Tell us about yourself and what you're working on!
I'm Tom Andrew, a 27 year old from Brisbane, Australia and the founder of WhatSong, Movie & TV Soundtracks. I'm a self-taught web programmer and designer, but to be honest, I have little real-world programming experience. I worked in a grocery store for almost a decade while studying finance full-time. When I graduated I realized I disliked both and tried to develop my programming and entrepreneurial skillset as much as possible.
I'm currently spending 100% of my time on WhatSong, preparing for our upcoming platform upgrade. WhatSong allows users to find songs they heard in the latest films and television shows, however our short-term future direction is towards a fully-fledged music discovery tool and community, with full integration of Spotify, SoundCloud, and YouTube.
What's the story behind how you started WhatSong?
I guess starting out I had no explicit goals, and I used WhatSong mainly as motivation to learn how to code. It also helped to distract me from a university degree that I found quite boring and hopeless, especially considering I had no interest in a corporate career.
I actually got the idea after watching Knocked Up, where the gang would write down how far into movies girls took their tops off. I figured I could probably figure out how to build a site that did the same for music. I think it definitely helped that I loved music and film at the time, as it provided a practical enjoyable setting for me to practice programming.
The very first test version I built using Dreamweaver, which I found very useful because you could see what sort of code was generated with each element you placed. I then started learning HTML, CSS, and SQL. I have a vivid memory of always struggling with PHP, and hired a cheap developer to perform a couple smaller back-end tasks. Fast forward a few months and there was a 3-4 page site with search and around 150 movies and a few thousand songs in an SQL database.
After a couple months of continually adding to the database I found myself disheartened, as I had built a product that no one was using. I didn't touch it for almost an entire year. After graduating in finance and finally realizing that I had less than zero interest in it, I decided to push really hard on the SEO front to get Whatsong going again. After almost a 6 months of applying SEO principles, I could slowly see the search terms rising on Google.
Since then, there've been a lot of ups and downs. For example, my site was hit quite hard by the Panda/Penguin Google SEO updates in 2013/14. As a result, all my visitors and revenue dropped to near-zero for around 3-4 months. I've always had an image of what the site would one day be able to acheive though, so I continually pushed through. We were also unlucky when the music streaming service Rdio was bought out, as we had just spent $5000 on implementing their API to allow full-length playback, and had to bin it as soon as it launched.
Today we have over 50,000 visitors a day, and we continue to grow as we launch new products like YouTube and Spotify integration.
How have you funded your development?
The expenses were quite low to start with, as I only had one back-end developer and I could pay him out of my weekly full-time wage. I was quite lucky, however, as within a year the website was starting to make enough money to support all development. I was then able to work less at my "real" job and dedicate more time to the website.
As revenue has increased, I've been able put money aside for any major future upgrades. For example, version 2.0 of the web app launch soon. It took almost an entire year's worth of profit to build out, as well as four months of my time. That said, I havn't made any substantial, non-cosmetic changes to the website in nearly 7 years. Version 2.0 has started completely from scratch with the latest technologies.
My current team, although expensive, are quite exceptional. I have been able to slowly afford better developers to help me out as the website grows. I currently have two professional contractors working on the web app, and they are helping us build out V2 from scratch with an entirely new stack. For those interested, we are using the MERN stack, which is pretty much the same as the MEAN stack, however we chose to go with React over Angular, simply because it's a little "lighter" than the enormous Angular framework.
As you can see, what is required changes over time, and I think developers these days get too caught up in trying to select the perfect technologies and languages before starting. Choose something that works and suits what you need, and adjust it as your business changes.
How have you attracted users and grown WhatSong?
I've always wanted users to organically find and use the site. It was never my intention to coerce, advertise, or actively market the site at the start, especially because I did not have the money to do so. I spent nearly all of my time concentrating on SEO, and spent almost an entire year monitoring, researching, and comparing various data I had collected to try to rank at the top of Google for my desired search phrases.
Marketing efforts also didn't really make sense for this sort of website, as we obviously were not trying to sell a product. To even call WhatSong a service would be a stretch. I'd always known that if I were to monetize the website, it would only come in the form of advertising or affiliate marketing.
