Dear founder,
Since I talked to Anne-Laure Le Cunff, the neuroscientist and author, on my Wednesday episode this week, we had a conversation about goal setting, running experiments, and trying to deal with uncertainty.
It made me want to take some time to talk about experiments that I’ve recently run in my life: in Podscan, in my media business, on Twitter, off Twitter, and just report on them. I want to give some insight into how my operations work, what things I’m trying to do, and what things I want to do.
Looking back at all these experiments now, I realized they share a fascinating theme: they’re all about removing constraints. These aren’t necessarily creative experiments that build something new, but rather they remove old assumptions.
The assumption that “this needs to be that way” — well, does it? That’s not necessarily true, realistic, or expected. Constraint removal has been a big theme across all the experiments I’ve been running over the last couple of months, weeks, and days.
Podscan Experiments
One of the biggest experiments I’ve done over the last two months involves the podcast data I have inside of Podscan. It’s now over 3 million podcasts with over 25 million episodes of transcribed data between them. This data has always been in my database, but I’ve now made it available to the public to a certain degree by building public podcast pages.
A couple of months ago, Chartable was closing down, and there were no more podcast charts easily available for anybody. So I quickly built a way to offer podcast charting information from Apple Podcasts and Spotify to my customers. I thought this might be interesting for non-customers too, so I made it available publicly on Podscan for non-logged-in users.
Then I realized I could show more information about these podcasts: who the podcast is by, what the last couple of episodes were about, a summary, maybe the description and thumbnail — just a landing page for people to see some demographics about it. I made all of this available through the website, and those pages have now started ranking on Google. There are over 20,000 pages on podscan.fm that have started to rank for both niche podcasts and the big podcasts that exist on the platform.
This has been a really successful experiment, but I had to overcome my inhibition to give my data away. I felt that one of the most valuable things about Podscan is all this data — all my transcripts and all the things I’ve scraped from different places, the chart history I’ve tracked. All that data exists only in my database. No other company has transcription data on these episodes. So it was hard to put this out there for free.
Obviously, not all the data is freely available. People still have to sign up and sign in to see the full data, and after a while, they can’t access it anymore without subscribing. But it was still hard for me to overcome my protective instinct with all this information. This experiment has led to a significant increase in sign-ups and conversions, so it’s been very successful.
In the wake of this, with those public pages existing, it also allowed me to add a feature that I always wanted to have in the product: “Claim Your Podcast.” I’ve always been interested in adding this, mostly because competing products have it and because it’s a great opportunity for me to improve data quality and hygiene in the product itself.
I was hesitant to add this feature because it would increase complexity. I need to make sure that the information people put onto their profiles is verified and valid. There are a lot of things that can go wrong if you allow people to deal with their own data. But on the other side, it also exposed me to a completely new kind of potential customer, some of which are paying customers now. They found that having control over their own shows, if they are hosts or producing companies, is quite significant. And then upsells can happen because if somebody keeps their podcast data up to date and they figure out, “Wow, I could actually track my mentions — my podcast has been mentioned several times over the last couple of weeks,” they’re already in the product that offers this capability.
This was an experiment that’s gone very well. I was fearful of the technical complexity of validating podcast ownership, which I thought was way more complicated than it eventually turned out to be. It was really just checking for a token that I determined on somebody else’s podcast feed, which can be easily added. So it turned out to be a non-issue, but it was something I purposely hadn’t built in the 10 months that the product has been on the market because I felt it wouldn’t align with the core features.
Looking at it now, it’s definitely a win-win feature, because it improves the quality of the underlying data in the system while at the same time inviting new users who are already in the professional field of broadcasting into a product that may have features they didn’t even know they needed.
Not all experiments worked so well in Podscan. I released a feature called “Topics” a couple of weeks ago that hasn’t really been adopted much. It’s only used by certain people who are interested in it. It tracks the overall themes of podcast episodes and allows people to keep tabs on every other episode with that particular theme. It was meant to surface data that was already there anyway, but it didn’t really do much. It wasn’t really much of a thing.
I’m trying not to overcomplicate the interface because a lot of my users are not technical at all, and they don’t see data through the same lens as I do. On the other hand, I’m trying to surface as much information as I possibly can, because that’s why Podscan is used and what it’s used for. So it’s been a bit hit-and-miss when it comes to feature deployments. The big feature deployments that people really liked were “Lists” and the “Claim Your Podcast” feature that I released recently.
Sales Calls and Metered Billing
Something else I’ve had to experiment with, because I was really uncomfortable doing it, was sales calls. I was on what can only be described as sales calls (even though they were never meant to be) for several days recently. This not only landed me a few clients on my most expensive plan, who are using the platform as a business data collection integration to get podcast data into their own business systems, but it also allowed me to finally offer and sign metered billing on the API.
This is also highly experimental. It’s something that I don’t even have as part of my billing system because it’s so new and so different from what I usually bill for. The billing for using the Podscan API to get podcast data has been capped at a daily rate of how many API calls you can make, and depending on the tier of subscription, that’s how much you get.
Well, this customer asked me, because they needed way more than I was offering, if they could get a metered plan where they could pay by the thousand. So I had to go deep into my wishful thinking document, do some math, and come up with a price per 1,000 requests. We negotiated this in a call, which was also very nerve-wracking for me and very new — I haven’t done much of this before. From that call, we agreed to do monthly billing.
