RCS Messaging Automation With AI Chatbot Software

Post by
Indrek Vainu
    Published: 
March 11, 2021
RCS Messaging Automation With AI Chatbot Software

1. Introduction

RCS is a new and improved version of SMS. Short for Rich Communication Services and called differently around the globe (joyn, Universal Profile, SMS+, Message+) it is a protocol agreed by mobile operators and private companies to enable sending rich media (pictures, vidoes, QR codes, carousels, read receipts) interoperably between phones. Currently working in Android phones only, companies can benefit from using automation and AI chatbots to communicate to billions of Android users at scale. Combining RCS messaging with RCS automation software such as  Conversational AI helps companies automate their customer support, provide product updates and personalized push notifications.


2. What Is RCS?

RCS Messaging is a new and better form of SMS, the text message format that we use to send messages from one phone to another. RCS is a new communication protocol set by the GSM Association, a global industry organization that unites more than 750 mobile operators across the world. 

RCS is not an app owned by a single company (like Messenger is owned by Facebook) but rather a common protocol that allows users to send rich media messages from one phone to another without needing to use a specific app.

RCS is marketed under different names across the globe - Advanced Messaging, Chat, joyn, SMSoIP, Message+, SMS+, Universal Profile. But the core of it remains the same, provide for more interactive SMS functionality with rich content.

RCS was dormant for nearly 10 years when Google picked up the leadership role in pushing for a unified standard. For a while it was required that operators turn on RCS capabilities in their networks for users to use it. However, Google bypassed that and RCS is now available for any Google Messages app user globally.



3. What Is The Benefit Of RCS?

Traditional SMS (know as Short Message Service) is a communication protocol that was defined in 1985. The maximum SMS length is 160 characters. As it is an old protocol, it is clear that user expectations for messaging have risen considerably. The reason apps such as WhatsApp, Facebook Messenger, Telegram and numerous others are used is because people want to share pictures, videos, emojis, GIFs and other forms of rich media that SMS does not support that well.

RCS is a way to make SMS better. What it allows is to send videos, images and carousels (like the product carousels you see in Facebook Messenger), typing bubbles and read receipts. In addition, RCS messages can be branded by custom colors. This gives the text message the look of a messaging app but with the simplicity of the SMS that is on every phone. 


RCS use cases
Receive rich messages like maps and tickets. Source.


4. What Are The Drawbacks Of RCS?

For a while, the main drawback of RCS was that it was not end-to-end encrypted. Meaning that messages sent via RCS were more vulnerable to hacking than messages sent via WhatsApp, iMessages and Signal which are encrypted. 


RCS encryption


However, in the end of 2020 Google rolled out end-to-end encryption for its Messages app. At the time of writing it is in beta which means that users have to turn on encryption on their own (it is not on by default). And when encryption is on, it is on between phones on which encryption has been turned on and only in 1-1 communications. It is not working in group chats.


5. Why RCS?

There are various reasons why RCS was made but two can be brought out. 

First, it was because SMS did not support rich media and users wanted more interactive experiences and features with messaging. Secondly, as users flocked to WhatsApp and Facebook Messenger, mobile operators started losing on SMS revenue. So RCS is a way for carriers to keep users on using the newly minted version of SMS with the benefits of rich communications that users expect.


RCS read receipts
Rich messages and read receipts. Source.


And thirdly, RCS is a way for companies to have direct communications with users over messaging. Many companies use SMS notifications about upcoming doctor’s visits or delivery of parcels as an example. With RCS is it possible to go beyond that by providing automated support, richers content (carousels, images, videos) inside the messages and through that enhance the experience of the users with businesses (called RBM).


6. What is RBM?

RBM stands short for RCS Business Messaging. It is a way for businesses to connect to users through messaging and provide them interactive experiences. With RBM businesses can send you receipts, tickets that you have purchased, connect you with live chat and do much more to make customers' lives easier. 

