top of page

The Making of "She'll Come Back"

(SPOILER ALERT - if you haven't yet watched the video, please do so before reading this article. It's only two minutes - you can do it!)


Like most of my best ideas, this one came to me very quickly. It was January of 2021 and I was sitting up in our attic/bedroom, contemplating the jumble of Christmas decorations waiting to be put away into the crawlspace when it struck me: what could be more sad than someone packing away all the festive memories they once shared with someone who is now gone from their life? It struck a universal chord, sure... but how could I turn that into something more interesting and unexpected? What if that missing loved one was dead? Or better yet, what if she was dead, having been bitten during a zombie apocalypse and her body was downstairs waiting for the inevitable reanimation? And even better yet, what if the whole story was a misdirection leading to the final, surprising reveal?

A story built on misdirection has a universal appeal and is always something that excites me as a viewer or creator. Hell, M. Night Shyamalan conquered Hollywood using (then abusing) this technique. But it's a tricky thing to pull off successfully. You have to sell the audience completely on the story they think they are watching, but at the same time pepper in enough clues so they don't feel cheated when the reveal finally comes. You want them to be blindsided but thrilled in those final twist moments, not scratching their heads thinking "where the hell did that come from... and why?" Essentially, you want them to experience the same feeling the entire world had when watching the climax of Shyamalan's "The Sixth Sense" for the first time. (and not the feeling they had at the end of "Signs" - terrific film but the ending lays it all on a bit thick in my opinion.)

Original prop note with signature

To pull this off, I came up with the handwritten note our protagonist carries with him at all times: "I'm so sorry it had to end this way". It sold the idea of a man in mourning who can't shake the hope that she might come back... or his simple denial. The original prop note I used is actually signed "Cindy" (his lost love) but that idea changed and instead of reshooting it, I simply cropped it out while editing. That, my friends, is called LAZY.

The concept and narrative arc were roughed out in probably an hour or two, tops. The story would take place in a single location (my house) over an indeterminate number of days. There would be no overt visuals or references to the apocalypse raging outside our protagonist's isolation (always handy when shooting on a zero-dollar budget) except for a couple glances by my character towards the distant sounds of chaos... which I kept in the finished video but could find no room in the mix for the matching audio cues. And I would take utmost advantage of the holiday decorations still to be put away, including the Christmas tree itself. I only needed myself as the principle actor and crew and some quick inserts of Cindy, at some point to be determined. It was falling together quickly and I was excited for a speedy, fun project. What could possibly go wrong?


Production began on the morning of January 18, 2021 with a series of storyboard videos shot using my iPhone 7. This is the crucial first step in nearly all my films. It allows me to quickly find the locations, framing and lighting best suited for the project. Also, and even more importantly, it gives me the footage I need to come up with a rough edit to gauge the pacing of the story as well as to determine the final shots I'll need to capture when things get real. 'Measure twice, cut once' is what my grandfather instilled in me a million years ago and I've always tried to hold to that.

The Storyboard edit
My creepy disembodied stand-in

Satisfied with how the whole thing would cut together, it was time to move on to filming for real. Now at this point in the Winter I had already made a couple shorts and my enthusiasm was riding high. I had spent months, courtesy of YouTube tutorials, learning as much as I could about the craft of cinematography and lighting, I had the basics down and knew that I could capture the look I wanted using my old but trusty Sony NX100, a couple cheap lights and a reflector. But doing everything myself always poses unique challenges, the biggest one being focus. As you can imagine, it's awfully difficult to gauge focus when you are in front of the camera, even with a flip screen. So I jerry-rigged a stand in for myself mounted on an old tripod. It's creepy but it works! Then it's just a matter of hitting "Record" and doing my best as an actor. I hate acting - hate it. Worst of all, I hate looking at myself while editing. But needs-must and I work cheap and never whine about lunch breaks or the catering so it's an equitable trade-off. So the little dance began - set-up, focus, hit "Record", jump in front of the camera, 'act' then jump back behind the camera to verify what I'd done was up to snuff.

This 'solo' filming took place over the next 3 days. I would shoot in the morning, dump the footage into Final Cut and then spend my late nights piecing it all together. As this rough edit began to evolve, I would figure out new story beats and in turn come up with additional shots I would need the next day. I love working this way - having a solid idea that I can take my time massaging into shape, allowing room for those flashes of inspiration that can only happen when working on it over and over again. There's never that sinking feeling of "damn, I wish I'd gotten this shot or done that differently" because I have the ability to simply correct things the next day... if I'm not being lazy, of course. LOL.

Fake blood, old shirt and sheet... presto - Zombie Cindy!

