Build AI-powered customer conversations in Google Maps and Search with Google’s Business Messages

Make sure that you’re signed in to the correct Google Account when you open the

Apps Script template. The current account can sometimes switch to

your default account without you noticing. Gemini is also only available in English, though Google plans to roll out support for other languages soon.

Also, anyone with a Pixel 8 Pro can use a version of Gemini in their AI-suggested text replies with WhatsApp now, and with Gboard in the future. To learn more about Business Messages, check out our developer website and join our community. You can also check out the Business Messages Helper Bot powered by this technology available in our developer support section here. After creating the JSON structure for a card, I click Save and re-enter the chat on my phone asking “Who made this bot? The document can be a URL pointing to an existing FAQ for a business or if you don’t have one, you can create an FAQ using Google Sheets, download it as a CSV, and then upload the CSV to initialize Bot-in-a-Box.

Building a chatbot that integrates Google Chat, Dialogflow, and Sheets (or another data source) is straightforward. For more details, watch our Integrate Dialogflow with Google Chat video, where we talk in more depth about the process, and check out the full source code on GitHub. Chatbots have existed for years, so let’s start by walking through the below video to visualize how generative AI changes the game.

Switching back  to responses grounded in the website content, the assistant answers with interactive visual inputs to help the user assess how the condition of their current phone could influence trade-in value. Once your bot is finished processing its task, it can respond back to the user in one of two ways. The first is with a simple acknowledgement (aka OK) response letting them know their action was handled correctly and close out the dialog.

Develop Google Chat apps

Finally, the message also offers additional annotations about the event and includes any argumentText supplied by the user already parsed from the command text itself. Google’s Business Messages is about enabling all businesses to welcome their customers and open a conversation, where and when they need it, as naturally as when a customer enters a store. Dialogflow is Google’s natural language understanding tool that processes user input, maps it to known intents, and responds with appropriate replies. Dialogflow is a natural language understanding platform that simplifies the design and integration of conversational user experiences for mobile apps, web apps, bots, and more. Generative AI App Builder’s step-by-step conversation orchestration includes several ways to add these types of task flows to a bot.

Developing your own custom Google Chat bot is a great way for users and teams to interact with your solutions and services both directly and within context as they collaborate in Chat. More specifically, Chat bots can be used in group conversations to streamline workflows, assist with activities in the context of discussions, and provide information and notifications in real time. Chat bots can also be used in direct messages, offering a new way to optimize workflows and personal productivity, such as managing project tasks or reporting time activity. Because use cases for bots are varied, you can consistently reach a growing audience of Chat users over time, directly where they work and uh-hum, chat. This no-code solution can easily be extended using Dialogflow’s fulfillment feature to pull in business information from a database or API, allowing you to support even more complex user journeys. For this example, I’m going to create a helper bot that can answer questions about Business Messages.

Create a Google Chat app that you can directly message and that

responds by echoing your messages. Now that I have the basics setup, I’m going to click on the Integrations menu item in the developer console and configure Bot-in-a-Box via Dialogflow. To get started, since I’m already registered for Business Messages, I’m going to go to the Business Communications Developer Console and create a new agent. In this article, I’ll give a brief overview of Business Messages, how to get started developing with the platform, and then walkthrough how to set up an AI-powered conversion using the Bot-in-a-Box feature. In the Google Cloud Community, connect with Googlers and other Google Workspace admins like yourself.

Step 3: Create and configure your GCP Project

I am prompted to either create a new Dialogflow project or connect to an existing one. I’ve already created a Dialogflow project, so I choose to connect to an existing project and then I follow the prompts to set up the authentication between my Business Messages agent and the Dialogflow project. After creating an agent on behalf of a business, the chat button isn’t immediately available to Google Search and Maps users. All agents must go through a verification and launch process before the chat button will be shown for businesses in Search and Maps.

The second part of the one-two punch of new Google Chat bots features are dialogs. You can foun additiona information about ai customer service and artificial intelligence and NLP. This is a brand new capability being introduced to the Chat bot framework that allows developers to build user interfaces to capture inputs and parameters in a structured, reliable way. This is a tremendous step forward for bot usability because it will simplify and streamline the process of users interacting with bot commands.

