How my app grew by over 1M users in one month

Building and promoting a new consumer product is one of the most challenging things you can do as an entrepreneur. While there are many approaches on how to design, test, build and promote apps, usually they don’t seem to bring real results. 

Then you start wondering, maybe it’s the product? Maybe there’s not enough market fit? Or is it bad execution? Or maybe we should grow the marketing/branding budget? Maybe we’re not targeting the right audience? Maybe we should build more features!

When you start questioning everything, things usually get even worse. You start defocusing from the main goal and start wasting energy and money on all kinds of wide approaches.

The worst is when you think it’s all a matter of growing your marketing or branding budget.

Your goal should always be one — improving customer retention. For those who are not familiar with what is customer retention, click here.

To make my point clear, I’ll let you in on a story I heard from a friend of mine, who’s the Co-founder and CTO of a very successful productivity B2C company. In 2012, they’ve released the first version of their app to the Google store and a crazy thing happened. A few days since the launch, 500K users worldwide downloaded the app. The reason for that crazy growth was due to no good apps in the productivity space back then. Over the next few months, they’ve grown to a few million users and raised over 5M dollars from VCs. 

Four years later, they still couldn’t reach a decent business model, and he realized that despite the big numbers, there were very few actually using the product long term. So he decided to dig into the data and look for the reason. He found out that retention was very low, and what’s even worse, that it hasn’t improved much in four years! That’s when it hit him to focus on retention instead of user growth.

Back then, VC’s poured millions of dollars into companies with large user growth because they didn’t know how to deal or measure the crazy scale that mobile app stores and websites brought with them. Today the case is different. The first thing you’ll need in order to raise money in B2C is to show retention growth. And there’s a very good reason for that. Back to my friend’s story, with no retention, it didn’t matter how many users have downloaded their app. After a week, 95% of users have stopped using the product. So even if they had a billion users, after a few weeks all it would be is just a number in their database.

If you have 100K users using your product every day, it’s 100X more valuable than having 100M users using your product once a month.

Most importantly, once you’ve reached a decent retention rate, you can be sure that your marketing budget will lead to a sustainable growth of your product and business.

Too many startups begin with an idea for a product that they think people want. They then spend months, sometimes years, perfecting that product without ever showing the product, even in a very rudimentary form, to the prospective customer. This is where the Lean startupcomes in. In short, Lean Startup is a methodology that every startup is a grand experiment that attempts to answer one main question — “Should this product be built?”. 

A core component of Lean Startup methodology is the build-measure-learn feedback loop. The first step is figuring out the problem that needs to be solved and then developing a minimum viable product (MVP) to begin the process of learning as quickly as possible. Once the MVP is established, a startup can work on tuning the engine. This will involve measurement and learning and must include actionable metrics that can demonstrate cause and effect question.

So that’s exactly what we did at Tiv.ai, with a few of our own twitches. FYI, Tiv.aiis a startup building an AI memory assistant which I’ve co-founded in January 2017. You can check out our product here.

Here are the steps we’ve taken in each iteration of building our product:

1. Define the most important product assumption

2. Design and build an MVP of how this assumption should be tested

3. Target early adopters to test our MVP

4. Apply the test results on the product

5. Repeat

This is how our growth looked in the first year (2017) of iterations:

User activity from March 2017 to January 2018

Slowly but surely right? Now let’s do an example together and see what happend after enough iterations.

1. Define the main assumption

We believed that there are no decent reminder apps that people actually like to use. The main reason in our opinion is that there’s a lot of friction in setting a single reminder. Either you need to fill a long form on a mobile app, or naturally ask an assistant like Siri, but realize she doesn’t understand 50% of your requests. So that’s when we defined our most important product assumption  —  If we could achieve understanding for almost every reminder request in natural language, users would use such a product long term.

2. Design and build an MVP

Since our assumption is focused on NLU (natural language understanding), we decided to focus solely on that. No branding, UX or other features. Just focus on improving the understanding. We’ve hired data scientists to build a state of the art NLU algorithm for understanding complex reminder requests. Secondly, since all we’re validating is this assumption, we’ve decided to build the MVP as a chatbot on Facebook Messenger, instead of going through the long and annoying process of building a mobile app. 

Please note! If we were to build a mobile app, this would not add anything to testing our assumption, and make our MVP longer and more complicated to design and build. Moreover, it might have even defocused us from the main assumption. For example, what if users just don’t like to use new apps anymore? We might’ve concluded that our assumption is wrong even though it was for a whole other reason. 

It’s important to narrow your MVP as much as possible, so there are no distractions from your main assumption.

3. Target early adopters

We needed English speakers since our algorithms only supported it. Also, we believed that millennial moms would eagerly want a product like this, since they’re always on the move and very busy, while constantly needing to remember things. So we targeted some Facebook pages (with no budget) which were based on a community of moms, and successfully brought onboard a few hundred beta testers to try it out.

