How I shot 18 interviews in 8 hours

In June of 2009 I produced a video for United Way of Santa Cruz and Applied Survey Research, sponsored by AT&T. The video is a presentation of 15 years of  the “Community Assessment Project”, an initiative that has changed the quality of life in Santa Cruz since it started because it measures hundreds of indicators about health, economy, environment, crime etc. The budget for the video was extremely tight and the expectations, of course, fairly high. The video had to feature several interviews, strung together in a seamless series of fragments that would create a single conversation from many voices. In this way I can keep the interest of the viewer up while at the same time delivering a powerful message.

The schedule allowed for only one day for shooting all the interviews. We had about 8 hours, from 8:30am to 5:30pm,  and the local meeting room at the United Way office. The schedule was to interview one or two guests every half hour. I had to transform a bare and stark room into something visually appealing and deal with 18 people who are not professionally trained in doing interviews or talking on camera.  There are psychological issues at hand that weight on all this. On top of this I am shooting, lighting, sound recording, basically doing everything except asking the questions. For that task I had, fortunately, the help of local radio professional Deanna Zachary who also pitched in to slate the shots.

It was imperative to have a setup that would work with people of all kind, with all kind of voices, from powerful to faint, and with all kind of heights. And we had no more than 4-5 minutes to adapt the scene for each guest.

Running lavs was out of question and quite frankly I’m not too fond of those. Give me a boom mic any day of the week. Using Green Screen for the background was also eliminated because of the cost involved in doing good chromakey work. A couple of days before the shoot I decide to use a projector connected to my laptop and showing a generic, pleasant background image. I did some tests and I was so pleased with the result that I decided to go for it. It’s more interesting than a draped backdrop and it worked perfectly for our setup. This technique was illustrated  brilliantly years ago by the Digital Juice website. Search for it. The only downside of this is the potential noise coming from the projector. Modern devices are much quieter and can be made ever more quiet by some shielding.

With my setup decided I went to the day of shooting with all my equipment, my trusted, calibrated, JVC HD-100 camera and I proceeded to setup the scene as it’s shown in the following pictures:


This side view shows the interviewee sitting, facing the interviewer, the camera is in front, the projector is behind the person interviewed, connected to the laptop and the image is projected far behind, on the wall. I had my HD-100 connected to another laptop running Adobe OnLocation to record all my clips directly to hard disk and to use the great features of OnLocation to check my focus and exposure via the built-in Waveform monitor. For this kind of situations the OnLocation solution is absolutely hard to beat:

Here is another view of the setup:


My lighting solution was very simple but effective. I used a medium softbox by Photoflex positioned at a three-quarter angle in front of the subject, adjustments were simple and fast for each subject. A half-gold reflector disc  was placed on the opposite side and a spotlight provided a bit of hairlight/rim light. Here is the result:

Actual frames from the video


The interviewer was sitting camera left so that the interviewee was naturally facing the camera at a three quarter angle. The idea was to shoot the interview with only the answers, no questions, and to “stitch” together all the answers in a single conversation carried by many voices. See the video for the result. By not having to change microphones I had  to simply adjust the height of the chair and the position of the boom. A few seconds to adjust the frame and check focus and exposure in OnLocation, all this while chatting with the new interviewee so that he/she could relax. Do a quick sound check, slate the shot and you’re ready to go.

After the shooting was done I was able to move all the clips from OnLocation to my editing machine, a MacPro with Adobe Master Collection CS3.  I imported the RAW m2t clips in Premiere Pro, no conversion necessary, edited, added After Effects comps directly in the timeline and saved hours that I would otherwise have spend in acquiring the footage from the tapes and then convert the clips to QuickTime. Native editing is the way to go.

Final compositing and grading was done in After Effects by simply importing the final Premiere project and adding effect and transitions. One last note. All the 3D text was created in Blender 3D.

This has been a very effective workflow that I’m sure I’m gonna use many times in the future. Keep thing simple, keep the people interviewed relaxed and keep the drama out of the experience. At the end of the day everyone told me how surprised they were on how smoothly the shot went.

I hope you found these few notes useful, leave a comment if you need more details.