Ha Duong
There has been quite some buzz around Facebook’s Messenger platform lately. And that’s not all. Almost daily, you can read new articles about chatbots, conversational interfaces, digital assistants, AI and messaging platforms. Is chat the next generation of how we connect and interact with the internet? Can Facebook’s Messenger eventually become the platform that connects all services that consumers need?
Since the launch of the Messenger platform with chatbots in April, Facebook has allowed businesses to deliver automated interactive experiences, e-commerce guidance and customer support with the help of chatbots. Thanks to the new Send/Receive API, these chatbots can send not only text but also structured messages, including images, hyperlinks and CTA buttons, within their customer interactions.
Users can browse product carousels to discover options and get directed to the web to finish their purchase. Recently, Facebook additionally announced that chatbots can now send promotions and subscription-based messages that make retargeting, up-selling and cross-selling possible.
Businesses have shown great interest and want to harness the capabilities of Messenger in order to enhance their customer relations, build up their brands and gain more profits.
But despite all the news and trending discussions right now, have chatbots delivered what was promised? Will they really have the potential to substitute native apps in the long run?
Chatbots — hype or reality?
The current interest in chatbots is obvious. They can be an entirely new way to engage your customers by providing personalized, interactive communication similar to talking to a human while being able to scale for the masses. They have the potential to increase user engagement and satisfaction for a brand. Eventually, they can perform more complex tasks than requesting a taxi or booking flights.
There is much talk of disruption because they could change the entire advertising industry through integrated marketing. And the best part is they provide access to digital services without the need of installing an app. But where are they now?
More than 18,000 bots have been built for Facebook Messenger. While some engage and delight consumers, most of them have not yet proven their value to the masses. In fact, beyond the tech scene, most people don’t even know what a chatbot is and what it is good for. This shows that a core underlying principle should not be forgotten when creating the next semi-intelligent chatbot: No one actually wants to talk to a bot.
Users want solutions to their problems. They want answers to their questions. They want their needs to be fulfilled the most efficient and natural way. What they don’t want is to type in a message on their small smartphone keyboards just to see the bot doesn’t understand it again and again.
For chatbots to work, developers must stop thinking about them as a bot. It is a new channel that can be used to engage with an audience via automated processes. No one expects an automated email campaign to solve all problems, but it is good in its own ways: distributing content and educating users.
This is also how chatbots should be used: not as a one-size-fits-all solution to all business services. It is only valuable when it can be the most efficient channel for customers to actually engage via a question and answer kind of way. It can be great if customers want to know something specific and a conversational interaction method is better than searching, browsing or button-clicking.
But chatbots currently just don’t feel right when I want to browse through all available options or when I know exactly what to do and a button is 5x faster than the text input. Not every service should be offered as a native app, and the same is true for chatbots.
In time though, once the technology is more mature and machine learning has been adopted by the masses, I can see the potential that a conversational interaction method may indeed be best to solve problems. It will eventually be possible to talk to your phone like talking to a friend, and the virtual digital assistant will automatically combine different chatbots to compile your optimal solution. This leads to the next step in human-computer interaction: the human-feeling user interface.
The bigger picture: Creating a human user experience
Chatbots are the interim solution before a more intelligent solution appears: In the future, a more or less General AI will be able to handle most of our questions, problems or inquiries. On the contrary, a chatbot (even an intelligent one) can be considered as narrow AI, with a limited array of possible questions it can handle.
Chatbots will soon be able to integrate information from our social media profiles, browsing and shopping history, geolocation, available smartphone sensors and even our behavioral biometrics. But still, they will be made to solve a specific problem; and once an inquiry goes beyond the spectrum of situations for which it was trained, the interaction will feel weird. The end goal will always be to create a human-computer interaction method that feels human.
“Human” doesn’t mean to have unnecessary questions asked or to have imperfection incorporated in the AI; it means that we can get solutions to our problems without knowing the specific syntax. Natural language processing (NLP) will enable us to talk with the system that is designed for voice; and through other fields of AI, like reasoning and knowledge representation, the response of the system will feel intelligent and human.
Once machines get smarter, traditional user interfaces can be simplified, abstracted or even hidden. The perfect user interface is no visible user interface, because then users can just interact with the system without needing to know the operating rules and patterns. And it should not only be the interface. The entire user experience should feel human. Interacting with humans is different because humans have contextual awareness. Humans empathize. Humans can think about you and your problems before you explicitly express them. Maybe one day, a general AI will be able to “feel” and “think ahead” to solve your problems before they arise.
Are we in any way close to having a general assistant that can do what we ask? All the big tech companies have created some sort of digital assistant. Facebook announced M, its own digital assistant similar to Siri, Cortana or Google Now a year ago. Contrary to other assistants, it is powered by both AI and people in order to make its interactions more human. Mark Zuckerberg said that through NLP and AI, in combination with human help, users will talk to Messenger bots just like talking to their friends.