Dialogflow uses a pretrained NLP model, and these training phrases are the realistic questions that help train a specific model for our use case. This is where we define how to use the entities we just created by defining Actions and Parameters. Bots in Google Chat are automated special accounts that provide you with certain services, such as looking up information, scheduling meetings, or doing specific tasks. It is followed by 90+ third-party bots like Zoom, GitHub, Polly, Trello, Salesforce, Zapier, and more. Satisfied that the Pixel 7 Pro is a compelling upgrade, the shopper next asks about the trade-in value of their current device.

As the user asks questions, text auto-complete helps shape queries towards high-quality results. For example, if the user starts to type “How does the 7 Pro compare,” the assistant might suggest, “How does the 7 Pro compare to my current device? ” If the shopper accepts this suggestion, the assistant can generate a multimodal comparison table, complete with images and a brief summary. To add trusted testers and learn more about testing interactive features, see

Test interactive features for

Google Chat apps. You now have working app code that you can try out (as described in the

following steps) and then customize to meet your requirements. We built our bot in three easy steps, which should look familiar to you if  you’ve completed the Deconstructing Chatbots video series.

Additionally, after the quickstart, you’ll have configured a webhook and created your first Business Messages agent. Agents include properties like the brand’s logo, the agent’s display name, the welcome message that greets a user, and more that define how the conversation will look and where the chat button will show up once launched. As shown in the example image below, I’ve searched “Bridgepoint Runners” and the results point me to a local Bridgepoint Runners store, which contains buttons to call, get directions, or go to their website.

In today’s world, where online collaborative work is crucial and maintaining productivity is key,  chatbots have an important role to play. Workers frequently need to incorporate information from external sources in their communications, and chatbots can help them find that information all in one place. In this post we’ll walk you through a bot that was inspired by a real use case here at Google. Well, Chats has over 90 bots, which will help you enhance your productivity and efficiency.

In large companies such as Google, it can be difficult to find out which person is responsible for a specific product area. When customer teams have a question they often have to go through many ad hoc trackers such as the one here in Sheets (sample scrubbed data) to find answers. You can then follow the given instructions and chat with it to perform actions or tasks. You can also directly use bots while chatting with others- all you need to do is type ‘@,’ and it’ll show you the available bots. Google Chats is a handy platform for G Suite users to communicate and collaborate with their team and colleagues. It comes with a host of features such as dark mode, chat rooms, smart replies, and more.

With these capabilities, developers can focus on designing experiences and deploying generative apps fast, without the delays and distractions of implementation minutiae. In this blog post, we’ll explore how your organization can leverage Conversational AI on Gen App Builder to create compelling, AI-powered experiences. In the near future we’ll be adding more posts with interesting examples of what you can do with chatbots, such as linking them to APIs and services, and even tapping into Google AI ML platform. In the meantime, check out some examples of bots that are built in to Hangouts Chat in this recent blog post. Remember that all of this is technically an experiment for now, and you might see some software glitches in your chatbot responses.

Here, you’ll create and configure your GCP project so that it can serve as the chatbot backend. Now that you’re ready to get started, here are the basic steps to build and test your own chatbot in Apps Script. Apps Script is a cloud-based scripting language and runtime environment based on JavaScript.

For the purposes of this example, I created an FAQ as shown in the document below and uploaded it to Bot-in-a-Box. Even without launching an agent, you can test the message flow by using the test URLs from a mobile device that are autogenerated when you create the agent. The test URLs for an agent can be copied or sent to your email from within the Business Communications Developer Console and are also available as a property of the agent if you’re using the API.

For developers, you can design UIs that are targeted to work precisely with the inputs you need users to supply your commands, without having to parse out arguments and logically infer the intent of users. In the end, dialogs will greatly expand the type of solution patterns and use cases that Chat bots can handle, as well as making the experience truly richer and more rewarding for users and developers alike. Business Messages’s Bot-in-a-Box makes getting started with conversational AI easy. Bot-in-a-Box takes advantage of Google AI tools like Dialogflow to easily convert an existing FAQ into an automated Business Messages solution. Within minutes, you could launch a lifelike virtual agent that provides relevant responses to the most common questions a business receives from customers.