On January 23 and 25th, I was able to wrangle Cindy into doing her parts. She's always been gung-ho and a great sport about participating in these silly things I do. At this point I don't think I had told her the whole plot of the film or the twist, only that she was a zombie reanimating in our living room and that I was going to shoot her. I like to keep her in suspense until the project is finished because she's always my first and best critic and if she watches it the first time with few preconceptions it gives me the crucial reaction I need to determine if I'm on point or way off target... or usually somewhere in-between.

We finished the interior shots in an hour then moved outside for the 'apocalypse' driveway scene. I bloodied my old clothes, she wrapped herself in some fallen tree branches (for some reason that remains a mystery to me) and we scattered newspapers and refuse as best we could. I even propped my old Jeep (Rest In Peace - its final film appearance) on an angle with its hood up - and got the fog machine out for some ambience. It was cold but fun and the footage, augmented with come CG flame and smoke, turned out as well as I could have hoped for this little no budget lark.


So I had footage I was pleased with and a story I thought was solid, fun and surprising. But what I didn't have yet was the engine that would propel the narrative. I had come up with a foundation seed I couldn't shake - the song "Last Christmas" by George Michael. It was going to be the title of the short and backbone of the narrative - the idea of a holiday staple that my protagonist hated because it's essentially a maudlin, depressing song about losing someone. I was convinced this was the way to go - I even filmed myself playing the notes to the song on the piano that remains in the finished video (minus the audio). But I knew that using the original recording or even my own version of it would be problematic at best and expensive at worst.

Original voice-over script

So how exactly was I going to make it work? My original thought was a voice-over that carried through the film, cryptically of course:


I remember her asking me once what the worst holiday song ever was.

Well that’s easy… (TITLE CARD)

Man, did we argue about that.

I mean, seriously… it’s literally about someone getting dumped on the happiest day of the year. People want to listen to that?

Maybe you’d feel differently if it happened to you,

she said… (sardonic chuckle)

Well what do you know…

I even recorded the voice-over, figuring it was the perfect amount of misdirection I needed to sell the concept. Yet however much I played around with the dialogue and the timing, it never seemed to find the sweet spot I was going for, never felt compelling enough to sell the story. So I took a step back and rethought my approach. Perhaps snippets of conversation between the characters of Scott and Cindy, recorded during the holidays before everything went south, would capture what I was going for. I even contemplated intercutting visuals of phone recordings leading up to the deadly zombie attack and the moment she is bitten, figuring I could weave past and present together leading up to the climax. But that thinking didn't lead me anywhere.

After a few weeks of mounting frustration and waning belief, I decided I had no choice but to put the whole project on the back burner. I had filmed a sequence of events that encompassed a story, but with no way of actually telling it effectively. It's every artist's worse nightmare - knowing a concept is good but not having the tools or skills to bring it to life.


Cut to six months later. Other projects have been started, some completed and others ramping up. Life has moved on. And LAST CHRISTMAS sits there on my hard drive, gathering dust, hoping to be rediscovered as so many other failed projects eventually are. Nothing truly dies... they just tend to wait for their moment in the spotlight.

I'm not sure when or why it occurred to me to resurrect LAST CHRISTMAS as a music video, but at some point the thought of a song called "She'll Come Back" (or some derivation of that) seemed to sum up everything the film was supposed to be about. It fit perfectly with the misdirection I was going for - a lament for heartbreak and hope that fit the visuals perfectly. In my mind, if I could hook an audience with a catchy song, I could deliver my surprise more effectively than originally intended. With some free time at my disposal and a window of opportunity to finally wrestle this thing to life, I decided to give it a shot.

Original demo tracks for "She'll Come Back"

Now when I score a film, I write music to fit the visuals. When I write a song, it's the exact opposite. So this was a strange hybrid I was embarking on. My instincts tend to steer me towards the fun and poppy, that's just who I am. I'm not going to moan about life like some Eddie Vedder wannabe - never. So after a few hours I had a basic piano, drum and scratch vocal idea I thought could work. It was short, to the point and fit the structure of the edit I'd already completed ages ago. There's was nothing revolutionary about the music - that wasn't my intent. My goal was to sell the story lyrically, in the sneakiest and catchiest way possible. My confidence grew the more I thought through it, as if it was meant to be all along. That's an exciting feeling for an artist, I can assure you... to exhaust yourself creatively in a blind alley and then step back and have that moment of perfect clarity.

