Why it’s difficult to draw the parallel between Facebook Messenger in America and WeChat in China
Over the past few months, I’ve spent quite a bit of time and effort exploring and experimenting on the Facebook Messenger platform, developing bots and interacting with the platform and other bots. After much exploration and discussion, I’ve come to the conclusion that the Messenger platform, as well as other similar platforms like the Slack one, aren’t going to go anywhere fast. Not only did 2016 fail to be the “year of chatbots,” I am now beginning to think that even 2017 will not be turning out as the “year of chatbots.” In fact, at this point I’m not sure if that year will ever come in America and the western world.
The one piece of evidence that everyone points to, to suggest that it should work — WeChat
WeChat in China is really the one thing that everyone in the community likes to point to, when arguing how messaging apps and chatbots would take over.
Some of the most referenced articles include Dan Grover’s write-ups on his experience in WeChat — this piece in 2014 and this piece in 2016. (I’m a big fan of these pieces, they are extremely informative and bring great insights.) Dan brings his product experience from WeChat to Facebook, and is now a Product Manager on Facebook’s Messenger platform. Matt Schlict also often writes articles making a lot of arguments for bots on Chat Bots Magazine.
Here is two reasons why it’s actually not correct to draw a parallel between WeChat’s reign in China and Messenger in America or the western world in general.
Messenger lacks ubiquity. It might be ridiculous to claim this, seeing as Messenger often boasts having a billion active users. Yes, by the number, this is impressive. However, think about your daily life and the reality. If you live in America, think of a scenario where you meet someone new, say making a new friend at a party or in school. If you were going to exchange contact information to get drinks next week, what would you do? You exchange phone numbers, and send texts (SMS) to each other.
This is the difference.
If you’ve ever lived in China in the past few years, you would see that nobody does this. If you’re exchanging contact information, you contact each other on WeChat. The assumption is that everyone has WeChat, and you’re weird if you don’t. Next time when you exchange contact with someone in America, try and say “Here’s my Messenger link/scan code”, see what kind of response you get. You may get anything in the range of blank stares, a double take, or just a straightforward “I don’t really use that.”
(Interestingly in the rest of the world, like in Hong Kong, or much of Europe, the corresponding app of choice would be Whatsapp, which is also owned by Facebook)
Much of this is thanks to the rest of the world not having affordable/unlimited SMS packages on their phones back when data began getting popular. I’m not here to argue what should or should not have happened, but the reality is SMS reigns king in America, unlike in China.
If you live in Silicon Valley, you may feel like some of the other messaging apps are ubiquitous in your circles, like Telegram or Kik. This is most certainly not the case outside of the Silicon Valley bubble.
When you don’t have that kind of ubiquity, it’s difficult to build a platform on top of it that could change everyone’s behaviors and use it.
This is reason number two. Commerce is one of the best ways to bootstrap the platform and get users to get used to being on it and using it as a way to get to everything. WeChat is extremely successful in this aspect; being able to buy stuff on it and even pay for the bill at restaurants using it. Facebook understands this. They want to be in that position in America.
But WeChat got its start at it because of a cultural tradition, red envelopes. This extremely informative piece written by Connie Chan of A16Z is a must-read for anyone in the industry. In short, because of how importantly Chinese people view the concept of red envelopes during Chinese New Year and other occasions, WeChat was able to capitalize on it to jump-start the commerce ecosystem on its platform.
As the messaging and app platform that everyone already used, it was a good bet. Getting users to link their payment information to your platform is no easy task. It takes a tremendous amount of motivation on the users side for them to come to accept this and do it. The red envelope tradition was the thing that pushed people over the barrier.
There is simply no equivalence, or even a similar concept, to this, in the western world. If there was, you bet Facebook would’ve utilized it already at this point. Facebook has built user-to-user payments features into Facebook itself and Messenger years ago. Traction is mediocre at best; most casual users would not even know that it exists.
Facebook has eyes on the long tail, but would not build for the viral growth
I previously wrote that I believe Facebook is close-guarding the platform to not make the same “mistake” as 2007; where a whole bunch of fart apps, and eventually games, got popular and took over the Facebook canvas apps platform. I still believe this is the case. This time around, they’re trying to not let the gold rush take place, and slowly build out the ideal user experience on the platform.
However, this also makes it hard for the platform itself to grow.
In a Hacker News thread that developers had with Dan, Dan wrote two informative posts: one with a long list of discovery channels that Facebook offers for users to get in touch with Messenger bots, and one with a fairly exhaustive list of use cases that he envisions where Messenger bots could shine in.
I agree 100% with his long list of example use cases, they are all scenarios where a Messenger bot makes complete sense in and I look forward to the day where they are all commonplace and we get to use Messenger bots in those ways.
However, they are all long tail use cases. Dan himself mentioned as much:
But these sort of problems felt by the long tail are things bots are particularly great at!
He’s completely right. They are long tail use cases, and they are particularly great for bots to solve.
But that’s also the problem. When the platform is not at the maturity level and ubiquity level (of say, WeChat in China), you can’t tackle the long tail use cases right away.
For all of these use cases to work, for the small businesses to even implement them, and for users to even know that it’s an option, you need the platform to be ubiquitous like WeChat in China. Like I mentioned in the previous paragraphs, Messenger is not there yet. (I’m hoping it will be in the future; I root for Facebook and I also look forward to having one such platform in America myself)
Here are some of my ideas of what I think would work in helping the platform grow and get there.
I mentioned this idea in my last article and admitted that it’s a poorly designed idea, but I think something similar could work.
Facebook itself enjoyed massive engagement when they first invented and launched the newsfeed. The Facebook canvas apps platform was able to publish content on it in the early days, which played a huge part in the explosive growth of the platform — since that’s where all the eyes are.
I think you can create something similar on the Messenger app. It doesn’t have to be as in-your-face as the Facebook newsfeed itself, but just a more expansive version of the featured Bots section currently on it. It could show the bots that your friends have been interacting with recently, and maybe even some published stories (similar to how Facebook canvas apps published them on the FB newsfeed before).
Yes, like Dan said, these don’t make a lot of sense when you think about the long tail use cases where a bot typically extends a physical or digital service or entity (company/business). But it makes a ton of sense for apps that are purely built on the Messenger bot platform. In a way, the goal of something like this is to help the platform itself grow, not to tackle the long tail use cases, for now.
Allowing bot mentions or functionality in user chats or group chats
When the Messenger bot platform was first announced, before any of us knew about much it, one of the most anticipated features was to be able to make calls to a bot in the middle of a chat with a friend, much like how they envisioned Uber to work.
In the Uber example, you would be talking in a group chat with 3~4 friends, and when you need a ride somewhere, you’d be able to summon an Uber directly from within that conversation.
Eventually, Uber itself did get made available to be used in such a way. Unfortunately, not so much for the rest of the bot developers.
Another example is the basketball mini game and the @fbchess feature. They work directly in chats between friends or group chats.
If this functionality could be used by 3rd party developers, it should help with bot discovery immensely and help the platform grow at a much better rate. A conversation with a friend is where your attention is as a user, and if bot functionality is shared, that’s one of the best ways a user could learn about the existence and usefulness of Messenger bots.
Spammy behaviors would indeed be a concern, but remember that at the end of the day, it’s up to the user to decide what to activate or trigger within a chat with a friend. If a bot behaves badly, it would naturally discourage the use of itself by users.
Those are some of my thoughts at the state of things as of currently, in January 2017. This write-up might sound really critical, but I’m actually actively rooting for Facebook to be able to get this right and build such a platform successfully for us. As both a user and a developer, I think we would benefit tremendously from this.