How TLDR Daily got 7,000 podcast downloads in 40 days
or, “How do you get podcast listeners in 2018?”
Background
When we first started investing in the audio space at Betaworks, I decided the best way for me to really understand the market was to start my own podcast. I wanted to use the tools, and to experience the distribution issues first-hand. At Betaworks, we had invested in Gimlet Media and Anchor (recently acquired by Spotify), and would ultimately invest in Breaker and make 8 more investments through our Voicecamp program.
In order to learn about the industry, I created my first podcast called Doing Your Business (“DYB”) where I interviewed founders of non-tech, non-venture-backed companies. The feedback from listeners, mostly tech people, was positive. The advice I got from Harry Stebbings of 20 Minute VC fame was to release the show with consistently. I aimed for once every two weeks. It was at a time when lots of people were creating new podcasts, but there was still room to get some attention.
How Do You get Podcast Listeners in 2018?
In 2018, it’s a different world in Podcastland™. While there are a lot more listeners, there is increasing competition for attention, and it seems like almost everyone has a podcast.
So I wanted to see what it took to get people to listen to a new podcast in 2018.
I had seen podcasts experiment with length and frequency — The New York Times’ The Daily is one of a set of podcasts which are, well, daily. The Memory Palace is interesting because it has episodes of very varying lengths, some are 11 minutes some are as short as three minutes.
Podcasts are normally monetized on a CPM basis, so the clearest proxy for success is total number of downloads. How fast I could get the greatest number of downloads? I had a few theories:
- Short: People now have their whole podcast lineup full and the bar for adding a new podcast is high. So my podcast should be extraordinarily short. A short podcast means low commitment.
- Frequent: I thought the podcast should be daily. Venture Capital is ultimately about speed. Would a daily podcast would get more listeners faster?
- Native: I like looking at emerging platforms because that’s where the clearest opportunities are if you’re willing to put in the work. Breaker is a social podcasting app (Betaworks is an investor) where one of the features is that when someone you follow is a guest on an episode, you get a push notification. So I very purposefully made my podcast “Breaker-first” making sure to interview people who had lots of followers on breaker or who were active and would share it around.
- Guests: I wanted to make sure guests had incentive to share the podcast beyond my network, which brings me to…
- Art: I also had a theory about the episode art. Some people tweet about podcasts, but what would make my guests and listeners really want to share out the podcast? What would make them put it on instagram? Certainly not yet another podcast with the host’s face on it. So I designed custom artwork for each episode, which featured that episode’s guest writing his or her own name in graffiti next to the podcast title, “TLDR Daily with Matt & Co.”
I called the new podcast “TLDR Daily with Matt & Co” and the premise was that each day an interesting person would give me the quick rundown on an article he or she found interesting. Again, it didn’t have to be about tech. I promised my listeners I’d do at least 15 episodes, I ended up doing 17 (plus a last episode asking if I should keep up the experiment, for a total of 18). Here were the results…
All The Data
Below is chart showing the daily downloads. In blue are the daily downloads for TLDR Daily and in red the daily downloads for my previous podcast, DYB.
For DYB, I’ve labeled in red when each of the episodes came out. My very first episode had 60 listeners. Not bad. You can see another bump a few days later to 80 listeners as it got shared around. The third episode led to the biggest boost, and I released part two of that interview as episode four. Episode 5 had another uptick and then Episode 6 got lots more attention. In 76 days I released 6 episodes, about one every two weeks. It took 76 days to hit 375 listeners.
Contrast that with the daily downloads of TLDR Daily. At 128 downloads, it did have a head start relative to DYB. Much of the initial buzz was not about the content of the episodes but about the format (primarily the short length) and, even more surprisingly, about the art.
In both DYB and TLDR Daily, the pattern is clear that episodes launch, they get passed around, and and new people listen (and hopefully subscribe). In the blue graph, I haven’t labeled when episodes came out because during the 30 day period they came out every day except on the weekends. What’s so notable here is that TLDR Daily was able to get to 466 downloads 3x as fast as DYB got to 375 — in 23 days, not 76.
More Downloads, Faster
The goal was to see if I could get more downloads faster with a short-form daily podcast. The outcome is clear when we look at cumulative daily downloads:
Earlier I mentioned that podcasts are normally monetized on a CPM basis, so the most important number is total number of downloads. I wanted to see how fast I could get the greatest number of downloads.
In 30 days, TLDR Daily got more downloads than DYB got in 4x the time, by a lot.
Influencers?
One strategy many podcasters use is having guests who are “influencers.” I had people on my podcast who had 800 followers on twitter and others who had over 200k. At first, I thought there was zero correlation between the number of followers and number of episode downloads:
I thought this was a really interesting insight, however my brother Michael Hartman does basketball analytics (think “Moneyball” but for basketball) and told me I was actually incorrect (“I actually think it is kinda correlated if you take out the ones with over 25k followers”) and sent me the following:
What about if it’s someone who was in the podcast industry? 0 for not in the podcast industry, 1 for in the podcast industry:
Wrong again! When my brother re-ran the calculation, he found that the Twitter followers accounted for most of the increase.
Breaker First?
There’s one other correlation I didn’t have enough data to analyze thoroughly but couldn’t ignore. The two most popular episodes were Leah Culver’s and Erik Berlin’s. Erik’s episode received more than 1300 listens, 3x the episode with the least listens and 1.7x the next most popular episode, which was Leah’s. What do they have in common? They are extremely popular on Breaker (they are its co-founders): Erik has about 3000 followers on and Leah has about 4500. Not all of my guests have followings on Breaker, so it’s a tough comparison, but I commented, Breaker notified all 1000+ of my followers and when their followers liked and listened to the episode, that showed up in their activity feeds, so their followers saw it.
More Tape, More Problems
One obvious catch here is that it’s more work to do a podcast every single day than to do an episode once every two weeks. However the work is mostly in the editing, and is a function of the amount of “tape” (the audio recordings) you have to listen to to edit down.
Here’s how the two types of podcasts compare:
- Six of the seven DYB episodes consisted primarily of 3 interviews I did, each about 80 minutes long each. It about five hours to create two episodes (I divided up each interview into 2 episodes). So that’s about 190 minutes (3 hours) per episode. It means the ratio of edit to record time is 2.25 mins of editing per 1 minute of recording.
- Each TLDR Daily episode ended up 5 minutes long. When I started, I talked with people for 15–30 minutes and edited it down to 5 minutes. Then the brilliant Allison Behringer pointed out to me that if I make the recording shorter, I’ll have less tape to edit. Now I keep the interviews to 5–8 minutes and it’s much faster.
What’s Next
I’m starting to record some new episodes and will release the first ones later this week, so subscribe on iTunes or Anchor or Breaker or go to TLDRdaily.com. Or your can just type “TLDR Daily” into your favorite podcatcher.
Thank you so much to all of you who have already shared what you’re read recently: Nisha Dua from #BUILTBYGIRLS, Leah Culver & Erik Berlin from Breaker, Ryan Hoover from Product Hunt, Peter Boyce from Rough Draft Ventures & General Catalyst, Naomi Hirabayashi & Marah Lidey from Shine, Maya Prohovnik & Michael Mignano from Anchor, Nir Eyal author of Hooked, Niv Dror of Shrug Capital, Caitlin Strandberg from Lerer Hippeau, Margot Boyer-Dry from Lorem Ipsum, Nathan Bashaw from substack, podcaster Allison Behringer, Phil Toronto from VaynerMedia, and Hunter Walk from Homebrew. I’m really excited to hear more about what you all and others are reading for season 2!