The assistant then asks if the shopper needs anything else, with the user replying that they’re interested in switching to a business account. This answer triggers the assistant to loop a human agent into the conversation, showcasing how prescribed paths can be seamlessly integrated into a primarily generative experience. When a user invokes your bot via a Slash Command, there is a slashCommand field attached to the message sent to the bot that indicates the call was initiated from a Slash Command. Remember users can still @mention your bot to call it directly by name without a / command and this helps you distinguish the difference. The message also includes the corresponding commandId for the invoked command based on what you set up in the bot configuration page, allowing you to identify the user’s requested command to execute.

Programs

Participate in product discussions, check out the Community Articles, and learn tips and tricks that will make your work and life easier. The other option is to respond with another dialog, allowing you to follow-up with a new or revised dialog useful for complex or conditional input scenarios. This is accomplished as it was originally when you called a dialog using a dialog card within an ActionResponse to get started. Please include what you were doing when this page came up and the Cloudflare Ray ID found at the bottom of this page.

To get started, read more about Gen App Builder and conversational AI technologies from Google Cloud, and reach out to your sales representative for access to conversational AI on Gen App Builder. We’ve been pleased to see the innovative results our customers have already achieved with pre-GA releases of Gen App Builder. For example, Orange France recently launched Orange Bot, a French-language generative AI-enabled chatbot. The chatbot stems from a long-term business vision to transform the customer relationship, optimize management costs, and offer ever more helpful and user-friendly experiences.

The first step to setting up Bot-in-a-Box is to enable the Dialogflow integration. Currently, Bot-in-a-Box only supports the Dialogflow Essentials (ES) version of Dialogflow. However, you can integrate with Dialogflow Customer Experience (CX) by calling the CX APIs directly from a configured Business Messages webhook and programming the conversion to and from the Business Messages APIs. Once the agent is created, I can select the agent to see additional details and access the various configuration options. Business Messages is a mobile conversational channel that combines entry points from Google Maps, Search, and brand websites to create rich, asynchronous messaging experiences.

Dialogflow integrates with many popular conversation platforms like Google Chat, Google Assistant, Slack, and more. Direct end-user interactions are handled for you, so you can focus on building your agent. Each integration handles end-user interactions in a platform-specific way, so see the documentation for your integration platform for details.

How we’re keeping up with the increasing demand for the Google Workspace Administrator role

Also, because Apps Script offers native G Suite integration (including authentication), it can be the most pragmatic choice for building G Suite-centric chatbots on Google Hangouts. Additionally, you’ll have access to the Business Communications Developer Chat PG Console, which is a web-based tool for creating and managing business experiences on the Business Messages platform. It provides the same functionality as the Business Communications API, but is a faster and more convenient way to get started.

Once you have configured a Slash Command to trigger a dialog, it will send an onMessage event when it’s invoked as it would before, but now it includes new details that indicate it is representing a dialog request. To handle this event you can use the example above with non-dialog Slash Command, using the commandId you can use a switch to determine what the user requested. As you experiment with Gemini Pro in Bard, keep in mind the things you likely already know about chatbots, such as their reputation for lying. With Bot-in-a-Box’s FAQ support, within just a few minutes, without writing any code, I was able to create a sophisticated digital agent that can answer common questions about Business Messages. Once the authentication is complete, I see an updated integration view like the one below.

Users can now simply type “/” in the message line to reveal a list of all the functions offered by the bots available to the room or direct message, and select the one to their liking to execute it. Slash commands can be invoked standalone (e.g. /help) or include user added text as parameters (e.g. /new_task review project doc ) that the developer can handle when invoked. To help make bot command discovery even simpler, the slash commands list filters matching commands once the user starts typing beyond the / (e.g. “/h” shows all commands beginning with H). This is super helpful as more and more bots are added to a room, and as more bots with slash commands are introduced by developers. Also included directly in the Slash Command UI is a description of what each command does (up to 50 characters), easing the guesswork out of learning. Enterprise search apps and conversational chatbots are among the most widely-applicable generative AI use cases.

The Business Messages agent is able to respond immediately with a matching answer pulled from the FAQ document I created. When you navigate to the test URL, the conversation with your agent will automatically open. This mimics google bot chat the experience that a user would see when tapping on a chat button for a launched Business Messages agent. Because different people talk differently, we use training phrases to provide different examples of user requests.

