Target Audience for this article

  • Techpreneur/Wannapreneur, who want to launch a tech product
  • Developers with some idea and want to start

What you might learn from this article

How to approach launching a tech product (mainly an app)


There is no such joy as creating something and seeing people using your creation - unknown


Let’s set some base (Self Promotion)

I am in Software, Android Engineer in particular and Full-Stack when needed. I call myself a Product Engineer, an Engineer with a product mindset. Like you, I have a lot of ideas. Sometimes I brainstorm some idea with myself and cannot sleep thinking it is the next big idea, the brilliant one. But later next morning, brilliant turns out to be a better idea and after a few days, it becomes a good idea and most of the times, an ugly one.

But sometimes, some ideas remain brilliant even after a few days. And that my friend, is something you should pursue.

This article is my journey which might help you learn a thing or two. At the very least, it should motivate you to take action. If it does nothing, I am sorry, I tried.

Software Business

There are many ways you can build a business in the software industry. I tried with apps. Thrice till date. Two products/apps are dead and third was a huge learning and little money. The third was acquired by a company and I thought to share some learning of all 3 with you. Do you know app business selling is a thing? I did not know before. It was a trend a few years back but now too (2019) you can find a few companies buying apps for the sole purpose of business. We will discuss later in detail.

NOTE: This article might seem full of common sense at several points because it is. One Take-Away

I get it, you have an idea and you think developing an app OR a website is a piece of cake these days. I will get it done by tonight and by next morning, I will start counting $€£₹ Trust me, if you want one take away from this article. “It is harder than it appears”. I will try to share my learning and formalize in an actionable way so you can get benefit from it.

Journey - Idea to Implementation to Acquisition

I love IOT. It’s like cute babies from hardware and software marriage. I had an idea in smart light domain. I thought why every smart light company is developing the same kind of software and simply burning money? On top of that, they are hardware people and software is just not their thing. I wanted to develop a white label app for all possible smart lights. I ended up developing a famous app for Philips lights, HueHello and sold the business before I could make it a white label. Validate Your Idea These days there is a saying: if no product exists for a problem, it is because users either do not need it OR they are not aware of it. If you find there are no existing solutions, no one can prove you right or wrong unless you validate your idea. If there are other solutions to the same problem, why do you think yours is better? The real answer is no-one knows unless you ask the real users to use your product. In both cases, you need to validate. How do you validate is very subjective! Few people consider developing a full product before gathering feedback, few settle for MVP, few even use the prototype without coding anything and few use just a form to gather feedback. It is up to you how do you want to proceed, my only suggestion is to burn as less money as possible and play smart.

A few learning

Learning 1

Research the idea before jumping into the field. It is cliche but it is, for a reason. The catch here is formalized your research. I would say,

  1. Write down key points of your idea, your strength. “Bullet by Bullet”
  2. Research the existing solutions and note down a few top products
  3. Match your key strength with existing solutions
  4. Listen to your gut feelings

Sometimes you are solving a problem by making existing solutions better and sometimes, you are creating a whole new solution.

My Case (smart light apps) I researched and found existing smart light solutions are average at best. Available solutions are doing 1 or 2 good things at most. My idea was

  1. One app for all smart lights
  2. Smart lights app should make the process of controlling lights faster than the normal conventional switch
  3. It should be as simple as turning a switch off at home. A simple thoughtless press of a button.
  4. Some X factor I need to come up with

In total, I settled on 1 app for all lights, faster than conventional switch OR same, Simple as much as possible. My gut feeling said this seems to be enough to start the product and it is something worth pursuing.

Learning 2

So learning 1 says do some research, note down your key ideas and compare with existing solutions. Next phase is not to worry about the distant vision and chalk down every possibility. If you have a vision, it is good. Stick to it BUT remain flexible. If you do not have a vision, it is even better as it makes you agile.

Your whole focus should be on how to validate your idea. Repeat again, how to validate your idea. That’s it. There are a few fancy words for it, MVP (Minimal Viable Product), MLP (Minimum Lovable Product), etc.

For me it was: MWP * 0.1: Minimum Working product which proves my idea to myself. If you are wondering what does that * means, It means the * highlight of the product. It should have at least one key strength which I jotted down in the idea phase and 0.1 means simply the basic first 0.1 version of the product. I am stressing again, not 1.0 version but 0.1 version. In my case, learning 1 was a white label solution, faster than conventional switch, simplicity and X factor in smart lighting domain which I settled with scenic views and some light shows. Remember, you should always gamble on your strengths.