4. Apply the test results on the product

After our first iteration, we’ve learned the following:

1. There were much more ways to ask for reminders than we thought. But users extremely enjoyed the ease of setting reminders with a simple request.

2. Users don’t always ask for reminders in a single request but break it to a few steps.

3. When working with chatbots, users would like the assistance of buttons to make it faster and easier to handle.

With these results, we went on to our next main assumption, which was to add buttons to the flow of setting a reminder (conclusion 3). And guess what, that assumption also turned out to be true. For more proven assumptions, you can read my article on How to improve your chatbot.

Little by little, we improved our overall product and retention rate on a weekly basis. We never went on to tackle more than one assumption at a time. Slowly but surely, we started discovering users who were with us for over 6 months! After a year of weekly iterations, we’ve finally decided it was time to launch our product. We’ve reached a Week 1 retention rate of 92% and Week 4 of 16%. It’s way above the market standard which was enough for us.

We’ve published our chatbot on FB Messenger on mid Feb 2018 and within a month, have grown by over 5,800% as you can see below. 

Daily new users from Feb 17 to March 17

This was mostly due to delivering a lean product that we knew people enjoyed and would recommend to others. Since then, we’ve grown to over 1M users worldwide and are growing by tens of thousands of users a day. 

User activity from Jan 01 to May 01

We’re continuing to work with this methodology, and it’s proven a success every day. 

Not only did this methodology help us focus solely on what’s the most important set of features users want, but it also helped us filter out features we believed are valuable, but that are actually not. 

As said with customer service that the customer is always right, the same goes for product development. Trust your customers, listen to them and engage with them, understand what they want and don’t want. Never build things out of your own intuition unless it’s for assumption testing. At the end of the road you’ll reach one of two conclusions:

1. The product does not have enough market fit, time to move on.

2. You have a product people want. Good job, you’re on the way to building a company!

Either way, you win.

Bots are here to stay. Here are strong reasons why.

For the past several years, I’ve been dedicating my life to learn, design, build and write about chatbots. The main reason chatbots fascinate me and conversational interfaces in general, is because they offer the most natural way for humans to interact with machines. Not only is the interaction natural, but also simple, clean and focused on what you need instantly. Think of Google’s search interface. All you can do is input search queries into a little text box. Everything that comes afterwards is magic.

Chatbots are still in their very early stage due to a few factors:

1. Lack of expectations between what chatbots can do and what users expect (or in other words, bad user experience). This leads to an instant disappointment and therefore to low usability. It starts with the largest bots such as Siri and all the way down to the very basic ones, which confuse users about their actual capabilities. In theory, Siri claims she can do almost anything related to your iPhone device, calendar and other internal Apple apps.

In reality, over 60% of what you ask Siri is not understood or results in general web search results. Just so you can get an idea of what she can do, here’s a list of Siri commands. Did you know she can do most of it? I didn’t before.

2. Educating users on creating new habits. The last bot I’ve built, was focused purely on conversation, and therefore did not have any buttons or menus. Retention was very low, despite the fact that most of the conversations between the bot and users were held successfully. What I’ve discovered, was that the more buttons and menus I added, the more retention grew. This led me to the conclusion that the majority of users are still not used to talking with machines naturally, but rather prefer to click on buttons as we’ve been used to for the past 30 years. Secondly, clicking buttons are faster than typing sentences. However, buttons are not faster than voice, which is why voice will eventually dominate the bot space. The transition from buttons to a natural conversation is growing but is still in its early adaption stage.

3. Artificial intelligence might have improved, but is still in its early stages. The reason this last and not first, is because I truly believe we can build great chatbots with todays current AI solutions (such as Api.ai, Wit.ai, etc.) if bots were more focused on how to create user habits and offering a well designed user experience that meets user expectations. You can read more about how to do this in a previous post I’ve written on how to improve your chatbot with 3 simple steps. Obviously, AI will only improve with time as more and more data is collected and trained across a multitude of domains.

For all the reasons above and more, we’re still far from seeing the true potential of chatbots. However, there’s strong reasons why chatbots are here to stay and will improve exponentially over time:

The optimal Human to Machine interaction

If you think about it, we’ve learned over the past 30 years to adjust ourselves to the complexity and limitations of machines. This has been done via websites, applications, buttons, icons, etc. But in my opinion, the optimal scenario should be just the opposite - Machines should be able to adjust themselves to us humans, both from a natural language understanding perspective, and personalization.

Humans should be able to ask a machine anything naturally, instead of having to learn new interfaces, products, and habits for every service they need.