M is part of a bigger game to make people more used to getting things done via Messenger (shopping, booking, scheduling, ordering etc.). Also, M is interesting for Facebook to learn more about consumers’ needs (what type of services they would like to have in Messenger). But we haven’t heard that much from M. Facebook’s Messenger chief David Marcus confirmed that M won’t be broadly ready for years. But still, hope remains that one day it can be the missing link that combines all separate chatbots into one intelligent service system.
Facebook’s Messenger is well positioned to become the holistic platform that hosts all different kinds of services in the future. As mentioned, though, they are not the only ones. Isn’t there already another messaging platform that is a couple of steps ahead?
Where Facebook’s Messenger lags behind other messaging platforms
If you look beyond the western tech scene, there is clearly a messaging platform that has already figured out how to transform itself from a chat app into a more powerful platform, portal and operating system: WeChat.
WeChat in China is what all messaging platforms aspire to be: It accompanies the users throughout their daily life in every possible aspect — making it indispensable in people’s minds. Through WeChat, Chinese can do everything from chatting with friends, groups or strangers to conducting business or e-commerce, accessing fintech applications, playing games, or even connecting to diverse IoT devices.
WeChat does it in a way similar to Messenger — by integrating third-party applications in its ecosystem. And with an extremely high average revenue per user of more US$7 per year, it has found ways to monetize its massive user base that other messaging platforms can only dream of. WeChat has already accomplished what Messenger is just beginning to build up.
When asking people about their experience with WeChat, you’ll only hear enthusiasm. By offering new possibilities to meet and interact with people, WeChat has fulfilled the needs of a society where loneliness is a common status (a whole generation grew up with everyone being an only child). WeChat has succeeded extremely well in tailoring the app experience toward the target group while offering services on top that make it essential to everyone.
More than that, WeChat has continuously been able to surprise and delight its users with new innovative features way ahead of other messaging platforms. For example, voice messages took over in WeChat years ahead WhatsApp or Messenger. Another example of how WeChat captures its users’ engagement is a stunt at Chinese New Year where users were able to send each other random amounts of lucky money — an important part of Chinese culture. Facebook’s Messenger is a strong general communication platform, but, so far, it fails to surprise and delight its individual user segments.
A strong differentiator is that WeChat actually is not only a messaging platform. It has elements of a social network where people can update their status or follow official accounts of celebrities or brands. Brands in China often use WeChat as their main channel for PR, content marketing, customer acquisition and brand marketing. Facebook clearly aims for this goal, as well, but it doesn’t enjoy the dominance in possible touch points with users as WeChat does.
The ultimate game changer is WeChat’s integration of payments. They make it not only easy, but fun, to send money to friends. It also is convenient to pay merchants offline and online, and people are additionally incentivized with discounts to use WeChat Pay. This makes it possible for Chinese people in metropolitan areas to completely forgo cash. WeChat Pay is what fosters its e-commerce functionalities and attracts millions of online stores to the platform.
Facebook Messenger has the function of P2P payments, but payments are not integrated into the bot functionality (so far). And unlike within WeChat, it is quite difficult for small businesses to set up an e-commerce bot in Messenger, even with options like wit.ai (acquired by Facebook).
All of these are just a few examples of what makes WeChat great. The point is, why is Messenger so far behind? Why does even its UI/UX have inferiorities to WeChat? Why doesn’t Messenger succeed in astonishing people around the world with innovative features? (Note: In Europe, we often don’t even see many cool features like P2P payments for years.)
Copying WeChat is obviously not the right way to go because WeChat succeeded by tailoring its app experience to the needs of the Chinese market. But there are definitely inspirations from the East that should have major impacts on the roadmap of Messenger. And inspirations from other messaging apps don’t stop there.
There are many more messaging apps trying to build up a platform play. Slack was among the first that not only revolutionized messaging in businesses but also engaged a massive developer ecosystem that is creating useful third-party applications within Slack. It is super fun to go and discover new bots on Slack’s bot store (App Directory), which can add value to team communication. Why doesn’t Facebook have an intuitive and searchable bot store? And why does Facebook still completely ignore the space of communication in professional/business contexts?
It is still extremely difficult to communicate with people on Facebook before adding them as friends. Friendship seems like a quite personal concept if I only want to speak with someone a couple of times or simply stay in touch without sharing private statuses. It looks like Facebook even wants people to go to Twitter or LinkedIn to connect with professional contacts instead.
Another competitor that Facebook can learn from is Snapchat. Contrary to Messenger, Snapchat is what younger people consider cool and more engaging through completely new possibilities to interact with peers or even brands. Snapchat realized that needs are changing and people want to share moments in their life without leaving a trace in the World Wide Web. Also, people were enabled to express their creativity through the application of fun photo and video filters. Because some filters change on a weekly basis, users are incentivized to revisit the app again and again to share even more fun and embarrassing pictures with their friends.