One of the current strengths of Bard is its integration with other Google services, when it actually works. Tag @Gmail in your prompt, for example, to have the chatbot summarize your daily messages, or tag @YouTube to explore topics with videos. Our previous tests of the Bard chatbot showed potential for these integrations, but there are still plenty of kinks to be worked out.

Future releases are expected to include multimodal capabilities, where a chatbot processes multiple forms of input and produces outputs in different ways. In this example, I’m responding with a simple text message, but what if I want to take advantage of Business Messages’s rich message support and respond with something like a rich card? I can do this by using Dialogflow’s custom payload option and use a valid Business Messages rich card payload in the response to create the card. Before setting up my Bot-in-a-Box experience, I want to make sure my agent is properly configured to greet new users. I click on Agent Information and from here I can set a welcome message and up to 5 conversation starters that help the user understand how to interact with the automated agent.

As with previous generative AI updates from Google, Gemini is also not available in the European Union—for now. I downloaded this Sheet as a CSV and uploaded it as the initial data set for Bot-in-a-Box to train with. The quickstart will have you deploy code to Google App Engine and the life of a message for your Echo Bot sample will look something like the image below. Once you’ve worked your way through the quickstart, you’ll have registered a Google Cloud Project and that project will have two APIs enabled, the Business Communications API and Business Messages API. Fulfillment is where we glue everything together, connecting all the APIs in a Node.js Firebase function. In this case, we use the Sheets API but you can connect to any backend you choose.

• As a developer, slash commands are straightforward to implement, and daresay essential in offering a better bot experience.
• The Business Messages agent is able to respond immediately with a matching answer pulled from the FAQ document I created.
• Slash commands can be invoked standalone (e.g. /help) or include user added text as parameters (e.g. /new_task review project doc ) that the developer can handle when invoked.
• The first new(er) feature that you can leverage to enhance the usability of your Chat bots are slash commands.
• This is a brand new capability being introduced to the Chat bot framework that allows developers to build user interfaces to capture inputs and parameters in a structured, reliable way.

Now with dialogs, users can be led visually to supply inputs via prompts, versus having to rely on wrapping bot commands with natural language inputs — and hoping they correctly executed syntax the bot could decipher. The first new(er) feature that you can leverage to enhance the usability of your Chat bots are slash commands. Released a few months back, slash commands simplify the way users interact with your Chat bot, offering them a visual leading way to discover and execute your bot’s primary features. Whereas the assistant generated earlier answers from the website’s content, in the case of the lens question, the response involves information that’s not contained in the organization’s site. This flexibility allows for a better experience than the “Sorry, I can’t answer that” responses we have come to expect from bots. When applicable, these types of responses include citations so the user knows what source content was used to generate the answer.

We are also continuing to add new features to Enterprise Search on Gen App Builder with multimodal image search now available in preview. With multimodal search, customers can find relevant images by searching via a combination of text and/or image inputs. Designing the actual elements that the dialog renders is where you draw from the Google Workspace Add-on Card-based framework. If you’ve built a new generation of Google Workspace Add-on, this part will be familiar where you construct widgets, add headers and sections, create events, etc.

It offers direct code access to a variety of Google products and APIs via its extensive library of services. So how is the anticipated Gemini Ultra different from the currently available Gemini Pro model? According to Google, Ultra is its “most capable mode” and is designed to handle complex tasks across text, images, audio, video, and code. The smaller version of the AI model, fitted to work as part of smartphone features, is called Gemini Nano, and it’s available now in the Pixel 8 Pro for WhatsApp replies. Despite the premium-sounding name, the Gemini Pro update for Bard is free to use. With ChatGPT, you can access the older AI models for free as well, but you pay a monthly subscription to access the most recent model, GPT-4.

Google does not allow access to Bard if you are not willing to create an account. Users of Google Workspace accounts may need to switch over https://chat.openai.com/ to their personal email account to try Gemini. A version of the model, called Gemini Pro, is available inside of the Bard chatbot right now.

Now with a working dialog, all there is left to do is handle user events once it’s displayed. Again this is similar to how you would handle events working with Cards within Add-ons. Your bot will receive an event that is type CARD_CLICKED with a DialogEventType set to SUBMIT_DIALOG. The actionMethodName value will let you know what element the user clicked to process the request, e.g. ‘assign’ as depicted below. The response includes the formInputs details which are the user provided inputs returned from the dialog, which you can process as your solution needs to.