It’s the whole #### “we need an app” all over again. Product managers and executives are tumbling to get budget for a chatbot development project. However, let's take the lessons we’ve learned from the app rush.
The app stores are filled with apps with weak use cases, unsupported iOS and Android versions and screen sizes. We have the power to do it different this time. Chatbots, on the other hand, make use of your app’s User Interface (UI) letting you create a chatbot without worrying about UI design. It’s the first easy win for chatbot creation. No need to update code and design when a new Operating System (OS) or screen size is announced. Channels like Facebook Messenger, Telegram, Slack, Skype, etc will take care of that for you.
The second win is the ability to have multiple channels and OS support with one bot code. It’s ideal scenario: “write once, use many” source code, and we are almost there. The bot code is only developed once and only a few specifics details per channel are in a separate file. And as soon as a new channel opens (we are looking at you here, WhatsApp) you only have to add those channel specifics in that file and deploy it. No need to run multiple instances. No need to redo the work from scratch for that new channel.
If you are an enterprise with an app to support, you might have different developers with knowledge of the primary coding languages; Swift / Objective C for iOS and JAVA for Android. With chatbots you can use JavaScript (or ES6). This is a language known by a larger group of developers. There is no need to learn new languages as you simply use existing talent from your web development team.
What is still the same with chatbots though, is the use case (intent). But luckily we have come far in User Experience knowledge. We understand getting a constant reply from a chatbot as “I don’t understand” is a bad User Experience. The intent of a chatbot must be clear. What is its purpose and what should do for people using it? Why should users communicate with the bot. For instance a chatbot for banking customer service could not give you the current weather in Seoul. But it should know all the Frequently Asked Questions of the bank’s call center, no matter how it is asked. The expectations of a user can be set in the early stage, while on-boarding the user and the standard “I don’t understand” could be replaced with a reminder of the intent. In case of a customer service bot, for instance a “press 1 to call me back” option.
Overtime with user feedback, the chatbot manager is able to improve the bot. Understanding the way people interact with the chatbot will give insights on the type of users it attracts. On the other hand, the level of entry for developing a chatbot is lower than developing a simple mobile app. One key advantage of creating a bot is the training and interactive response of the bot can be done by a non-technical person. This lessons the burden in creating a bot and allows for a shorter creation time bring the bot to market faster.
Developers have the power and the knowledge to build great chatbots that serves a specific purpose. Give the audience what they expect. Build great bots. Don’t know how to get started? Check out our bot development platform Recime which we’ve built to help developers worldwide bring bots to life! We give you the developer tools and knowledge you need to build amazing scalable multi-channel bots. With Recime, you build it once and deploy it across Facebook Messenger, Slack, Telegram etc. Don’t know how to build a bot? Let us help you. Contact us for your bot development needs. With the Recime bot development platform you also have all the tools to do so.
