Thursday, April 11, 2013

The randomizer dropped me on Thursday, April 11, 2013, which at first looked like one of those dates where the archive was going to make me work for it. Twitter had some ordinary-life stuff. I was at Burgermaster around lunchtime. Later I was “Suburbaning with Heidi” at Fred Meyer. I made some jokes with Jen Zug about volcanoes and magnets that are now missing enough context to be completely useless. I tweeted a picture with the caption, “The dog did it. She’s a good dog. She’s a good dog. She’s a good dog.” The photo itself is gone, so I’m sure that was hilarious. The day before I had gone to see Jurassic Park in 3-D at Cinebarre and declared that I felt thirteen again. The day after this I would be complaining about not eating lunch and dealing with doctors and labs and the ongoing aftermath of back surgery. In other words, April 11 looked like a fairly normal day in the middle of a fairly complicated stretch of my life, until one tweet at 11:08 in the morning caught my attention. It was a retweet from Bootstrapper Studios: “A few BTS pics from a fantastic day on the set of @pipelinedeals new commercial.” And suddenly I knew exactly where the archaeology needed to go.

This was Chaos.

Or, as I now think of it, The Project That Would Not Fucking End.

Five days before this random date, on Saturday, April 6, Bootstrapper Studios had finally shot a commercial for PipelineDeals that had been kicking around our lives for so long that by the time I got onto the set I was absolutely itching to make the thing. I directed it. Cindy Lemos played the boss at the center of this increasingly ridiculous office. Tyler Kalberg was my DP. Jory Randall was assisting me and took charge of all the extras moving around in the background. We had somewhere around eighteen extras, a Steadicam, an office full of carefully manufactured disorder, a Nerf gun, a mustachioed man with a coffee mug, potato chips, an apple, a lot of moving bodies, a lot of business, and a concept that had started as something much more ambitious than we could actually afford. On April 11, I was already in the first rough-cut stage. At 9:14 that morning I emailed JP Werlin, Nic Cary, and Jory a YouTube link with the subject line ROUGH CUT and explained that this was a very early version where we were only looking at continuity and pacing. Audio, music, sound effects, color, all of that would come later. I had no idea when I wrote that sentence how much “later” there was going to be.

The project had actually begun the previous summer. On July 5, 2012, I sent PipelineDeals an estimate for a brand video that included our StoryLab process. By July 11 they had decided to move forward, and on July 12 I invoiced them for the first half. Alex Hendler was the writer I was working with, and this was also my first experience really figuring out what it meant to direct something somebody else was writing. Our first round of concepts went out in July and, to be completely honest, I didn’t like them enough. They weren’t bad ideas. They just weren’t grabbing me. One involved a Magic 8 Ball, another had a football thing going on, and I remember writing to Alex that I wanted to “pull the brake” on a couple of them, and specifically making sure he knew it wasn’t because they weren’t my ideas. Pipeline was spending real money and they were going to use this thing for a while. I wanted to get it right. I wanted it to be funny instead of feeling like another corporate metaphor with people standing around pretending they were inspired by software. And more than anything, I wanted to make something that felt cinematic.

I sold Alex on the visual idea using the opening of season one, episode one of The West Wing. That was the thing in my head: movement, choreography, the camera traveling through a working environment while the story unfolds around it. Alex immediately understood what I was reaching for and came back with another reference: Goodfellas. That was it. That was the moment I knew we were speaking the same visual language. He was a storyteller, much more naturally than I was, so when we eventually went in together to present the concepts to JP and Nic I have the feeling I mostly let Alex go. I don’t remember who said exactly what in the room, and I don’t want to invent a pitch meeting twenty-three years later just because it would make a cleaner story, but I know the lineage. I brought Alex the West Wing idea. Alex said Goodfellas. We built from there. On July 27, 2012, the concept deck included something called “Chaos, Sorkin-Style.” Pipeline didn’t instantly choose it in the room—we were still emailing afterward while they considered their options—but by early August JP was back from vacation asking, “time to talk about Chaos?” And that was the beginning of the commercial we eventually made.

