Credit: Michael Coury
Listen Notes has been around for ~9 months, but I spent only 1 month building it early this year. Afterwards, it had been in autopilot mode for 8 months or so. Things changed recently. I just left the company that I co-founded last Friday (Sep 15, 2017). Therefore, I can focus on building & improving Listen Notes now.
I want to document my thoughts & plans on Listen Notes in this post.
Design principles
For the same topic, different reporters write news articles from different angles, because those reporters value different things and have different incentives (e.g., values, politics, money, whatsoever). Similarly, different developers come up with different implementations for the (seemingly) same idea.
When you search “Podcast Search” on Google, you’ll find a bunch of Podcast Search Engines. Some of them are developed by well-funded startups, while some of them are developed by individuals (like me). Clearly, developers behind different Podcast Search Engines have very different personalities and have different incentives, thus, different design principles.
What are the design principles of Listen Notes?
Episode-centric
A Podcast is like a website, while an Episode is like a web page. When using Google or other general-purpose Search Engine, oftentimes you are looking for web pages and usually you don’t care which websites these web pages belong to.
There are more and more Podcasts. As a listener, I don’t want to subscribe to too many Podcasts. I couldn’t listen to every single Episode in the Podcasts that I subscribe. Oftentimes, I binge-listen to a bunch of Episodes of the same topic, where those Episodes may come from different Podcasts. For example, if I want to learn more about a celebrity, then I just find a bunch of Episodes that interview this celebrity and listen them all (e.g., Chris Sacca).
Most Podcast-related apps / online services are still Podcast-centric. People rarely bookmark a lot of websites nowadays. I bet that listeners will not want to subscribe to tons of Podcasts in the future, due to recent Podcast boom. Want to find interesting contents? Search! Hopefully search on Listen Notes :)
Simple, but actually works
Many Podcast Search sites are way more technical than Listen Notes, where they use fancy AI / Audio Search / NLP / other-buzzwords-tech to power their search engine — it’s easier to raise VC money by calling yourself AI-company nowadays :)
Best doesn’t always win. Those AI-powered fancy Podcast Search sites may work well for a small set of carefully hand-curated Podcasts by doing automatic + manual audio full text indexing — Imagine that Google indexes only a thousand web pages; not matter how good the search quality is, you can’t find interesting things, due to the small set of indexed contents.
Life could be simple and happy. Let’s have a complete database and just index the text description. Make it work first, then improve it whenever those fancy AI technologies are ready.
Find what you want, then leave
I don’t want to build yet another Podcast player app. I don’t want to trap listeners to Listen Notes. You come to Listen Notes and find the Podcasts or Podcast Episodes that you want to listen, then you leave Listen Notes to use your favorite Podcast player app to listen.
Under this principle, Listen Notes shows RSS & brings traffic back to official websites of Podcasts. Many Podcast-related sites don’t show RSS, because they want to build a walled garden to make visitors stay there as long as possible.
What’s next
I do have some plans.
Polish the website UI
As a practical engineer, I strongly believe that you should
- make it work first, then make it beautiful;
- release fast and release often.
Listen Notes works, and people do use it, and people do like it. However, it’s bare-bones (Thanks, the Verge), for now. As I said in the beginning of this post, I spent only 1 month developing Listen Notes and polishing UI was not a priority then — an ugly web service that actually works is better than a beautiful toy that doesn’t work.
A very logical next step is to make the website UI a bit polished, which is what I’m going to work on immediately.
Improve Search Usability
Next, I need to improve two very basic things:
- Search quality. Make search results more relevant.
- More search options (e.g., filter by language, category, fields, …). It’s like Google Advanced Search.
Community Feature
It’ll be something like “Goodreads for Podcasts”. Well, I haven’t figured out yet. It’ll take me quite a while (half an year ~ one year) until I get to this part.
How does Listen Notes make money
To fund the development of Listen Notes (and slow down the burn rate of my personal savings, as technically I’m unemployed now), I have to monetize Listen Notes from day one. But how? 1) Ads & 2) Developer APIs.
If you have better idea for monetization, please email me: hello@listennotes.com
I’ll send out monthly updates to those who are interested in Listen Notes development: http://eepurl.com/gfSdfj