So now Podscan has a tiered per-1,000-requests price in addition to the existing subscription tiers, which allows very precise billing for use of the platform and gives me an opportunity to supply API calls and listed data to customers who are really willing to pay for it. The price is based on a negotiation, but I would definitely offer it to other customers at a similar rate — final pricing depending on the overall expected load, may be higher or lower for others.
I really like this because the customer needed the data quickly and reliably. They had just turned on their system and were hitting their limits. Knowing how important that customer was for the future of the business, I not only agreed to set up this kind of billing system but also implemented several things for that particular customer to be able to use the API most efficiently.
Obviously, that kind of behavior — building whatever one specific customer needs — won’t be smart forever. But if it’s a customer that can supply one of my most wanted use cases (intense integration of the API into another business) and is willing to pay several thousand dollars a month for it, why would I not do that?
Making Time for Hobbies
Personally, outside of Podscan, I’ve been trying to make some time for hobbies again. Something I really missed over the last months was taking time away from the computer. I had a phase a couple of months ago where I would read a lot — I would just sit down and read books for four hours a day when all the important fires were doused and all the things that had to be done were done. I would just lay back with some fantasy book. That phase was really nice, but it ended at some point.
What I’ve done now is get back into miniature painting and my 3D printing hobby, which is a nice thing you can do between coding sessions — which, more than ever, are just me instructing an AI to deal with a certain problem and then coming back five minutes later. Well, five minutes later, that’s a coat of paint. That’s a little space gun painted, or a little layer of cloak color added to a model.
So I’m going back and forth between building little spacemen (my old Warhammer 40,000 hobby) and building a business. I’m also just talking about it, sharing information, and doing a couple of webinars recently — just a lot of different things that bring diversity to my day.
When you’re only working on the same thing day in and day out for many months, burnout comes knocking. You can feel how you’re more stressed, more anxious. I’ve felt this often enough in the past to know that the way out is embracing other things to bring clarity and diversion from the things that bother me all day. It also helps with not touching the critical parts of the software for a while.
I know that I need to take care of myself to be able to keep running the business, and I owe it both to my customers, myself, and my family to be present and understand my own limitations. That’s important. It’s important for me to know that I’m limited in ways, and that I can choose to act to make my experience an enjoyable one.
Trying to step away from work when there’s always work to be done is hard. It’s a thing that, for me in particular, feels like it’s going against my identity. As a founder, it’s not that I’m on the grind set, but my founder identity says, “If you use your time productively, it’s a smart use of time.” Painting little robots or space aliens does not often feel like a productive use of my time, but allowing myself to understand it as a necessary part of my health regimen makes it a productive use of my time.
By phrasing it as a mental health activity — therapy, if you want to call it that (which it obviously isn’t, but we could jokingly call it a therapeutic activity) — it allows me to feel like I’m still an active founder and I’m taking time for myself. Because if not, I would just be a nervous wreck and very anxious. I don’t want that.
That’s one of those little internal changes I made in my own self-perspective that has been very helpful. I’ve found that I’m much more reflective on business challenges or technical challenges. A lot of my problems in the software part of the business have kind of solved themselves as I just fiddled with plastic glue and my exacto knife. By using my hands and my eyes, by doing some actual crafting, I’ve found that my mind was at peace because the action it was focused on was so simple that it could just start exploring, instead of being fully focused on building a complex thought.
When I’m cutting a piece of plastic or trying to shave off some excess here or there, the task is so simple that my mind gets to relax a little bit, and that is usually the moment where revelations happen. So I’ve been seeing very good results from making that little change in my life.
Media Business Experiments
On the media business side of things, I’ve been changing very little. One of the experiments I did over the last half year or so has been very productive and helpful. That’s been dropping the requirement to put out an interview episode every week and only interviewing people that I actually think are important for me to talk to right now.
As you probably have noticed, the interviews on this podcast have been sporadic in nature and almost kind of random, but I prefer to have sporadically happening interviews over being forced to find people to talk to. The last few guests that I’ve been talking to have almost exclusively been people that I really wanted to chat with, that have written something I really liked, or that are talking about certain topics that I really enjoy — just kind of vibe interviews with people who I enjoy and want to hang out with.
That has been very helpful for me to keep the quality up and to feel like it’s not a chore but an opportunity. Even though I have minimums for my sponsor agreements, it still allows me to be selective and talk to people when I feel like it, and then deploy the episodes when it best benefits them, just like I did this week with Anne-Laure as she had launched her book on the day that the episode came out.
So this has been an experiment in allowing myself to not feel too constrained. I think all of these things are like this if I think about it. Allowing my data to be more public — that’s removing an internal constraint. Allowing myself to only interview people that I like — that’s removing an external constraint. Allowing my customers to administer their own data — that’s removing a constraint that I put on them. Allowing myself to spend my time with hobbies that are very different from my work — that’s removing a constraint too.
Constraint removal has been the underlying theme for all of these experiments, and they’ve largely been successful. I think there’s something to be learned here: sometimes the most effective changes aren’t about adding new things but about taking away limitations we’ve placed on ourselves or our businesses.
🎧 Listen to this on my podcast.
If you want to track brand mentions, search millions of transcripts, and keep an eye on chart placements for podcasts, please check out podscan.fm — and tell your friends!
Thank you for reading this week’s essay edition of The Bootstrapped Founder. Did you enjoy it? If so, please spread the word and share this issue on Twitter.
If you want to reach tens of thousands of creators, makers, and dreamers, you can apply to sponsor an episode of this newsletter. Or just reply to this email!