Eventually is doing a lot of work in that sentence. Our original production timeline had us shooting on September 14, 2012. We did not shoot on September 14, 2012. First we needed the script locked. Then the location. Then casting. Then feedback. Then more feedback. Then clarity on feedback that everybody thought had already been clear. We would get close to a production window and something would move. Pipeline had decisions to make. Bootstrapper was also busy and our own schedule filled in around them. At one point Will Foster took over a lot of the production planning while I stepped back into more of a support role. In October I sent Pipeline an email that literally said there were three things we needed before we could schedule: find the location, review and lock the script, and decide on food. It is hard to imagine a more perfect summary of production than “script, office, sandwiches.” We kept moving. Slowly. In November I told them our next two weeks were booked and that the earliest new window was after Thanksgiving. By the end of the month I was telling them November had been a blur but we were finally free to move their video forward again.

Then December happened, because apparently this project had not accumulated enough obstacles yet. I injured my back badly enough that I needed surgery and spent a large portion of the month staring at the ceiling. The Spine of the Story gets into the larger history of my back and the surgery, so I don’t need to turn this into that story again, except to say that when I returned to work in January I had converted my desk into a standing desk because sitting for more than a few minutes was miserable, and Chaos was still standing there, next to my desk, waiting for us. On January 2 I wrote JP that I was ready to move again if they were, and even offered to slim the whole Sorkin-style idea down if they wanted to simplify it and finally get the damn thing out the door. They still wanted it. So on we went.

February is when the project really came back to life. Amy Silbergeld at Pipeline met with me and then sent an email saying she wanted to move forward with CHAOS “immediately,” which at that point was probably the most beautiful word anybody could have written. There were still a million little pieces: script edits, actress, extras, wardrobe, hair, makeup, location, props, catering. We had spent months looking at actresses. There was even an earlier joke from JP asking whether one candidate accepted shoes as currency, which lodged itself in my memory and later merged with the fact that the actual shoes for the commercial were apparently around $850 and important enough to the narrative to justify it. Cindy Lemos eventually became our lead. She was non-union, $600 for the day, able to provide wardrobe and handle her own hair and makeup, and from the moment we actually put her in front of the camera she inhabited the boss role. She had exactly the demeanor we needed, but she was also easy to work with, patient, and took direction beautifully. Honestly, everybody did. For something designed to look like chaos, the cast was remarkably cooperative.

And still we weren’t shooting. The BlueBox office location had construction problems. A possible March weekend disappeared. Then March 30 wasn’t going to work because Tyler, my DP, was at a wedding in Tacoma, and given that the entire idea depended on Steadicam work I was not comfortable just swapping in somebody else. I had already waited this long. I wasn’t going to compromise the central piece of the visual treatment because the calendar was being annoying. So we booked April 6. Finally. If I measure from the first version of Chaos in the July 27 concept deck to April 6, that is 253 days—eight months and ten days from concept to camera. The original September 14 shoot date had slipped by almost seven months. So when I say I was itching for shoot day, I mean I had spent most of a year waiting to see whether this thing that had lived in documents and meetings and email threads and revised timelines would actually become a piece of film.

And the funny thing is, the shoot itself is not what I remember as difficult. I remember it as fun. Really fun. The atmosphere was chaos, but controlled chaos. I had a good team. The first shot started in the elevator lobby, and Tyler began setting his lights and shaping the scene the way he wanted. He was more experienced than I was in a lot of the technical areas of filmmaking, and I had absolutely no problem letting him be more experienced than me. I knew what I wanted. He knew how to get the camera and lighting there. As he finished setting up that elevator lobby and I started seeing the image come together, I had that feeling: holy shit, this is actually working. After most takes Tyler would show me clips, little confidence checks throughout the day. I didn’t know whether we had every single planned beat as perfectly as I might want it, but what he showed me was enough. I knew the footage looked good. I knew I had something I could work with. I knew we had nailed the shoot.

