Paul Graham, co-founder of Y-Combinator has said entrepreneurs need to, “build something people want”. In the spirit of building in public, I am planning on writing posts that shed light on all things related to our Android mobile app, Signils.
Our app was launched in the Google Play Store on 3 September, 2020. Since then, it’s been downloaded 111k times, is on 22743 active devices in 145 countries, and has 4214 registered users that have added 17035 Bluetooth devices to Signils.
Our original business concept
We would combine education with credit monitoring. Split (my-split.com) would help people with threats to their online safety and their financial lives. It would be a community integrated with Experian’s solutions.
We spent 2019 working with offshore developers. Since the contract was for 7 months, we should have been code complete in June. In reality, we saw little progress. Instead of pulling the plug, I pushed harder. This resulted in a launch date that passed without explanation.
By December 2019, I had to consider alternatives. To hopefully have a better outcome, I inadvertently used the innovation process (see graphic below). Shilpi Kumar’s Medium article includes this great quote, “Design thinking is a lens that helps us challenge the basic premise of the idea while keeping people at the center of our thinking. Ideation and design are steps that define and make the idea real. The combination of these four things make up the innovation process which creates value and ultimately lead to impact. ”
Where is the Blue
Instead of building a Bluetooth card skimmer detector (long story) which was my inclination and desire, I decided to look at a wide range of existing Bluetooth apps. I concluded that we could “build something people want” by combining key features into one app and doing it with a better user interface and experience.
Those key features included visual techniques to view devices, a “poor man’s” device locator, and a way to find devices that utilizes signal strength (aka SSI). Later, we decided to add the card skimmer detector. We also decided to support a wide range of Bluetooth technologies including Classic, Bluetooth Low Energy (BLE), and Beacons (Beacons, iBeacons, Eddystones, and Ruuvitags).
While I was an iPhone user, I decided to build an Android app.
- Android has the lion’s share of the smartphone and tablet market
- Android gives developers more control
- We needed to control our costs
I named it, “Where is the Blue.”
I am not a developer. I also have not been able to find technical co-founder capable of handling our software engineering work. I researched development firms with experience in iOS and Android app development and had several discovery calls. Another valid reason for hiring a professional firm is they typically have a deep bench of technical resources. It’s common for them to be capable of providing: designers, backend/frontend engineers, database administrators, quality assurance people, and experts on specific technologies like Docker.
I reviewed multiple time and material estimates and did not ask for fixed cost bids. I ultimately felt the most comfortable with Vironit in Ukraine. My previous experiences formed my expectations and concerns. Their detailed estimate, contract, and robust project management processes were convincing. You won’t find zero terms (my previous developer) or contract terms that only favor Vironit.
You may be curious about the initial estimates we got from Vironit. They ranged from $30,000 to $46,000. Time estimates to complete the project ranged from 2.5 to 3.5 months. Contract signed, we started the app design process with my initial mockups.
The designers took my ideas and turned them into what we now know as Signils. You can see better visual elements like images, icons, and colors, a sort function, and either a one or a two-column display. Many of our users have ten or more Bluetooth devices, so these features make managing Bluetooth devices much easier when using Signils. Our current top three Bluetooth device leaders have 87, 67, and 65 devices, respectively.
The original design included the My Devices screen locked down to three devices and a registration feature to unlock it. That also meant users would need to log into the app with a username and password. I did worry quite a bit about this feature. I was concerned about making people log into an app that was meant to give them access to their Bluetooth devices. I’ll post about our experiences in production and what we’ve learned from our user community later in a future post.
Right before our launch, the USPTO denied our trademark application. Evidently, you can’t include an existing and protected trademark in your mark. So, “Where is the Blue”, “WITB”, and “The Best Way to Manage Bluetooth” all got denied. It makes sense, but I had no idea there was a problem. That forced us to abandon our domain and app name and rename it.
Signils is actually a combination of the word “signals,” representing wireless communications, and “nil,” or no. The intent is “no signals” but gives a nod to the fact that we focus on wireless technologies like Bluetooth and in the future Wi-Fi and RFID. Users can block suspicious devices to prevent communications. Besides, Signals was already taken.
We also quickly realized that we needed a new production server. It’s not a good idea to try to run development and production on the same environment. In fact, the best way to operate a server environment is to separate development, staging, quality assurance, and production. To avoid a manual rebuild of our backend stack and delay our release by weeks, I cloned the EC2 server, did some DNS magic, and there you have it.
The developers updated the build, uploaded the APK to our Google Play Developer Console, and we were now considered “launched.”
Our actual costs from the start of the project in January through our launch in September were $68,477.20. Those costs included the Bluetooth credit card skimmer detection functionality that was not originally included in the scope and our backend management application.
Our development costs from mid-September through November (we will be billed for December soon) were $14,168.60. That work includes bug fixes, enhancements to Bluetooth discovery and identification, modifications to the backend, ads, and work performed on the trial/free/paid version.
Initially, we got no organic traffic to our app listing. We tested both a Google Ad and a Google App Ad. We were able to generate traffic to our listing with the app ad that resulted in downloads. As we increase the ad budget, we saw increases in our number of downloads. Our average went up over time and we’re now averaging about 1,000 downloads per day. Our spend on adds was $5,253.98 or $1,313.50 per month. If you like metrics, that’s $.05 per download or $.23 per active device.
If you are an entrepreneur or an aspiring one, find a partner or a co-founder that shares your vision. Vironit has demonstrated high levels of performance, commitment, and capability. Their design capabilities are awesome. They give me input on what they think is useful and what isn’t.
One of the things I knew but didn’t understand was that once we launched, our project was “complete” which meant our original team members move on to other projects. Fortunately, the fact that our contract was time and materials meant we could get the team back working on post-launch support. Had I realized, I would’ve made better after launch support arrangements.
Our developer struggled to implement In-App Purchases (IAP) but did get there. Vironit also didn’t have much experience with ads. This is another learning experience. Have a monetization strategy in place before you start development. It will help drive your search for the right specializations to complete your project. If you’re a technical founder able to code, that’s great but that is not one of the skills I bring to the table.
We did supplement the team with Halkwinds Technologies to get more experience with the Bluetooth protocol, ads, and MixPanel.