Since people started sharing less and less content on Facebook, the platform has never really figured out a new way to invert that negative trend. Shouldn’t Messenger, as a personal communication platform, be able to motivate people to share pictures and videos again?
It is clear that Messenger still needs to work hard to accomplish its goals and exceed its high expectations. Messenger has a strong position with its vast user base across the world. But it needs to keep innovating in order to continuously engage and delight this user base in new ways. For that, it needs to stay user-centric and always be on the user’s side instead of focusing on how to make more money with businesses. Here are some options how Messenger can improve itself through inspirations from competitors, while staying true to itself and shaping its own path.
Opportunities to shape Messenger’s future
In my opinion, Messenger needs to evolve much further and in more fields to become dominant in the future. On a strategic level, it could invest in these three pillars to reinvent itself:
- Messenger should extend its use case beyond the private communication sector in order to capture business relationships. If people get more used to using Messenger in a professional context, interacting with brands, carrying out e-commerce transactions and directing digital assistant applications won’t be that unusual anymore.
- Messenger should focus on improved user engagement by developing a creativity ecosystem. It should let its users interact with each other in new ways. The users should be able to define how the interaction can look. It should let the users co-create their own messaging experience by letting them be creative and do things they couldn’t do before.
- Messenger should target the masses beyond the tech scene with valuable services offered through the conversational interface. Bots will never take off if their users will only be techies in the Valley because it’s a wide user base that attracts brands to the platform. It should make bots discoverable and more present in the online and offline world.
These strategic aspects reflect what other competitors have been doing well. WeChat operates in personal and business contexts. Snapchat engages users by new offerings again and again. Slack makes it easy to find and integrate new apps/bots into its platform. Kik lets friends share bots in group chats or in private messages, which creates some virality.
If Messenger succeeds in catching up in these fields it can really become the underlying platform that connects the new wave of applications — pushing away Apple and Google from its iOS and Android thrones. So what can Messenger improve in the near future to come a step closer to that vision?
Small incremental Improvements: These are improvements that make the messaging experience itself less inferior to other messaging apps. Messenger can start by allowing multiple-people video calls, enable voice messages to go beyond 1 minute and work on small UI/UX elements that make the app better. It can make interacting with bots faster by getting rid of the “Get Started” Button at the beginning of the conversation. And it can make group interactions with bots possible. Bots have two discovery issues: The users need to first find the bot and then need to figure out what the bot can be asked. Having an intuitive and searchable bot store within Messenger would be the first step to making people discover and find out more about the bot.
Bigger substantial improvements: These are improvements that require more investment and may make the brand go out of its comfort zone a little. Messenger could include some sort of daily or weekly change in features that continuously surprises the users in a positive way. First attempts were already made, e.g. with birthday balloons, but we need more of these! And we need features that are not only nice to look at but that let us interact and engage with others. The same way that people love to discover and play with new Google Doodles or Snapchat filters, Messenger could build up something similar that is engaging in new ways. Facebook is already on the right track. Through the recent launch of Lifestage, the company shows its wish to engage younger audiences with new ways to express oneself. In the near future, Facebook could integrate FB Live within Messenger or go a step further with live VR content within Messenger to allow new interaction methods for users.
Bold and totally new functionalities: These are new developments that have probably not been often thought of in the context of Messenger. Facebook can work out a new B2B product that integrates Messenger and its bots directly on websites for better customer support through live chat. It can even go beyond its own bot ecosystem and build up a development framework that allows easy cross-platform development of bots, similar to React Native for native apps. This way, developers would be incentivized to build new bots for Messenger that can also be launched on other messaging platforms. And maybe Facebook can even figure out a new way to integrate Instant Apps for Android or similar functionalities within Messenger that make possible executing small app functions within a minimized web view.
All of these aspects show there is still plenty to do for Facebook’s ecosystem. The current disillusionment of dumb chatbots is only temporary; in time, some types of narrow or more general AI assistants can be brought to the masses. Messenger has great potential to achieve this. I admire what Facebook, as a company with strong culture, has already done, and I believe that it has what it takes to reinvent communication. I’m excited to see how we can make the world even more open and connected all together.
This article was meant to show where Facebook and Messenger stand right now and which opportunities lay ahead. Let’s hope that whatever comes next will enable entrepreneurs around the world to create innovative services and solve big problems. I always welcome discussions, so shoot me a message with your thoughts on this article!
Special Thanks to my friends Michael Ortolano (Program Manager at new AI-focused and Asian-based accelerator zeroth.ai), Ellen Hoeven (Design Thinker), Nick Helleberg (Head of Operations at Dormando), Ivy Dai (Freelance Consultant for Chinese Business) and Amy Ly (Business Designer at IDEO) for their input about WeChat’s messaging platform.
Comment