My Case I did some brainstorming and found white label solution is gonna take a lot of time in v0.1. Faster than conventional switch needs RnD which is possible after the base is ready. X factor can be built if the base is solid. So I settled with the simplicity of the UI/UX. I called it The Remote UI 0.1 release. I thought smart lights app should be similar to a TV remote. Simple, fast and familiar. Note: Familiarity with a design does not need the learning cycle and users can adapt to your product faster. Secondly, it should be aligned with your key strength. If Rock n Roll UI/UX is your key strength, go for it. Create that awesome alien design.

Summary of learning 2 is research, compare, plan 0.1 release with at least 1 key strength.

Learning 3

Discuss your idea with others. DO NOT KEEP YOUR IDEA TO YOURSELF. Keeping a secret is a recipe of disaster. I am not saying to publish it on a public forum (I am not against btw). All I am saying is to brainstorm with others. Discuss with people you trust, even though they are not from the same domain. Try to explain like they are 5 and notice their expressions. Listen to their negative feedback. Understand why are they opposing your idea.

One take away here is most of the people won’t get it and they will oppose. Do not get demotivated. Not everyone is visionary/business-minded/risk-taker and gutsy. But they can be your potential users, try to understand why are they saying what they say. Discuss with your close friends. At the very least, listen to your gut feelings. There is no right or wrong here. If other arguments could not convince you why not to pursue, just do it.

My Case I am fortunate to have a few technical and like-minded friends. I called them and discussed the idea. Honestly, I did not get much negative feedback but I could clearly see they do not understand the domain much. But not having strong opposition made me pursue.

Learning 4

Revenue and monetization. Everyone has different views about this. I will share mine. If your idea has the potential of becoming the next everyday product, you might be able to get away with this. Most of the superb businesses in history started with a problem solving with a no-money mindset and they became the most successful software businesses of the 21st century to date. But you know the idea more than anyone else. If you know the mass public is not your audience, and you are building something for a niche (smart light owners in my case). YOU MUST HAVE SOME MONETIZATION PLAN IN PLACE.

Nothing is written in stone and you are not forced to follow only one approach. In fact, you have to be flexible to pivot from time to time. But the reality is you have to ask this question to yourself: How will you make money out of it? It does not have to be 100% accurate but if you can not find even single reason why people will pay for your product, DO NOT pursue it is my advice. If you can not find a single way, but you anyway want to try the product. You are lying to nobody but yourself. I am not saying your product cannot earn at a later stage, I am stressing having a monetization plan in place helps you making the decision and prioritizing/testing assumptions faster. Try to answer this format:

People will pay for my product {product_name} because … … ….. your reasons here….. ….. ..…

Do not fill it with the vague statement “because I am better” “because I know the domain” “because I have money”. Try to be as specific as possible.

In my case, I thought People will pay for my smart light product because it is better than existing solutions:

  • It is simpler and it adds value to their smart light investment
  • No App is doing remote based UI
  • If they have spent ~$100 on lights, they can spend a dollar OR two on the app controlling them.

There is no such joy as creating something which people pay to use - unknown


Learning 5

Now you are all set and you know you want to pursue the idea or not. Next learning and very important one is JUST DO IT. Do not overthink/brainstorm. Start the development as soon as possible. Do not wait for perfect flows. Perfect illustrations. Perfect design. Perfect architecture. Perfect time. Perfect Love. Perfect person. Perfect Kiss.

Life is imperfect, so should you be. Perfection is the enemy here. You can not write perfect code. There will always be something you can improve. Remember, you are here to prove your idea. 99.5% you think about a product in a very different way than the users. You are too much attached to it and you can trap yourself in your thoughts. That’s common. The good part is you are aware of it and you will take action to avoid it.

On the contrary, do not build a crappy product. Do not trade speed for quality. Readable and Reusable code goes a long way. Writing readable code takes a day or two more than writing crappy code. If you are a developer yourself, you know what am I saying here. If you can not code yourself, my advice is to do not hire someone just because he/she is asking for less money. If, have some technical domain knowledge. One trick is to give a sample project and measure how much time a change requires. If small changes need a long time, RUN.

My Case I am a developer myself. My strength is Android. I thought not to build ios app to prove the idea and go for only Android. Remember, play on your strengths.

Learning 6

Without analytics, your app is like a muted TV in a blind school. No one knows what is happening. Spend some good time on what analytics are important for you and what not. If you understand the funnel of your product users is going to experience, it is really good. DO not collect everything but do not miss anything important. If in doubt, gather as much data as possible. Ensure you are compliance with laws.

Do not forget. Crash Reporting and Performance. There are many tools which help in debugging the crashes and pointing if there is any performance bottleneck.

My Case I collected anonymous app data which helped me understand what users are doing in the product but I did not have individual data. It is your call how you want to set it up. There are many free tools like Firebase available. More on it later.

Learning 7

I presume you have some product ready and you want to launch it. Do not forget Friends and Family. Remember learning 3. You discussed your idea with friends and family. It is time again to poke the same people. This time, be nicer and bring chocolate/ice-cream when you show the preview of your real working product. Take feedback, make bullet points and rework/improve if every person pointed the same flaw. For mixed feedback, listen to your gut feeling. If your X factor is not as good as you thought of, try to brainstorm what can you do quickly. Most of the times, our brain works fast under pressure. DO NOT ARGUE WITH PEOPLE. They are trying to help you.

My Case I showed the product to the same friend and he pointed out a couple of UX flaws and discrepancies. It was my decision what to improve and what to keep for later. Remember, gut feeling.

Learning 8

It is time to soft launch, getting the first few real users, NOT YOUR FRIENDS. You must be feeling excited thinking hard part is behind you. Let me burst your bubble. It is going to be harder. Coding is ~20% of the Job. Yes, you read it right. I absolutely understand it is hard to develop something beautiful but trust me, having coding skills only won’t launch a product which people love. 80% of the Job is left. The good part is it changes with software lifecycle. In the early days, a soft launch is very crucial. For a growing company, marketing takes over and for a grown company, innovation, customer service, etc.

Find the platforms where you can ask people to test your product. Do some research and note down options like product-hunt, Reddit, beta-software websites, public forums of your competitors, top-#-product for your problem, etc. Get as many early adopters as possible. Without the first 100 users, you can never reach first 100,000.

My Case I chose Reddit. There was a channel discussing Hue lights explicitly. I posted there a humble request that I am the Founder/Developer of this Hue App. I would like to gather some feedback so I can improve. I offered the app free to first 20 users and discount to few. People did help. There was a bug in in-app purchase. Some guy reported, I fixed, apologized and refunded the money.

Learning 9

It is time to launch it on play store. Upload the app. Take a break. Celebrate. Keep an eye on performance and crashes.

Learning 10

Now you have uploaded the app to the play store but you do not see people downloading your app. At least not what you expected. You dreamed of users in thousands but you see in one zero. That’s because nobody knows who you are and what your product is. It is time to spread the news. Depending on the audience, you need to use as much free advertisement as possible. Ask your friends and family to share the product.

  • There are a few free and few paid medium to spread the word
  • If possible, hire a freelancer who knows what to do
  • If you are doing yourself, stick to one or two ways instead of juggling between everything

My Case I figured friends and family will not help in my case. People with Hue hardware can use my app. So I went into stealth mode. I asked a friend and he suggested.

Now wear the user hat and search for the Hue App. I went to play store, searched Hue App and similar keywords. A couple of apps came but not mine. Search Engine works on a few parameters but when your app is very new, all it knows is keywords within your description. Once people start downloading your app, search engine learns. So, my only hope was keywords.

I tried to play with the keywords inside the description and used Google Adword “App Install Campaign” In short, I paid to get people to install the app.

Other than that you need to create some brand name. I searched “Philips Hue Apps” on Google and many blogs came. I simply visited so many and commented about my App wherever possible.

Learning 11

You might have planned many features and you are eager to add more features into the app. Hold your horses. This is the time to validate your product not validate your idea. Validation of the product can be done by a few mediums:

  • Data You are tracking
  • Daily Active Users
  • Weekly Active Users
  • Monthly Active Users

It seems the data you are tracking plays a very important role now. It is the quantitative form of the qualitative product you just shipped.

My Case I was tracking how many times people are switching lights on and off. How many times people landed on a feature and how many times they used. It helped me understanding which feature people are not using at all and which one is becoming the most used. All if this information will help you planning what next. For me, it was surfacing a few features better. I believed in them but thought people are not using it because they are not visible. I could not be more wrong. The features were simply not useful. Improved them in future.

Now, the app started generating data. I could understand more about geography, user segments, etc. Everything combined helped to shape the next few versions.

