How We Do The Omni Show

Brent here. I’ll drop the editorial “we” in this case, since I’m mostly talking about what I personally do.

I’m writing about how we do The Omni Show podcast because it might be interesting — but also because it might be helpful to other people who are thinking of doing an interview show, or who might want to do a similar show for their company or organization.

Here’s how it works…

The Guest

First I pick the next hapless victim interview guest. It’s still early days for the show, so I’ve been trying make sure we get someone from each department — and I haven’t achieved that yet, but it’s getting close.

And then I ask the person if they’re willing to go on the show, and so far nobody has said no. (It’s totally optional, of course! We don’t make anyone go on the show — except Jim.) I tell them when and where we’d meet, and revise the schedule if needed.


I don’t know most people’s backgrounds, education, and former jobs in advance, so I go to our about page, find the person, then click on their picture and read their story, which usually gives me a few things to start with.

Then we meet for a pre-interview in the cafeteria right before lunch. It’s almost always a Wednesday, because I like having a set day for this.

It usually takes about 15 to 30 minutes to learn about what the person does at Omni, what they did before Omni, where they grew up, and what they like to do for fun. I type a bunch of short notes on my laptop as we talk, and take special care to note anything that seems especially vivid or lively or funny.

I also ask two questions: can we publish a photo? Can we link to you somewhere on the web? (Both of these are optional, as they should be.)

And I remind the guest where the recording will be, and I advise them to bring some water.


I organize my notes — using OmniOutliner — into the structure we follow for the show, which is this:

  • What you do at Omni
  • What you did before Omni
  • How you came to Omni
  • What you do for fun (hobbies, etc.)

These notes are all just short lines, like “DOS as a kid” and “Knitting.” This makes the notes easy to scan while recording the show later.

You might wonder why we have a structure: it not only makes it easier for me, it makes it easier for the person being interviewed — who has likely never been interviewed like this before, but who has probably listened to at least a few of the episodes. This way they know what to expect. (A lot of decisions are about helping people not be nervous!)

Once I’ve got these organized, I create a new text document with the show’s opening and closing, and with these now-organized notes in between. I make sure I’ve added this person’s name in the right places in the opening and closing.

I also think about something funny to say or ask about right up front. I don’t always come up with something. But the idea is to do something — usually dorky on my part, by design — to help put the guest at ease right at the start.


Just before 3 pm (same day, usually a Wednesday) I grab a bottle of Perrier — yup, that’s how podcast hosts roll — from the snack area and head off to the recording.

Our intrepid producer Mark Boszko — himself a podcast veteran — is there to handle all the many things that are a complete and utter mystery to me. The guest arrives.

We adjust chairs; Mark checks audio levels and helps the guest get set up with the microphone. We remind the guest that this is being recorded, not broadcast live, and that it’s okay if they want to reword something or pause for a minute to collect their thoughts. We’ll edit that out and make them sound their best.

Mark takes pictures of the guest, if they’ve agreed.

Mark places an iPad with a big timer where I can see it. We try to make the podcast 30 minutes long, and it helps me to know, as we go along, how much time remains.

I remind the person to say their name when I say, “Say hello, [your name].” It’s a silly and ancient joke, but it has its points. One is that it sets the tone: we’re not hip, but we’re secure in our own dorkiness. Another is that people like saying their own name. Another is that the first step in getting to know someone is hearing them say their own name.

We start rolling. I read the opener, call for music — because my notes literally say “-Music-” at that spot. I don’t have any questions written down for the interview shows, but I have notes, and we just talk for about 30 minutes as I try to guide us through the show’s structure.

The notes — on my iPhone at this point, usually — are just a guide: other things come up, and some things in the notes never make it in. That’s totally fine. A goal is to keep it loose, as if we were at a coffee shop, while still hitting the notes I want to hit, and being awake to the possibility of cool and unexpected things.

The show should sound effortless. You the listener should feel like you could be a guest — or as if you could be the host. As if there’s nothing to it but some talking.

After the recording I thank the guest and Mark, and head back to my office.


Soon after recording — sometimes the same day — Mark gets me the unedited version. I create the show notes and pick out 30-ish seconds for an audiogram — a video file that can be shared on social media — which we use to promote the show.

To create the show notes, I start by listening and typing every proper noun that’s mentioned. Afterwards I go through and find links — usually by DuckDuckGo-ing — for each. I don’t end up doing every single proper noun, but I do many of them. The links are often Wikipedia links, but Wikipedia’s cool.

I also write a few paragraphs for the top of the blog post, before the big list of links, where I introduce the person and mention some interesting things about the episode.

I get all this to Mark, who then queues it up on the website.

Mark then edits the show. Mark could, I’m sure, write an entire blog post about this part. I’m betting that he uses some heavy filtering and trickery on my voice. I have a great voice for print. :)

Mark sends the audio to a service that does a transcript, which he then edits, and that also gets queued up on the website.

Mark also creates the audiogram and picks out the best photos to use, which he checks in to source control where I can get those files.


The podcast episode is automatically published when the right Wednesday morning rolls around. And then we promote it via Twitter,, and our main blog — including the audiogram and photos, depending on where the promotion is.

By this time we’ve already got one or more other podcasts in progress or scheduled soon.


You might wonder why we do the podcast at all. I don’t have some grand marketing philosophy, other than 1) make great apps, and 2) look out and let other people look in. This is about #2: it’s about letting the world get to know us.

Omni’s always been darn good about that — note, for instance, how accessible CEO Ken Case is on Twitter — but the idea behind the podcast is to kick it up a notch. (Same idea with the Slack group, which we’ve found very valuable.)

It’s not turtles all the way down, by the way — it’s behind-the-scenes all the way down. If the podcast is behind the scenes, then this blog post is behind the scenes of a thing that’s behind the scenes. :)

The podcast also has a nice side effect: co-workers get to know each other better. Which is cool, and might actually be my favorite part.