I let it sit and percolate in my mind for a couple weeks while I tended to other matters. And then on July 13th, with open days ahead of me, I dove in with enthusiasm. That first day I locked the opening verse section musically and the next day I had the bridge. Writing pop sings is something I do infrequently these days, mainly because I don't listen to pop music often and I'm never sure what kind of style I could write in that would excite and challenge me enough to exert the effort. But this was a short piece to serve a larger goal - it came unburdened with any of my usual preconceptions. It just had to be catchy and support the narrative. So with little pressure placed on myself, I flirted with and abandoned ideas with great rapidity, focusing on the essentials while keeping things interesting for myself: a supportive but colorful bass line, a subtle synth ostinato that would fill out the mix without calling attention to itself, and then some percussive and acoustic colorings to fill in the gaps and add interest around the vocal line. Standard stuff but it all came together rather effortlessly, to my relief and delight. It also gave me the opportunity (as most musical pieces do) to stretch and expand my musical tool kit.

Tweaking Kyle's initial performance

The drum parts, for example. The music program I use, Logic Pro, has a terrific 'virtual drummer' tool that allows you to choose between various styles, from Jazz to Hard Rock. These 'drummers' even have names - I chose Kyle, the So Cal drummer (goofy, I know). It's like hiring a pro drummer for the day - I popped him into my timeline and let him do his thing. Then, like a producer relaying instructions from the control booth, I was able to go in and tweak all aspects of his performance until I had something that fit the bill. It's an exciting marriage of happenstance and control that I really enjoy. Once I had that track in place, I converted it to MIDI notes and broke out all the individual drum kit instruments (kick, snare, cymbals, etc) into separate tracks which I could then exercise full dynamic control over - levels, reverb, separation, etc. Having the MIDI tracks separated by instrument also allows me to move beats and hits around however I see fit - move a kick drum hit here, move or add a few extra tom toms there. Exciting and fun stuff for a studio geek like myself!

Temp 'til you get it right!

As I mentioned previously, I rarely listen to pop music. But what I find invigorating and exciting about the new music I do hear is the fearless disregard for historically-established instrumentation. Things just aren't drums, bass and guitar anymore and I love that. Modern music is anarchy and I embrace it. So I was determined to mix some orchestra into my pop song. Why chug along with a guitar line when I can use violas and cellos instead? French horn to add color - why not? This thinking led me down some inevitable dead ends - for awhile I was convinced the entire ending section would be orchestral bombast, forgoing the band completely. But as I bounced between the edit and the music, honing and tweaking, I finally resolved myself to bring the band back in at the end, settling for a mix of the two while the orchestra alone would only carry the emotional, penultimate scenes. The tremendous luxury I have in being both editor and composer is that I am beholden to neither role, allowing myself the freedom of endless experimentation.

Solar-powered green screening


Saturday the 17th was filming day, ironically one day short of a full 6 months since this whole journey began. I was tempted to use the spiffy new Blackmagic camera gear I'd recently acquired and shoot in 4K but in the end I wanted to match the previous footage shot on the Sony. Set up and execution for the music portions was simple. I also needed to capture a few inserts for the story itself, shots I'd had planned since the project first began. So out came the old bathrobe, pajamas and prop note (which I'd saved thankfully). No technical hiccups there either. With the footage imported into my computer and my song structure relatively complete, I was off to the races.

Sunday - final mix day. The moment in the life of any new song that scares the bejesus out of me. Why? Because I like to compose in headphones. I can hear every nuance of the instruments more clearly and where they sit in the overall mix. And with a vocal song like this one, I can have a live mic in front of me to record scratch vocals as they occur to me. If I did that in speakers, the feedback would blow my head off. But the downside to headphone writing is that I eventually, through repetition, arrive at a mix that has become ingrained in my ears and when I play it through speakers, things inevitably sound unbalanced (at best) or horrifying (at worst). Now I've gotten better over the years and these inconsistencies have been reduced, but it's still a daunting moment hitting the Play button that first time. Why mix in speakers at all you may ask? Because (a) that's still how most people listen to music and it gives me the best chance to strike a balance that worross multiple platforms and (b) for whatever reason, a good speaker mix magically transforms into an even better headphone mix... I've found this to be true every time. So... a few hours of tweaking levels and EQ later, I had myself a mix I was happy with.

I imported the mix into Final Cut and finished the edit, adding the effects and moving things around just a bit more. One final EQ pass on the low frequency side of the song and this 6 month creative nightmare was finally over! Time to give it over to the world but more importantly, time to sit back and enjoy the fruits of my labors. Because, at the end of the day, the only genuine satisfaction I get from these projects is impressing myself. If I can enjoy my own work in those heady celebratory hours after a project is done, watching it over and over again with a smile on my face, I know I've truly succeeded. Compliments and praise from the outside world are nice, sure. But the real reward is my own satisfaction. And this one was particularly sweet.

So if you've made it this far, thank you sincerely for indulging me and my ramblings. Until next time... cheers!


Featured Posts
Recent Posts
Search By Tags
No tags yet.
Follow Me
  • Facebook Basic Square
bottom of page