According to statistics RBM can help businesses move the needle on customer engagement. 72% of people are more likely to purchase online if they can ask questions in real-time and 35x are more likely to read RCS than emails. 

Users still need to opt-in to receive RCS messages from companies but this is like with any marketing channel. The difference though is that engagement rates with RCS is higher than with email or other traditional marketing channels.


7. Can I Use RCS On Every Phone?

No. RCS only works on Android phones. Apple has its own iMessage that iPhone users use for sending SMS to non iMessages phones and more rich iMessages to iPhone users. Apple has not indicated that it will support RCS on iPhones. So for now, users with Android phones get to use RCS. 


8. What Is The Difference Between RCS And iMessages?

For one, RCS only works on Android phones as Apple has not indicated support for RCS. iMessage works on iPhones and also on other Apple devices.

Secondly, iMessage is server based. Meaning that messages go through a server and as such you can access iMessages through your app from several devices (your Macbook, iPad etc). RCS is tied to your phone, that one device as RCS is not server based. There is no central storage of messages as they are device based. Messages go from your device to another device. 

And third, encryption. iMessage is end-to-end encrypted by default. RCS in Google Messages is encrypted and has to be turned on by the user and is only on in one-to-one communication (if the other party has also turned it on) and not in group chats. 


9. How Can I Use RCS?

First, you need to have an Android phone. Second, you need to use Google’s Messages app. Third, in your Messages app you need to go to Messages Settings, click Advanced Settings and enable RCS. Then you can start using the RCS features. You can use RCS in Messages if you run the latest Android version. You can send RCS messages to other parties that also have RCS. If other parties do not have RCS, a regular SMS is sent instead. 

10. Business Cases In RCS

Billions of people globally have Android phones. This makes it an attractive channel for businesses to communicate with their users. Having rich messages through RCS available in the Google Messages app is an attractive value proposition for companies. Instead of sending only text based information through SMS companies can now provide for a more enriching experience for users. Here is a list of some of the most common use cases that companies can consider.

10.1. Make Customer Service Better

This area is perfect for automating through RCS because there are large numbers of users that are looking for answers to similar types of problems. Users are looking for companies to take ownership of their products and services they offer through customer service automation. This repeatability of similar queries makes customer service one of the perfect fits for RCS based AI chatbots and Virtual Customer Assistants.

10.2. Make Online Shopping Experience More Delightful

Ready to buy those shoes online but still have a few questions? Simply start a chat on your Google Messages app with the store and get relevant info about availability and material. As people shop more on mobile it is important to connect users to businesses on mobile and RCS provides a way to do that. 

10.3 Push Notifications

Businesses can send promotional messages to users that have opted-in for receiving promotional messages. So when there is a new sale of products in your store or arrival of new products, you can let your audience know directly. 


11. Why Do You Need Automation With RCS?

The common denominator in all of those business cases is that businesses need to be able to carry out one-on-one communication at scale. In messaging users will inevitably have questions and simple button based options will not do. So businesses need to have a Conversational AI based Intelligent Virtual Assistants (some also call them smart AI chatbots) at the forefront of their RCS automation campaigns to carry out conversations with users.

Users expect answers in 10 seconds or less. That is because the on-demand economy has raised users’ expectations to what is good customer service. And the great service they receive from one company builds expectations for another company. So you as a business have to provide service on the level of other businesses irrespective if they are in your business area or not. Providing fast response at scale to millions of users is almost impossible with human support because of the high cost of having that much staff on payroll. 

Hence, Intelligent Virtual Assistants come to the rescue. An improvement upon a simpel click-based and if-else-logic-based chatbot is what's called a Virtual Customer Assistant (VCA). These intelligent agents are not only capable of presenting a multiple choice selection of answers to the user but also understand user intent from free text. Understanding natural language and being able to interact in multiple languages is a significant leap.