Jory was another big reason I was able to enjoy directing instead of losing my mind managing eighteen human props. He told me he had worked on a set with a lot of extras and knew how to make a room look busy. Great. Congratulations, Jory, that is your entire problem now. I let him take over the background choreography so I didn’t have to reset every extra after every take. He knew how to give people bits of business that looked like actual work instead of eighteen people standing around desperately pretending they had somewhere to be. Tyler owned his lane. Jory owned his. Cindy inhabited the boss. I could pay attention to performance, story, the overall feel, and what the camera was telling me. The machine was humming. That is the phrase that keeps coming back to me when I remember that day. The commercial is literally about an office spinning into disorder while one person moves through it, but behind the camera the reason it worked was almost the exact opposite. People knew what they were doing. I trusted them. They did their jobs.

My original dream had been to do the whole thing as one shot. No cuts. That was the real West Wing/Goodfellas ambition in my head, and if we had considerably more time and money I would have loved to really build it that way. We didn’t. What we ended up with was a compromise that I was still very happy with: a strong primary tracking shot carrying the spine of the action, with cutaways to let us actually sell the jokes and tell the story clearly. There is a perfect example where Cindy fires a Nerf dart and it suctions onto the coffee mug being held by this mustachioed guy. There was no realistic way to guarantee that gag inside the moving Steadicam choreography and still have it read, so we got it as a cutaway. And it works. It works really well. That became one of my favorite beats in the piece, and it is also a nice little reminder that sometimes the thing you give up in production is not actually the thing that makes the finished work worse. The one-shot dream bent so the jokes could land. The cuts were a compromise, but they were also filmmaking.

JP was there as the client, and I remember him being pretty chill. He’s a funny man and I really enjoyed working with him. He stayed out of the way while we worked, but I also wanted to foster the kind of set where the client never felt like they had to stay silent just because the film people were doing Film Things. If JP or somebody from Pipeline saw something they didn’t like, I wanted to know while there was still a camera, an actor, and a location in front of us. Fixing it on set is almost always better than discovering later that the client hates a shot and now the only available solution is to carve around it in the edit. That relationship with JP lasted beyond this shoot. We became a little closer over the following years in that professional-friendship way. We would see each other at events. He later invited me onto a bus tour he organized for startup CEOs where they would take day trips together. Pipeline eventually called Bootstrapper back for more work. So despite the occasional frustrated email during this endless process, this was not a story where the relationship blew up. Quite the opposite. We survived the project and still liked each other afterward.

Which brings me back to April 11, the date the randomizer actually chose. Five days after the shoot, I was looking at the first rough cut and I already knew what we had made was a good compromise. It wasn’t everything I had pictured when I first told Alex I wanted The West Wing. If somebody had handed me a giant pile of money and a week in the location, I would have pushed it much farther. But under the budget constraints we had, I was super happy. We had the primary tracking shot. We had Cindy commanding the space. We had the background world Jory built. We had Tyler’s images. We had the dart on the coffee mug. We had enough. The footage itself had worked. The shoot was not something we needed to rescue in post.

And then post took forever anyway.

The rough cuts started multiplying. JP wanted Cindy kept in frame more. The iPhone notification needed to be clearer. We went back and forth on the apple and potato-chip gag. There were questions about dart speed, the mustache reaction, the laptop ending, whether we had a shot of Cindy looking at her phone with a knowing smile. We didn’t. There was discussion about replacing an iPhone screen in After Effects, which turned into a separate rotoscoping estimate because reflections made it more complicated than it sounded. I remember deliberately choosing the version where Cindy crushed the potato chips with the apple because the alternative of aggressively shoving the apple at somebody felt too mean for the character. Then there was voiceover. Then there was a misunderstanding about whether voiceover was holding up the rest of post. JP got frustrated and told me, basically, this was not cool. He wasn’t wrong that our communication could have been clearer. I apologized and moved things forward. Then the ending turned into its own miniature design project because now we needed to demonstrate the actual PipelineDeals product: moving a deal, customization, goals, reports, activity tracking, secure access, any device. Jody Ferry from Pipeline got pulled in to help make screen captures and static graphics. We changed music. We changed audio. We changed product screens. We changed the voiceover energy. The version numbers just kept climbing.

