningful results. We were speaking the same language now because of time spent understanding the ideas being communicated.

I am very much in the camp of visual learners. For every new skill or piece of software I’ve taught myself, there has been a huge helping hand from YouTube and any other source of video content I can get my hands on.

With the help of Articulate’s E-Learning Heroes community, I was able to map out what I felt was a well reasoned roadmap for the project. There was only one additional thing I would need to figure out before work could commence.

I had poked around Storyline while following the tutorials and saw the ability to trigger JavaScript. This seemed like a sensible way to reduce the number of triggers I would need to compile the final ranking by checking the variables. I was sure there would be a JavaScript that could check the values of all the variables and then communicate that ranking to another set of variables that would display the correct graphic.

I turned to a developer friend for the specifics of executing my scheme. And found more confused looks than answers. My inability to answer a single follow-up question told me the conversation was over before I’d even brought up the topic.

Sure of my thinking and the existence of this special bit of computer trickery, I went back to Claude. Over a few minutes we discussed my idea and how I felt it was going to improve the project by making it come together more quickly and with the possible removal of dozens of triggers I had deemed unnecessary. By the end of the conversation I had my answer—the elusive JavaScript—and something that ended up being much more helpful: an explanation.

This was a bad idea.

What I had pictured as cleaning up a process was actually going to add an additional step without removing the need for the Storyline triggers. My developer friend wasn’t able to answer my question, not because there wasn’t an answer, but because my language gap was too big on either side of the conversation. The words he understood like JavaScript and computer were nullified as soon as I started adding qualifiers and other attempts to get creative.

What I needed was a system that could translate my gibberish into a workable question, then translate the answer to that question back into something closer to Tyler-ese that could close that gap.

What came out the other side is the tool at the top of the page: a handful of plain-language questions, nine content types and three ranked recommendations at the end. Take it for a spin. The results make for pretty good conversation.