Many times users are looking to articulate their specific concern to the machine in a similar manner they would do to a human. User has a question and asks that specific question from the machine e.g. “When will I receive my payment from Bank ABC?”. The main drive behind this is that users are looking for a quickest way to get an answer to their specific question.

In our example, the user is looking to understand in hours or minutes how long it takes for the payment to arrive from Bank ABC. Not Bank XYZ or GFD but Bank ABC. And the VCA has to provide this answer in the form of “It takes 6 hours during weekdays to receive a payment from Bank ABC”. In such a situation only the most relevant answer matters and for the users it does not matter if the answer comes from a machine or a human. As long as it is accurate and  relevant. 

One can see how a traditional chatbot falls short in this example. The chatbot might show an illustration of transfer times from other banks or give a link to a self-help article. However, the user needs a concrete answer. Hence, Virtual Customer Assistants with their ability to understand specific intent from free text, are helpful here. For more information on vendors, click here for our top 17 Conversational AI software platforms and here we put together a comprehensive list of the top 30 chatbot software providers.


12. How To Build A Chatbot With RCS Automation Software

1. Go to https://alphachat.ai and click Sign-Up. Enter your email and choose a password. Go back to your inbox and click on the verification link. Now you have verified your account.

 

 

Alphachat.ai no code maching learning platform website.

We also have an AI-based Virtual Assistant embedded right on the front page that you can chat with. The best part? You can chat with it in any language. English? For sure. Spanish? Of course. Swahili? We got it covered. 

 

Intelligent Virtual Assistant built with Alphachat.ai
Intelligent Virtual Assistant built with Alphachat.ai

 

2. You can use the product with all its features for free for 10 days. During the first 10 days you have access to all the features and can build your own Intelligent Virtual Assistant.

Signing up does not require a credit card. So if you just want to build something quickly and try out for free, you can do that. If you do not wish to pay for the product after the 10 day free trial, no problem. You don’t have to do anything. After 10 days your account will close and unless you enter your credit card and pay the €399 monthly fee, we will delete the account in 30 days. 

3. After signing up you will first see the Floweditor. This is the heart of the no code machine learning tool where you can build the content of the Virtual Assistant. For the sake of simplicity we have built a demo IVA for you already. 

The blue items are actions. Each action is a topic that your no code AI knows. In the welcome action you will see the greeting. It is polite for your IVA to say something nice to the person in the beginning. 

Floweditor - the heart of the no code machine learing tool for building Conversational AI.
Add topics that your AI can chat about.

You can change the text in the action by simply clicking on it and adding your answer.

Change the content of the Intelligent Virtual Assistant.
Change the text of the answer.

 

You can add pictures, carousels or gifs as well.

 

Add pictures to the Intelligent Virtual Assistant.
Add pictures to the answers.

 

 

Add links to the IVA.
Add links to the answers.

If you want then you can also add links at the end of each action. The link can be to any URL. Simply click the button titled “Button” and add the text that the user will see and the URL in https:// format.

 

4. Next is the menu. In the menu your IVA will present the user with a few popular topics that they can select. If you are a company then for example these can be your Products, Pricing and Value. We have provided these by default. You can change the content by simply clicking on the topic and then editing the text area.

 

Menu provides quick navigation links to topics in the IVA.
Menu provides quick navigation links to topics in the IVA.

5. Topics in the menu can be easily changed by adding new ones through buttons on the top or deleting old ones. Each topic is linked to a topic in the main folder system. For example Product is linked as Main.Product where Main is the root folder and Product is the name of the topic in the Main root folder.

Change topics in the Conversational AI menu.
Change topics in the menu.

6. If the Intelligent Virtual Assistant is unable to help the user you have two options. You can log into the chat panel in Alphachat and toggle on the live chat button. Now when users do not get an answer they like they can choose to be connected to a human agent (i.e. you). 

 

Human-in-the-loop AI
See incoming chats from the VCA in the chat panel.

If you are online the IVA will direct the conversation to you and you can see it in your chat panel. If you are offline, then the IVA will take a message from the user and send it to your email. You can set the email where you wish to forward your conversations from the forms panel.

 

7. Now, let’s turn to training the no code AI. No technical skills needed here as the no code machine learning interface makes it easy. Users will have all kinds of questions for your Intelligent Virtual Assistant. So let’s train it. First, you should add a few topics that the AI should know. As an example we have added topics like Product, Pricing and Value. If you are a company then you might want to edit those and specify what product you offer, how much they cost and why customers should buy them. 

Add topics underneath the menu in the Conversational AI
Adding topics underneath the menu.

 

If you want to build the Intelligent Virtual Assistant to talk about things you care about or your hobbies or you, then simply add a new topic by clicking the + button on the top left. Add the text and you have another thing that the IVA knows. 

Add new topics to the Virtual Customer Assistant.
Add new topics to the Virtual Customer Assistant.

To train the AI, go to the Train tab on top of the action editor. Add here phrases that you think your users might ask if they want the IVA to answer with the action you have just created. For example, if the action is Product then you might add phrases such as “what products do you offer”, “what do you sell”, “what kind of product you provide”. Add a minimum of 5 phrases, 10-15 is good, 20 is optimal for starting out. 

Adding training phrases to the topics of the Conversational AI.
Add training phrases to each topic.

Once you have added the phrases for all your actions, click Train. This will train your AI. It is using a combination of neural networks and machine learning algorithms to train the Intelligent Virtual Assistant to understand user questions. There is no way to write down all the ways in which your users will talk to your bot. And you do not have to. It will learn from the training data that you just entered and make its own guess when users will come and talk to it. You can read more here about how to train your AI after you have put it live.

Training the AI.
Training the AI.

8. Now that you have trained the no code AI, give it a shot. If you have the AI panel open, you see the chat icon on the bottom right corner. Click that icon and ask away. This is what your IVA looks like.

Test your Conversational AI.
Click the bottom right corner in Alphachat to test your AI.

 

Also you can click on the URL in the AI panel. It is a unique URL for your AI where you can test it. You can also share the URL with your friends and colleagues so that they can also try out the IVA you have built. 

 

9. All the questions that the AI gets are visible in the Inbox. From here you can easily see how the IVA recommends to train itself. You can confirm the choice for an action for a phrase that the user has said by clicking confirm.

Inbox for training the AI on actual customer data.
Click Validate in Inbox to train the customer phrase to a topic.

 

Or you can change the action by selecting a different action from the dropdown menu and click confirm. Press Train, wait a bit and the IVA is trained based on the new information.

Training the AI with AlphaChat.
Choose the correct topic from the dropdown menu and then click Validate.

10. To deploy the IVA to RCS you need to have an account with a messaging provider that has an RCS API (e.g. Sinch, MessageBird). If you do not have it yet, reach out to us and we will do it for you. Once you have your API tokens, the IVA will be connected to send and receive messages in RCS.  

You can place the IVA on your website. Simply go to the Widget panel on the left. Copy the code and paste it on the website you wish to deploy the bot. It can be copied to Wordpress, Wix, Hubspot and other website builders. 

Sample code to deploy chat widget with the AI to a website.
Example of code for copying the AI onto a website.

Once the IVA is on your website you can encourage users to chat with the IVA and then ask them to carry on the conversation via Google Messages app. Once users opt-in, they can continue the conversations in Google Messages.

13. Summary

That’s it. You have built your AI. Without a single line of code. And connected it to RCS. Needless to say there are more features in the Alphachat product (e.g. APIs and customizable actions with custom code). But the steps described above are enough for you to get started with your own AI with the no code machine learning tool.

Feel free to play around with Alphachat and explore the ways in which you can use AI chatbots to automate your communications through RCS messaging. If you have any questions, feel free to reach out to us or chat with our virtual assistant on our website.