For example, let’s say you’d like to know the weather. Until recently, you’d have to find and pick a service of your choice out of many alternatives and learn how to use it functionally. Now, since every service tries to be innovative and different from its alternatives, this usually results in various UX/UI which means more learning and effort required by users. The optimal solution would be if you could just ask. Thankfully there’s many great solutions today for weather assistants (such as Siri or Poncho), and there’s many more to come in other domains.

Domain specific AI

Not very long ago, companies built virtual assistants which tried to go very wide and open, but quickly realized how hard it is to understand natural language. Going back to Siri’s example, Apple tried to capture many domains in order to display Siri as the ultimate personal assistant. This ambition failed very quickly. On the other hand, AI solutions that have succeeded, are the ones that focus narrowly on one specific domain.

Take for example Getdango - an AI solution for predicting emojis. They’re doing a great job predicting emojis based on natural language and it’s due to their narrow focus. Another example is Meekan, a scheduling assistant for teams on Slack. Meekan is a chatbot dedicated to providing the best solution for scheduling events as easily as possible.

The power of synergy where individual bots focus on specific domains, is the right approach for solving bigger AI challenges. You can see companies moving in this direction like FB Messenger’s latest release of handover protocol which enables two or more applications to collaborate. More importantly, Amazon partnered with Microsoft to collaborate on Alexa with the help of Cortana in order to provide a more powerful virtual assistant. If every bot was to focus on one specific domain, the race to AI in a whole, would be solved faster and more efficiently. Happily, that’s where we’re heading.

The power of long term relationships

The way most products are designed today, is to maximize instant short term value for users once they enter an application. While web and mobile applications focus on short term value, bots can and should focus on long term value. Bot developers should focus on how to build relationships with users over time so that value is gained and grows constantly with every single interaction. With time, bots should know enough about what they need, to maximize and personalize the user experience and minimize input friction. 

For example, say you’re looking for a travel planning service. The way you would go by it today, is to look at travel sites, fill the proper forms, and basically teach the site about your preferences and filters every single time. The way a bot should work, is to know which information is relevant to learn about the user, like personal information, preferences, budgets, places the user has already been to and etc. Bots should constantly learn from user’s behavior and offer much more personalized responses. 

The optimal way you’d be conversing with such a bot after some time, would be as follows:

The bot should know by now how many people are in your family, their ages, where you’ve already been to, where you’re at right now, what you like to do, what you don’t like and much more. In other words, the bot shouldn’t be any different from a real travel agent.

To see how much are users willing to share regarding their personal information within a conversation, I’ve conducted a research with one of my bots. The bot started asking users basic questions like “how old are you” all the way to more personal questions like “what are you most insecure about?”. Guess how many users answered all questions truthfully?… Over 85%. More specifically, 89% of women answered all questions truthfully, while “only” 81% of men answered all questions. So if you’re a bot developer, don’t be worried about users not sharing their information. Worry about what questions you should be asking in order to enhance the users long term value. This kind of information retrieval is something todays applications cannot achieve, and where chatbots have a huge advantage.

Cross platform usability

In just a few years, mobile apps have transformed to must-haves for smartphone users. But despite the increase in app usage and app choices, the number of apps used per user is staying the same, according to a recent report from Nielsen. Most people are tired of downloading mobile apps and learning about how to use new interfaces. 

In addition, research says that US citizens own in average 3.6 connected device. These devices can vary from mobile devices to Smart TVs and products like Amazon Alexa. That’s a lot of connected devices! Now obviously, users would like to interact with your service on any device they’re using. But what do you do? Build an application for iOS, Android, Smart TV, Alexa, Smart watch, iPad, Windows and Mac and more? Sounds like a lot of work. And it’s going to be very hard for you to get users to download your app in the first place, since they’re already flooded with other apps. 

This is where the beauty of messaging platforms comes in. At present, approximately 75% of all smartphone users use some sort of messaging apps such as WhatsApp, WeChat, Facebook Messenger, etc. Over 1.2 billion people worldwide have and use Messenger on their devices, all people that have mobile devices have SMS and obviously most have an email account. The list goes on. Instead of building applications and spending hundreds of thousands of dollars, just focus on building your bots back end. For the front end, just integrate your bot across multiple messaging platforms that are already on users devices and you’re set. If your service brings value, users will come. More importantly, turns out it’s what users want.

---

The future and success of chatbots depends not only on the big 4 tech companies, but on developers and entrepreneurs who continue to innovate and push boundaries in AI and conversational interfaces. Most of todays mistakes in the conversational user interface space, are tomorrows improvements and solutions. Eventually, bots will bring great value that cannot be achieved with most todays applications.

To learn more about chatbots go ahead and read the chatbots beginners guide. If you want to start building one, read this post on how to develop a Facebook Messenger bot.

Chatbots - The beginners guide

If you search for chatbots on Google, you'll probably come across hundreds of pages starting from what is a chatbot to how to build one. This is because we're in 2016, the year of the chatbots revolution.

