AdDuplex Post-Mortem. Part 6: Pivots
This is part 6 of the AdDuplex (2011–2023) story. AdDuplex was the widest ad network for Windows apps. You can read this part in isolation or start at the beginning.
With the business growing in a linear fashion and all our “traditional” attempts of increasing the velocity failing, we needed to do something different. We’ve made all our invested money back and with that in mind coasting on the linear growth train didn’t seem like an option. In retrospect, I guess, I could’ve tried to return the investor money (with some interest) but to be honest this didn’t even occur to me at the time. And I’m not sure what the response from our investors would be. I decided to shake things up beyond the core business. And that’s what we tried to do between 2014 and 2018.
Those attempts are what I’ll cover in this part. But first we need to address the most frequently asked question which is very appropriate for this part…
Why didn’t we expand to Android and iOS?
When I started AdDuplex in 2011 it would very likely be a very good idea to start on iOS and/or Android. Chartboost started roughly at the same time and when I visited their first(?) San Francisco office in 2012 they were a team of about ten people. They were addressing a slightly different industry and in a somewhat different way but from the birds-eye point of view we did roughly the same thing. But our genesis stories were what made us approach the market the way we did. They came from the mobile gaming industry, and I was a .NET developer. Naturally, they started a network for iOS/Android games, and I started a network for Windows Phone apps and games.
AdDuplex wasn’t conceived by MBAs in a conference room throwing 50 random ideas onto the whiteboard and then eliminating 49 of them until THE ONE was left. It was birthed organically out of a directly observed need by someone in the Windows Phone community. Additionally, the timing was nearly perfect. In 2011 iPhone was 4 years old and Android was about 3. When AdDuplex launched Windows Phone was about 3 months(!) old. We were perfectly positioned to be the dominant player in an area which was a priority for one of the largest corporations in the world. So, I still think the bet on Windows Phone may have been an unlucky one, but it definitely wasn’t a stupid one. We very consciously focused on this market for the next 2–3 years.
Not many remember this, but in 2011 both Gartner and IDC predicted that Windows Phone will overtake iPhone in market share by 2015. And who am I to argue with Gartner? 😉 Especially, when they say what I want to hear…
By 2014 it became clear that this won’t happen. Neither in 2015 nor, most likely, ever. At the same time, not many remember that while Windows Phone was a distant third in mobile it wasn’t a total fluke in absolute numbers. And even in market share it showed double-digit penetration in some markets. Just not in the US where most of the pundits live. Anyway, that is not the point of this story.
We realized that we needed to do something and going Android was on the table. By that time, I’ve attended a lot of conferences and other industry events, talked to a lot of developers, and it was clear that the competition for that limited ad space on that side of the fence was massive. We would need to raise tens of millions to even attempt to play. So, I turned to exploring a familiar path — going through the community that already knew what AdDuplex was.
Many Windows Phone developers weren’t exactly swimming in money on the Windows side and were exploring other pastures. Maybe we could target just them for a start. Game developers using cross-platform engines like Unity were already doing it and, in a way, were lost to us (in the scope of the “cheap” market entry plan). That shifted my focus to app (as in “not a game”) developers. The Microsoft developers trying to pivot to cross-platform were leaning heavily to using Xamarin as their technology platform of choice. So, that sounded like a natural point of focus for the first impact. However, after talking to some developers and, eventually, to Xamarin leadership I realized that most of those indie Windows Phone developers were using Xamarin not for their indie Android/iOS apps, but rather for launching their cross-platform contractor careers. Meaning, in most cases, that their customers were some companies with little-to-no interest in running ads in those apps.
Unlike when launching AdDuplex itself, I did too much research before deciding not to proceed with the expansion to Android. We will never know what would’ve happened if we did, but at the time it seemed it would be easier to come up with something new for those markets than trying to be ad network number 137.
And something new we did.
Fidesity
I think this was pre any research into actually expanding to other platforms, but it was under pressure from constant questioning about our overreliance on Microsoft that I came up with the idea for Fidesity.
Back in the day, for some bizarre reason there were no refunds in the app stores. And some bad actors abused the system poisoning the well for all the rest. So, after reading some blog post or a chapter in a book on how “money-back guarantees” increased sales, I decided that we can build a service like that.
This was the pitch:
In-app purchases are driving most of the app revenues, but a few bad actors poison the popular perception for everyone else. As a result, people are wary of freemium apps and free-to-play games and good developers suffer. Fidesity provides a trust seal and money-back services to eliminate spending fears and help publishers earn more.
We’ve approached this one in a “timferrissesque” fashion — made a landing page and a signup form but didn’t write a single line of code for the actual product before figuring out the demand.
We made some mockups, a couple of t-shirts to wear at the events, and these amazing “do not disturb” signs:
After trying to attract customers for some time we realized that this wasn’t top of mind for most of them and decided to scrap the whole thing. Luckily, we didn’t invest much into it and there was no substantial harm done to our finances. And I also used those doorknob signs for years to come.
Additionally, it turned out I wasn’t the only one who thought the no refund policy reality was insane and Apple and Google were forced to revise their stances on refunds. So, by scraping Fidesity before we even started the development, we dodged that bullet.
AppRaisin
AppRaisin was our most successful “pivot”. Primarily because it was not a pivot but an extension to our core business.
It was a Digg-like (for those who remember it) or Reddit-like (for those who don’t remember Digg) service for app news. Publishers or other users posted about the new app and game releases or updates and everyone else voted or commented on them. There were a “rising” and latest posts feeds, and also a “spotlight” feed where we posted sponsored or otherwise notable items.