Learning 12

If you do not measure it, you can not manage it. In this context, you need to know one very important piece of data.

COST OF ACQUISITION AND REVENUE PER 1/10/100/1000 users

  • If you are spending money to acquire users, you need to know your cost per install. It varies from the platform to platform you use.
  • Secondly, you should have some structured way of finding the revenue per x users

Once you know both of these data points, you get to know are you making OR losing money. It is very common in the software business to lose OR just break even in the early days. But if you do not know this information, you simply can not improve it.

Tweak the cost of the product a little bit and gather more data. Check what people are willing to pay.

My Case I was gathering two data points. CPI adword showed and how many people converted to paid users. I figured I was making $5-10 per day.

There are many ways to optimize this ratio. Your goal is to make more than you spend.

  • Spend Less (decrease CPI → less installs → Not good)
  • Make More (Increase Cost → Test the price → If people pay, good otherwise NOT Good)
  • Improve the funnel (In my case, I observed 1 out of every 6 people who reached the payment screen becomes a paid user. THE POWER OF DATA.

What I Did I kept testing the price.

I increased ways people could reach the payment screen. Gradually more people started converting.

I was charging $1.99 when apps were launched and $6.99 when apps were acquired. It is because of features added with time and improved funnels.

Learning 13

This is the world of rating and reviews. If you do not have internet credibility, people do not trust you. It is very true for apps.

Request your users to rate your app in the play store BUT do not bug your users to do so again and again. Behavioral science and psychology plays more role in software than you imagine.

My Case I requested users to leave a review when they became pro. Just after they paid, showed a success dialog with happy graphics and nice animation. Literally Wrote “We need your help” Please rate us and help us grow. Who does not want to help the company they just paid for?

Learning 13

I get it, You are running a business to make money. Keep in mind, you can not do this if your users are not happy. Customer satisfaction plays a huge role and goes a long way.

My Case

  • I responded to each and every review < 5 stars and requested the users to email
  • Responded to each and every user email with no pre-defined format. Every email was personal and users felt appreciated. Can’t say the exact number but more than 40% converted from 1 star to 5 star
  • Refunded every user who asked for a refund

At the time of selling the app, it had 160k downloads and 2k ratings with 4.5 average ratings.

Learning 14

Your users know better. Set up some way of asking them ideas and suggestions and be clear when/if are you going to implement them. They feel appreciated and valued. You get free ideas to work on. Together, you improve the product.

It might not be a good fit for your product but worked for me.

My Case I set up a google form and added a link in the app asking users to share ideas. I was very explicit in the description of the app that we do not implement every idea you share. I tried to respond to as many users as possible. It was a routine every week to sit and read user ideas + respond. 10-20% feature ideas came from users. Keep in mind users can not state solutions. They are good at stating the problems. DO NOT follow their solutions blindly. Try to find the problem they are facing and solve it in your way.

Learning 15

Know the tools. I used A/B testing, remote controlled switches for new features, remote controlled app update dialog, etc. I believe it helped in improving the user experience.

Learning 16

“Web Presence” Ask yourself: Can I add more value to my users’ life? I believe almost every company can get benefit from the good content they generate. Every good company is writing good content in the domain they work. E.g Money transfer companies write about taxation laws, conversion rules. Dating companies write about “how to prepare for a date”, “What to ask” etc.

For us, it was “How to use Hue lights and Smart lights well” How to use HueHello features, etc. Started a medium blog and added a link in the app so users can visit.

Gradually started shooting a video if needed for any feature (It need not be professionally edited). and created a youtube channel.

Setup a website with app information and content links.

Everything combined helped in web presence and improved app downloads.

Learning 17

I will call it a mistake 1. Know your users and act accordingly. I believe localization is very very important and we brought a little late. If you have an international customer base, localization will surely help. DO NOT IGNORE IT. People love their native language way more than you can imagine.

Learning 18

Mistake 2. Professional content writing. I did all the content writing inside the app and on play store pages. I believe taking help from professional content writers can help clear the message and increase downloads. Understood it very late.

Learning 19

Mistake 3. Play store page is very very important. It is the gateway to your app. NO one is going to come inside your home if the gate sucks. Use good screenshots, clear messages, localization here as well. A/B Test if required.

Learning 20

Keep trying and do not give up. But know when to shut it down if not working. Do not ask a dead hen to lay the eggs.

Hope you learnt something. If you did, share and spread the word. If not, let me know what would you like to read?

Subscribe to newsletter and keep learning good stuff

Email