This basically led to my growth strategy being entirely based around my Google rankings. I started to realize we could outrank most of our competitors, including Amazon and iMDB, by sticking to our niche, and so I stuck to phrases that applied to the site. Instead of trying to beat iMDB at a search like "The Great Gatsby", I would heavily target phrases like "The Great Gatsby List of Songs" or "The Great Gatsby Soundtrack". If we could get a soundtrack up quick with scene descriptions and a few social likes, it would usually go quite well. For example, I remember we got "Deadpool Soundtrack" to the number one position just after it premiered, and that single page got over 1 million views in two weeks. All without spending a single dollar.
In 2015 we also bought out a smaller TV music competitor and ingested their database so we could be more prominent in the television market. This gave us another boost in domain authority, and also gave us the start we needed to branch into shows, especially considering our major competitor has such a strong hold on the this segment.
How are you growing your revenue?
Considering I never expected to earn money with this, revenue has always been low on my priority list. I've always concentrated on simply trying to create a product that I personally would not only use, but come back to. Given personal bias, it's a hard thing to know, but I feel as though it's getting closer and closer to that point.
Obviously to get it to that point, revenue is necessary, and I think it has come as a direct consequence of trying to create a product for the users, rather than a product to simply make myself money. I never wanted to charge users for site access, or impose on them with advertisements. At the start, this remained the case. I was able to get away with making enough money purely from iTunes and Amazon affiliate marketing. The site had a click-through rate of about 15%, and most of these were in turn buying something from the online stores.
However, as online streaming services like Spotify became more and more prominent, especially towards the end of 2014, our affiliate marketing revenue began dropping each month. It actually still continues to drop to this day. Back in late 2014, we were easily generating $4,500 per month from affiliates, however this number today is closer to $2,000.
The site is still free, and always will be. However, I had to implement two Adsense units on each page to supplement revenue so the site could continue to grow. These can easily be removed by simply logging in. Even though only non-authenticated users could generate Adsense revenue, it began earning more than twice what our affiliation channels were.
I'm not sure what the future of the site will bring for revenue generation, but it's stable and sufficient for now. We are concentrated on the product right now, and will look towards other revenue channels towards the end of the year. Our current business model generates around $6,000 - $8,000 in revenue each month, and peaks during holiday seasons closer to $10,000.
If you had to start over, what would you do differently?
I guess the main thing I may regret is not building it entirely myself. It was always my initial intention to use it as a method to learn programming. While I'm not a novice developer, I know I could have learned more if I'd done it all myself. I would have always liked to find a second technical partner so that we could learn and build the entire business together. Over the last 3 months I have taken a lot of steps to refine my programming skills, allowing myself to competently lead the front-end development in the near future.
What have been your biggest advantages?
The fact that the website was able to be built in incremental stages has kept risk and expenses relatively low. I was lucky that I had a product that didn't need to rely on heavy marketing or upfront costs. I built the minimum viable product and progressively developed it over time, with minimal expense.
Although it was a lot of work, it also took considerable luck, especially given my lack of initial experience. It's also something I didn't learn from, and when I tried to launch another startup with some friends last year, we decided to sink a lot of money upfront without proper research. We ended up with a costly product that people didn't seem to need, and with no money left for marketing. So make sure you start small, spend what you can afford, and be sure people will actually use it.
What are your goals for the future?
The intention is to turn WhatSong into a complete music discovery tool. User retention is a big issue for us at the moment, with most people using the site and leaving to the iTunes or Amazon stores within a couple of minutes. Full-length music playback, user playlists, suggestions, and third-party integration are all on our shortlist right now, with the intention of favorably increasing the user experience. Mobile apps will also come later in the year. However with a finite amount of financing, we tend to opt for development of one feature at a time.
What's your advice for aspiring indie hackers?
I'm not really sure I'm in the position to give advice, but I will say the most important thing is patience and perseverance. If you have a product that you impartially believe people will use, stick at it. Keep expenses at a minimum to start with, and if something isn't working, change it. And remember to keep learning.
Where can we learn more about you?
I don't actually have a personal blog or a Twitter account, but you can email me at email@example.com or follow @what_song on Twitter. We've been pretty slack on the Twitter front, but it will start to kick off again soon.
If you have any questions for me you can ask below in the comments: