From Idea to Live Product

We’re testing, testing, testing: me with my co-founders Nicolaj and Tarjei (in the back).

In the fall of 2015, our startup journey was marked by landing on Onboard as our billion-dollar startup idea after running ten hackathons, the completeness of our founding team, and a looming question: What minimum viable product (MVP) would we build?

In this ninth part, I'm sharing how we learned what MVP to build based on conversations with game studios who signed up for our beta and what happened to our dream of joining the renowned startup accelerator Ycombinator (YC).

Here's an overview of the full series:

• Part 1: How the Megacool journey began

• Part 2: Why we chose San Francisco to start our new company

• Part 3: Embracing Failure – How we tested 10 different startup ideas over 10 weeks

• Part 4: Hackathon 1-3: From AI travel agent to photo management adventures

• Part 5: Hackathon 4-7: The messy middle

• Part 6: Hackathon 8-10: Wow, that's megacool!

• Part 7: How we validated the Megacool SDK

• Part 8: How we built our co-founder team

• 👉 Part 9: From idea to live product 👈 YOU ARE HERE

• Part 10: Our bootstrapping hustle

• Part 11: How we raised $1.6m in funding

• Part 12: How we got acquired

• Part 13: From Alpha to Acquired: The product, growth, and business model journey

• Part 14: The emotional founder roller coaster

• Part 15: The epilogue: Reflections on the whole journey

The main challenge that game companies faced

The more time we spent talking to game companies, the more I realized I wasn't likely to leave gaming, the industry I had joined two years prior. Despite all the other hackathons being non-gaming related, gaming had found a way to pull us back in with Onboard.

Speaking with game companies revealed a universal struggle: producing a great game wasn't enough for success. Despite developers' hopes, players didn't come storming down the doors post-launch by themselves. 

More precisely, game studios' main challenges—after building a great game—were the rising advertising costs (CPI = Cost per Install) and getting their game discovered on a crowded app store.

Through our research, we'd learned that CPI was, on average, $3 at the time, but most mobile game developers didn't earn enough per user (LTV = Life Time Value) to justify that spend. To be profitable, you want to make more money than you're spending—LTV>CPI. In 2015, the total mobile ad spend was $70b, and gaming had the most significant portion of that spend.

By addressing this challenge, it was clear that Onboard had a considerable market opportunity. If we could build a turn-key solution that would boost a mobile game's organic growth, the effective CPI would go down, and hopefully, they would be able to launch a profitable and successful game.

We hypothesized that friends inviting friends to a game would have higher engagement, retention, and monetize better than the average new user.

If our product worked, small scrappy indie companies would increase their chance of success from the increased, hopefully viral, organic growth. At the same time, larger studios would get more new users on the same marketing spend.

This, in a nutshell, was our pitch. 

And it resonated. We had mobile game developers sign up from near and far. Large and small.

But we had yet to figure out exactly how we'd help boost organic growth. We had multiple ideas based on the opportunities we had flagged in our game dissection posts, but we needed to validate and test them out with mobile game companies.

London Venture Partners shared our summary article with their portfolio! We suddenly have people like the founder of Badland signing up for the beta!!! We have reached 28 registered beta testers and have 3 test companies ready so far. — Journal entry from 2015-11-04

Deciding what to build

In our journey to figure out how to increase organic growth for mobile games, we landed on two potential solutions after chatting with various mobile game companies: GIF sharing and incentivized invites.

GIF Sharing

We noticed that manually making and sharing gameplay GIFs was a popular but tedious activity in gaming communities. Taking cues from Fun Run, where players were keen to share screenshots but were held back by the manual sharing process, we hypothesized that players would likely share even more if it were more straightforward and fun. 

With the belief that "a picture speaks a thousand words," we thought a GIF could convey even more than a single image.

Our idea for "GIF sharing" was to make sharing moments from games a breeze by recording the gameplay automatically – since it's always in retrospect that you're thinking: "Damn, I should have recorded that"! Then, at the end of the game, players would see a 5-10 second highlight that they could share across various platforms like WhatsApp, iMessage, and Twitter, ensuring memorable in-game moments weren't missed just because a player didn't think to record.

Incentivized Invites

For "Incentivized invites", we drew inspiration from the impactful referral programs of companies like Dropbox and Uber. Instead of sticking to the traditional method of having users manually enter referral codes, we thought of building deep linking capabilities. This would let us know if Friend A sent an invite link to Friend B, and reward both Friend A and/or Friend B when Friend B starts playing the game, providing a heartfelt thank you for bringing new players to our game.

We found ourselves at a crossroads even though we didn't see direct competitors in the GIF sharing arena and only a few in the incentivized invites space (none of which were focused on gaming). 

Discussed focusing on GIFs or invites first. — Journal entry from 2015-09-24

Our big question was whether to build one or combine both features, especially when the common advice for startups is to do 'one thing' exceedingly well.

Another feeling of going down the roller coaster. We discussed how we can differentiate from [Incentivized invites competitor]. It's clear from those game developers we talk to that they don't know about them. There will always be competition. We need to stay in the race and believe and work towards bringing something new and better to the table. I think we can figure it out with enough time. — Journal entry from 2015-10-08

The "wow" factor from speaking with game companies was more around GIF sharing, guiding us to commence prototyping and building the MVP, with its visuality serving well for demos.

Thus, our MVP started with a simple capability: recording the last 5 seconds of gameplay and converting it into a shareable GIF.

Applying for YC

Notes from a practice session with a YC alumni who told us what to focus on in our pitch.

Earlier, I felt so empowered. I was meeting people. I knew what to say. I knew how to make them feel excited. I felt in control. Now, not so much. I started to write the YC application only to lose focus and belief. I lost confidence in what I was trying to do. What if I'm not doing what I should do? — Journal entry from 2015-10-02

The YC application was tough and long. Answering questions perfectly, such as:

• Why did you pick this idea to work on? Do you have domain expertise in this area? How do you know people need what you're making?

• What do you understand about your business that other companies in it just don't get?

• Please tell us something surprising or amusing that one of you has discovered.

We spent much time writing it and getting feedback from other entrepreneurs and YC alumni. It felt like an endless endeavor that was never good enough. Luckily, it came with a deadline. 

Eventually, we hit send.

HOLY SHIT! HOLY SHIT! HOLY SHIT! We're fuckings on to the YC interview! — Journal entry from 2015-10-28

We're having such a momentum and moment now. It's almost too good to be true. Waiting for the crash. Just hoping it's not a YC crash.. — Journal entry from 2015-11-06

I do what I usually do when facing a huge challenge. I calm myself ahead of what awaits by talking up the not desired outcome so the blow doesn't strike as hard.

The day before the day. Unsure of how I'm feeling. What's important to remember is that we aren't depending on YC. But we will do a good job during the interview regardless. We are going to be engaging. Trustworthy. And get across all the fantastic expertise we inhabit. We're making Megacool to solve our own problems [as game developers].

Before applying, I told Nick: "We do this to learn, and whether we get accepted or not, this is worth our time." I find myself thinking that if we don't get accepted for an interview that I want to even more "prove them wrong". — Journal entry from 2015-11-10

YC ran ten-minute interviews from their office in Mountain View. YC partners, including Twitch co-founders Justin Kan and Michael Siebel, conducted the interview. Despite extensive preparation, the rapid-fire questioning and abrupt topic shifts made the ten minutes vanish in a blink. I remember being so nervous that I felt my heart imploding.

In case it's helpful, here are the questions they were able to squeeze in during the quick chat:

• What do you do?

• How do you know game companies want this?

• Where do you see this in 5 years?

  • What if you are integrated into all game companies? Where are you then?

• Show us the demo.

• How will this make money?

  • Do you think game companies would use you if they knew you would provide their data to their competitors?

  • How will you compete with the large companies when they make this themselves?

• What's the equity split between the three of you?

Following the interview, a YC partner will call you if you're accepted into the program and email you if you're not.

A check-in I wrote to myself after the interview:

The mood was good throughout the interview. We talked well, but I completely forgot the time and didn't ensure we covered the most important things. They didn't like our business model, and we didn't answer the question: "Where are you in 5 years?" well.

I find myself checking my email all the time. I'm pretty nervous since I don't know what getting accepted will mean. Nick believes we can raise a seed round either way, but I'm quite unsure of it all.

Unfortunately, the partners did not believe enough in us, and we received a rejection email:

Well now we know: We need to prove them wrong. — Journal entry from 2015-11-11

It was a huge blow, but thanks to the YC application process and its focus on traction, we had both game studios eager, drooling at the thought of our SDK solving all their growth problems, and an MVP showing how GIF sharing could work. 

We were hungry to prove YC wrong.

Regarding the YC's concern: "We feel that at the point where you require large games to pay for this feature, they are going to build it themselves", was legit. I had met a PM from [One of the largest mobile gaming companies] at a growth event. I sent the PM our summary blog post and asked if we could meet. 

We met with  [One of the largest mobile gaming companies] today, and they showed up with four PMs and the CTO! That is a clear signal we're hitting a nerve with our product offering, even though they prefer to develop these things in-house. — Journal entry from 2015-10-21

We knew from experience that just because it's technically possible to build something yourself, the likelihood of doing so is very small. At least to start. 

One example of this was back at Dirtybit when we discussed ways to boost our Fun Run growth further. Whenever the ideas required development resources, they were postponed indefinitely. 

For instance, incorporating screenshot capturing with Twitter sharing in Fun Run, which eventually evidenced potent user engagement, took over a year to be prioritized but only half a day to be implemented.

Imagine what could have been achieved with the growth feature implemented from day 1? Or when momentum was at its peak? 

Another tailwind was the macro trend for game studios to buy vs. build, given it was so hard to build a good game in the first place.

Henrique Olifiers (Co-founder and CEO of Bossa Studios, and grateful to have him as one of our advisors starting the summer of 2016) said: 

"I love the idea of making sharing so frictionless and platform-independent -- all the things that hold back the likes of Kamcord etc. Rather surprised no one has done this before, given .gif is the #1 format games use to spread on Reddit and Twitter."

Going live

To qualify sign-up leads, we asked why they wanted access to Onboard and the email address. This created friction to avoid bots and random sign-ups, increasing the chance of high-relevance sign-ups.

After signing up, we emailed those passing our initial screening (see screenshot below). We prioritized games built on the iOS native platform as we couldn't support Android or Unity developed apps yet. 

Based on their answers, we'd ask for a meeting to learn more. From there, our goal was to narrow down to five gaming apps built on iOS native that could integrate and go live with us over the next few months.

Among our many meetings, one game studio really stood out: Candywriter. They had pioneered screenshot sharing from gameplay to Facebook, so much so that Facebook had to block them. 

Candywriter was a small game company of two very talented people. Their forté was image quiz apps. A few years later, they launched Letter Soup, and today, they are most known for BitLife. Their success at constantly creating new mobile gaming categories got them acquired for at least $74m in 2021.¹

In addition to being at the forefront of innovation, Candywriters games were built on the iOS native platform. Exactly what we were looking for! Candywriter was eager to work with us and ready to become our first user. They would integrate us into their whole game portfolio if it worked well!

Safe to say, we got to work on a stable beta version and documentation.

We decided not to listen to the common startup wisdom of doing one thing well. Our beta would include both GIF sharing and incentivized installs as we believed they worked optimally hand-in-hand.

By offering both, we envisioned providing game companies with a dashboard of insights² into their players' sharing habits, from knowing what moments from their games were most sharable to general analytics on the number of installs, sharing conversion, etc.

Speaking to Pieter Kooyman from MiniClip, today one of the largest mobile game studios in the world and publisher of Agar.io—one of the games we dissected—was a massive fan of the GIF sharing idea. He referred to it as a "Trojan Horse" (in the positive sense) since it seemed to be what made game companies most excited, but then they also got all these other benefits (analytics) from working with us.

We're truly an amazing team! We've been working so freaking hard. And tomorrow, we're integrating a test with our first customer! — Journal entry from 2015-11-30

Had our first integrations with Candywriter, and there is definitely room for improvement! — Journal entry from 2015-12-01

Very anxious about whether we can launch something stable enough for Candywriter to launch with. — Journal entry from 2015-12-03

Someone has uploaded their app to the fuckings app store with our SDK integrated! Someone trusts us that much! — Journal entry from 2015-12-08

And today that app is live! — Journal entry from 2015-12-09

GIF shared from the game What’s the pic? with our SDK

We were live. Finally! It had been six months since we moved to San Francisco and started the Megacool adventure. We:

• had gone from idea to launching our beta,

• were live in one mobile game, and

• had over 50 games companies on the waitlist!

We felt on fire, but, …

…when working on our budget back in February 2015, we'd estimated we had savings to last us for two years living in San Francisco. It would give us time to experiment, find our groove, and close funding. 

Since that cold February morning earlier in the year, the Norwegian krone (NOK) had weakened drastically against the dollar. Now, with handling our new co-founder, Tarjei, expenses, and the drastic change in the exchange rate, our NOK savings would only last us ~12 months from our real start back in June 2015.

Now, six months in, the reality of our finances started to creep in. We needed funding! I got a minor heart attack every time I opened my bank account. How hard would it be to fundraise? For how long could we really bootstrap? Could we bootstrap without burning out?

Continue reading:

A huge thank you to

CansaFis Foote

and

Leo Hepis

for providing feedback to an early draft of this.

1. https://www.gamesindustry.biz/stillfront-group-acquires-candywriter-for-at-least-usd74m

2. A note on privacy: We were extremely privacy-minded, and anything we tracked would be done with the highest regard for someone's privacy while keeping the sharing experience possible.