What started as an app for creating lip-syncing videos has since evolved into a full-featured communication platform that has been downloaded by more than 100 million people across the globe. Before embarking on Dubsmash, co-founders Daniel Taschik, Roland Grenke and Jonas Drüppel had worked on two apps that failed to pick up much traction among users. But perhaps the third time’s a charm. After only a week of launching the video lip-syncing app in 2014, it became a number one hit in Germany. Since then, Dubsmash has gained popularity around the world and caught the attention of big names, including the likes of Rihanna and Jimmy Fallon. Surprisingly, all of this has come without the startup spending a dime on marketing.
After noticing that users were sending videos in an asynchronous way and using Dubs—or clipped lip-syncing videos—like emojis to express themselves, the team decided to focus on letting people share video content and start chats within the app. This led to their latest release: Dubsmash 2.0. The newest version comes with a revamped design and allows users to send video messages to friends directly in the app through one-on-one or group conversations.
Following the recent launch, we visited the company’s headquarters in Berlin and spoke to CTO Daniel Taschik about the future of video communication, how learnings from their two previous apps informed the way Dubsmash was built and the importance of taking an iterative approach to app development.
Where did the idea for Dubsmash originally come from?
The three of us met at the end of 2012 at a tech meetup here in Berlin. It was a lucky coincidence that we bumped into each other and found out that we shared an interest in both mobile and video. So we started working on some projects and found out that we worked pretty well together. Eventually, we thought, ‘Let’s start a company to make video creation on mobile as expressive and easy as possible.’
Dubsmash’s latest update includes a messaging service that transforms it into something of a communication platform. Can you talk about version 2.0 and the thinking behind it?
Instead of spending time on social networks, there is a trend towards spending more and more time sending private messages because it’s a more relevant way to exchange information with people you’re close with. Of course, there are people who want to broadcast messages to a big audience, but the majority of people are interacting with a very limited, close group of peers. That’s why private messaging in the mobile space is growing so quickly.
To give you some perspective, shortly before our Dubsmash 2.0 release, we saw 35 videos shared on the app per second. Out of these 35 videos, 34 were shared privately into messaging apps. Having noticed these trends as well as reflected on our own lessons learned, we were convinced that building the latest version was the logical next step. Now, you can connect with your friends one-on-one as well as send Dubs in groups. Our vision is to bring joy to video communication by enabling people to easily create videos as well as to share them with their most important friends. Dubsmash 2.0 lays the foundation for that and is the start of getting this kind of communication platform going.
Given the app's popularity, what do you think it is about the user experience that has made it such a ‘sticky product’?
I think on one hand, it’s the simplicity of creating the videos and on the other hand, the entertaining part of it. If you look at other messengers like Whatsapp or Facebook Messenger, you can send videos through it but no one really does it. It’s not very fun and also very difficult. For most people, when they see a red record button blinking, they feel intimidated, don’t know what to say or might be afraid to say something wrong. Also, listening to your own voice can be awkward, which is another reason that limits people from creating and sending videos. With the Dub format, though, we overcome that by putting a very well-known pop culture soundbite to it – that makes it so sticky. I mean, I think there are many other aspects to it, but a lot of it has to do with the fact that people can express themselves through video without the intimidation.
How have your two previous experiences building apps informed the way you developed Dubsmash? What were the key lessons learned?
Almost everyone carries a smartphone around with them and, most of the time, the device has a processing power that is strong enough to process a video and is equipped with a camera that is almost as good as a DSLR. While it was a common thing for people to take photos, put filters on them and then share them on social networks, people weren’t using video in the same way as much. We were wondering about this a lot and why that was. And it seemed like there wasn’t the right tool in the app stores to create expressive videos.
We built two other apps before Dubsmash. One was to mix up YouTube videos with personal videos and the other one enabled users to create 30-second music videos with different effects. Both apps didn’t really take off but we came away with a lot of learnings that we built into Dubsmash. The first lesson for us was that both apps were just way too difficult to use. It took people too long to create a video. Secondly, we expected people to share it publicly, on Facebook and Instagram, but they preferred to share it through private spaces and messages. The third thing we learned was that a pre-existing sound helps users to more easily create expressive videos.
User-testing undoubtedly plays a large role in the app development process. Can you shed some light on what your current user-testing and iterating process is like? How has staying lean impacted the way the product was built?
We’ve been working in a very iterative way on Dubsmash since day one. When we were working on previous projects, we were building and building and building – and then trying it out. But with Dubsmash, we work on a feature for a couple of weeks and then ship it out to our first users. We also leverage quite a bit of A/B testing, so we use technical tools to show certain new features to a very small fraction of the users. Instead of rolling out a feature to our whole user base, we choose a small user segment of 10,000 to 100,000 users and measure how well the feature is performing. This allows us to iterate really fast. We don’t have long development cycles. We just develop the basic functionality, ship it, measure it – sometimes throw features out – but often rolling it out to all users after two or three iterations.
What have been some of the biggest challenges you and your team faced while building Dubsmash?
Definitely the scaling aspect. Pretty early on, we started building the app and the back-end behind in a very lean way. We didn’t have the resources to tailor to super scalable back-end systems from the beginning, but as the platform grew, we applied technology and know-how to build systems that could continue growing with our multi-million user base. One of the best things that we came across was the cloud. Having computing resources at hand as we needed them, helped us to elastically scale up and down depending on the load.
What’s in store for Dubsmash in the coming year? What will the company be working on?
For us, the new release is just the foundation. We’re going to continue building features that make it easier for users to create and express themselves through video. We won’t only stick to the Dub format, but will provide other ways to open up the format for users to create videos. We’re currently playing around with sound manipulation and overlays. These are just some examples of how we’re going to continue giving people better and easier ways to lower the intimidation barrier when creating and sharing videos.