We hired Lazyworm Apps (RIP Atta) who were one of the best Windows app developers at the time to do the design and initial UI, and assembled a dedicated team of 3 to work on the project. It was one of the best rated and popular Windows 10 apps, but Microsoft reneged on the promise to upgrade all Windows Phone 8 phones to Windows 10 Mobile, and as we found out later shadow banned AppRaisin in the store for some stupid internal reasons.
All of this resulted in AppRaisin not reaching our estimated usage thresholds for breaking even. But users loved it, media loved it, and we loved it too. We decided that maybe it was the product we can take cross-platform. We even found a partner potentially willing to invest in expanding to Android. They were a savvy game developer and asked us to conduct series of experiments and optimizations to prove that we can scale it by pouring money into user acquisition. Basically, that if we were to acquire users for AppRaisin via paid channels we could make more on the other end. We tried, we optimized, but at the end of the day couldn’t make the math work.
With a dedicated team of three this wasn’t a negligible expense for a company with 15 people total, so we stopped development at first and after some time discontinued the service altogether.
I’ve written a long post-mortem for AppRaisin right after its demise. It’s one of my most popular posts. You can read it here.
AppStretch
In parallel with AppRaisin we’ve started working on another new product we called AppStretch.
The pitch for AppStretch was this:
AppStretch is a platform to crowdsource feature ideas for your great apps, crowdfund implementation of these ideas and crowdpromote their availability to users.
It was based on the premise that app and in-app item prices were in a race to the bottom, so nothing but mainstream apps and games could survive in the app economy. At the same time, niche apps had a core of enthusiastic customers who would pay way more than 99 cents to get some additional feature or improvement.
With that in mind we decided to build a sort of a crowdfunding platform for apps with a twist. In addition to “classic” crowdfunding users could also suggest the features they want to crowdfund. And those who don’t want or can’t commit money to seeing those features implemented could pledge support in social media marketing (Thunderclap was a thing right around that time).
This was our most ambitious project after AdDuplex itself. And it was cross-platform (platform-agnostic) from the get-go which made our investors happy. While we knew exactly what we were doing with AppRaisin, AppStretch presented a lot of new challenges. From investigating crowdfunding regulations to actually implementing charging supporter credit cards months after the support was pledged.
Braintree (our trusty payment processor for AdDuplex) turned AppStretch down (crowdfunding is a big red flag), and after a lot of attempts we finally found a local processing partner who gave us greenlight on what we wanted to do… Only to renege on it when time came to actually charge the cards for the first time.
By that time, it was clear that AppStretch didn’t get the momentum we hoped it would and failing to provide the most complicated part of the whole thing with no obvious options on how to salvage it was the last straw. We’ve invested a lot of money (for us) into AppRaisin and now AppStretch. And our bank account balance didn’t look as calming as it did just a few years ago. The decision to “fail fast” with AppStretch seemed to be the only logical choice.
Extendee
Our pivot attempts failed, Microsoft had finally officially dropped the ball on mobile, and “big Windows” didn’t attract as much indie developer attention as Windows Phone did in the early days. We’ve burned our “fat” on the pivots and our core market and core product were on the declining side of the curve. We had no resources left to do almost anything but ride it into the sunset.
On the technical side we were back to just me and Paulius, and while there were occasional changes and fixes, we needed to implement, we weren’t extremely busy with day-to-day duties. So, it made sense to make one final attempt at doing something productive (with our new financial constraints in mind).
Over my career as AdDuplex’ CEO I’ve traveled to infinitely more events than I ever did before (which was zero, if I remember correctly), and time and again I realized that I was at the same airport and often on the flight with others travelling to and from that event but only found out about it later. While there were a lot of services trying to improve the “conference experience”, they all focused on the “at the event” part. And that’s how Extendee was born.
Extendee was a tool to connect you with other people going to the same event before you even get there. The original name for Extendee was premeetup and I almost bought a domain for it (or maybe I even did buy it) only to realize that “meetup” is not actually an English word but a trademarked name of that business. Hence Extendee — extended attendee experience.
I figured it will make money from various affiliate opportunities around travel but first we needed to test if people would actually use it. Luckily, Microsoft MVP Summit was around the corner, and we did a push to have an MVP (sic!) in time for that. The MVP program leadership greenlit the endorsement of Extendee as a useful tool for attendees and it got promoted to everyone attending.
Even though it was rough around the edges, it served its main purpose really well. Lots of MVPs connected on the way to the event, shared taxis, etc. Yay!
After the initial success we continued improving the product while we’ve seen several attempts of attendees of various events self-organizing their extendee experiences. It became clear that it can only work if endorsed and promoted by the event organizer.
We had two major tech events round around the corner — Microsoft Build and Google I/O. We set our sights at conquering those. And… were eventually turned down due to “privacy concerns.” Oh, the irony! While I understand that by 2018 it became less common to enter your sensitive information (flight data in our case) into random websites, hearing this from Google of all places was funny. It wasn’t fun though. We failed to secure any meaningful events in the next months and decided to just leave it be.
Not long after I became the sole factual employee of AdDuplex and potential issues maintaining zero revenue product seemed like an unnecessary risk, so I shut it down. Could’ve waited another year and blamed it on COVID! 😉
In the next part we will wrap things up and reflect on the ride and lessons learned.
—
P.S.: with the announcement of AdDuplex shutting down and this blog post series I keep getting asked about what I’m up to these days. Check out marker.js if you need image annotation in your web app, and the upcoming JavaScript diagramming library.