By June 3 I sent version seven and JP replied, “Looking much better. Feels right. Let’s move.” Then came version eight. Then version nine. The surviving copy I still have on YouTube is actually the version-nine link I sent on June 5, which makes me laugh because of course the artifact that survived all these years is not technically the last one. JP responded to v9 that the sound was much, much better and gave me two more little changes: delay the music until after the elevator ding and let us hear a few authoritative clicks of Cindy’s heels, and tighten the zoom on the customization. “Other than that,” he wrote, “I would say we are done.” I made version ten that afternoon. On June 6 he finally wrote, “Ok, looks good. Thank you for all your hard work.” The original full-length Chaos commercial was done.

Except it wasn’t done.

Because in July Pipeline came back and wanted a thirty-second version for a campaign. This was a new paid re-edit, which is important because I don’t want to make it sound like they simply trapped us in revision hell forever under one invoice, but emotionally it was still Chaos, the project I had first estimated in July of the previous year, crawling back out of the grave and asking whether I had another cut in me. On August 1 I sent JP the first attempt at compressing the whole story to thirty seconds. On August 8 he said the story worked but there was a little cut glitch and the music felt too low-key. On August 12 I sent another version with the audio tightened up. On August 14 I put a version in the shared Dropbox. On August 20, 2013, I sent the last email I can currently document in that thread: “How’s this one look to you? Are there any changes? Have you tried to upload it and use it yet?”

That is where I am calling the finish line, because archaeology needs evidence and that is the last piece I have. From the first PipelineDeals estimate on July 5, 2012 to that final documented check-in on August 20, 2013 is 411 days. Thirteen months and fifteen days. The Chaos concept itself did not exist until July 27, so yes, if you want to get extremely pedantic about it, the named concept lasted a little less than that. But the project that became Chaos started with that July 5 estimate and was still producing deliverables on August 20 the next year, and I feel comfortable letting the title have this one.

I am also very deliberately not telling this as “Pipeline made us wait thirteen months,” because that isn’t what the archive says. Some of the delay was Pipeline. Some of it was unclear or slow feedback. Some of it was casting and locations. Some of it was office construction that was out of both of our control. Some of it was our own workload. Some of it was me handing production planning off and then taking it back. Some of it was my back exploding in December and requiring surgery. Some of it was DP availability. Some of it was post-production workflow and communication on our side. Some of it was the fact that the concept continued to grow even after we had shot it, because suddenly the ending needed to do more product explanation than it had originally been designed to carry. This thing kept reaching the doorstep of being made, or being finished, and then somebody would move the doorstep. That’s what made it feel endless.

But the weird thing is that I don’t remember it primarily as miserable. I remember being impatient. I remember wanting the shoot to happen so badly. I remember the tail end taking too long. But once we were actually making the thing, I loved it. The first shot lighting up in that elevator lobby. Tyler showing me clips after takes. Cindy becoming the boss the second the camera rolled. Jory making the background look like a real office instead of a community-theater approximation of one. The controlled chaos. The dart landing on the coffee mug. Knowing at the end of the day that we had enough. Knowing at the end of the edit that, even if it wasn’t the enormous one-shot version I had dreamed about, we had made a good version of the thing we could actually afford to make.

More than anything, I am proud that I got it across the finish line. This was a big project for me and it stretched basically every part of the job. Client management. Producing. Directing. Editing. Knowing when to push for what I wanted and when to accept the reality of the budget. Knowing when somebody on my crew knew more than I did and giving them room to be good at their job. Knowing when a client’s note was worth fixing and when I needed to explain why I had made a choice. Knowing how to keep going after the exciting day with the camera was over and the project had become version numbers and screenshots and voiceovers and invoices and another email asking, hopefully for the final time, “How’s this one look to you?” I learned a lot because everything pushed me pretty far.

And I came out of it more confident.

If I could pull that shoot off, I could do more!

That was a nice feeling.

So apparently that is what was hiding inside Thursday, April 11, 2013. I thought I was digging into a random Thursday where I ate Burgermaster, went to Fred Meyer with Heidi, made jokes on Twitter, and sent around an early rough cut. Instead I found myself standing five days after one of the biggest shoots I had directed, eight months after I first started trying to get that particular idea onto a camera, still four months away from the last documented email on the final thirty-second version, right in the middle of the project that refused to end.

Thirteen months.

Fifteen days.

Apparently we didn’t scare JP away, because Pipeline hired us again.

So I guess it worked.

And, I’m pretty sure Cindy kept the shoes.