I've been introduced to many people who are new to this space, and who are very interested and motivated in entering it, rather they're software developers, entrepreneurs, or just tech hobbyists. Entering this space for the first time, has become overwhelming in just a few months, particularly after Facebook announced the release of the messenger API at F8 developer conference. Due to this matter, I've decided to simplify the basic steps of entering this fascinating world.

What is a chatbot?

To fully understand what is a chatbot and its potential, lets start by watching the following example:

Get the idea? The conversation example above, was conducted between an end user and a chatbot, built on the Facebook messenger platform.

So what is a chatbot? It is a piece of software that is designed to automate a specific task. More specifically, a chatbot is essentially a conversational user interface which can be plugged into a number of data sources via APIs so it can deliver information or services on demand, such as weather forecasts or breaking news. 

Why now?

Chatbots have been around for decades right? So what is all this noise all of a sudden? This question has many different answers, depending on who you ask. If you ask me, there are two main reasons:

1. Messaging has become the most essential and most popular tool for communication. 

2. We're closer to AI (Artificial intelligence) and NLP (Natural Language Processing) breakthroughs than ever before. This means that talking to a chatbot can closely become as real as talking to a human. Today, developers can find many APIs that offer AI/NLP services, without even understanding how AI/NLP works - This is HUGE. A few examples I recommend are Crunchable.io, Chatbots.io, Luis.ai (a must!), API.ai and Wit.ai.

Basically, the point I'm trying to make is, that messaging platforms are the place we all go to on a regular basis. So why not bring all the other places into this platforms? This is what Facebook did with Facebook Messenger.

Facebook Messenger is far more than a messenger app. It is a store for thousands of apps which are integrated into our daily conversations. Furthermore, as stated above, Facebook has released its chatbot platform in April, 2016. Since then, more than 11,000 bots have been added to Messenger by developers.

Where are the chatbots?

The first chatbot I built was on WhatsApp. The reason I chose WhatsApp, is because all my friends use it as their main messaging platform. Unfortunately, WhatsApp doesn't offer an official API. What this means is, that WhatsApp doesn't approve building chatbots on its platform (not a surprise since WhatsApp is a Facebook company, which itself offers an extensive API). This doesn't mean that there aren't any work arounds. If you're as stubborn as I am, take a look at yowsup and start from there. You'll also need a registered phone number before starting the process. So to conclude, WhatsApp is probably not the place you'll find rich and powerful chatbots. 

Platforms that do offer official APIs are:

1. Facebook Messenger

2. Slack

3. Telegram

4. Kik

There are other deployment channels such as Android and iOS (via SMS), Skype and even Email. However, the listed above are the ones I would focus on.

You can find a rich list of most of the chatbots out there by clicking here, thanks to our friends at Botlist.co that did an amazing job. 

How do I build a chatbot?

This requires a long answer. An answer I will save for my next blog post, in which I will describe how to build your very first chatbot using Node.js and MongoDB.

If you're not a developer, or is looking for an easier approach which does not require programming, here are a few solutions for you:

1. Chatfuel - My first choice. No coding required. Easily add and edit content— what you see is what you get.

2. Botsify - Build a facebook messenger Chatbot without any coding knowledge. 

3. Meya.ai - Meya helps with the mechanics of bot building so you can concentrate on the fun stuff.

There is some downsides to using a service instead of building your own. Using the above services limit your creativity in many ways, enabling you only a glimpse of what can be done. Secondly, you are using a third party hosting service, which means you're stuck with them. Nevertheless, these are great solutions for services that will get you started with chatbots, without the need for any coding knowledge.

Summary

There has been a lot of controversy rather bots will succeed or fail in the near future. To understand the controversy, you have to understand the differentiation between "stupid" bots and "smart" bots. "Stupid" bots work with structured input, while "smart" bots process your natural language and provide a more human-to-human experience.

The main issue with "stupid" bots is that as soon as people start bundling things up, changing their minds, going back to what has been mentioned earlier in the chat, etc., the bot falls apart. Therefore, as long as chatbots can't fully conduct a conversation naturally, while understanding the intent of the user at every stage, bots will be limited and ineffective. 

Having said that, in my opinion, chatbots don't have to be smart in order to succeed. There are thousands of use cases in which a "stupid" chatbot can simplify both the end users experience, and the business side productiveness. Take for example ordering Pizza. You can create a flow in which the user needs to enter inputs based on questions and options. You can deliberately state the input you're expecting from the user, and therefore the need for NLP or AI becomes irrelevant. I would prefer ordering pizza from a "stupid" bot then over the phone, or some cheap website any day. 

To fully summarize the above and much more, have a look at the Chatbot ecosystem, brought together by Business Insider.

Stay tuned for my next blog post, about how to develop your very first Facebook Messenger chatbot, using Node.js and MongoDB.