# How to create a human-agent service for a website Conversation Management is a conversation tool that lets you create a human-agent service as well as chatbots that create a live user experience, answer queries using AI in real-time, and clear all the doubts of a user while visitors exploring your website. Excited? Then, let’s take a tour of how to create a human-agent service that embeds with a website and makes communication simpler. ## Creating a human-agent service ‌To create a human-agent service, sign in to your [Conversation Management](https://chat.quickwork.co) account with your registered email ID and password. If you don't have an account yet, click on the **Create new account** link and create one: ![Figure 1: Conversation Management login page](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQePnhwPGu-RBzvLOa%2Fimage.png?alt=media\&token=d090eddd-48a2-4006-a5c1-aa857e1ab453) Once you have signed in, you will be taken to the **Conversations** dashboard as shown here: ![Figure 2: Conversation Management dashboard](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQeynmVHnxe17jGhvJ%2Fimage.png?alt=media\&token=b620b563-992f-411a-91f0-93e4560566ee) To create a human-agent service, click on the **Settings** tab on the left-hand side panel and click the **Inbox Settings** tab. Then, click the **Add New Inbox** button. You will get a list of channels that Conversation Management support. It supports the following channel options: * **Website**: This is used to create a human-agent service/chatbot for a designated website. It will create a live website chat widget through which visitors can interact with the agents and bots. * **Multichannel**: This is used to create a human-agent service/chatbot for a designated website and social media channels such as Facebook, WhatsApp, LINE, Slack, etc: Since we are creating a human-agent service for a website, select the **Website** channel option: ![Figure 3: Creating an inbox ](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQfwnNPSdkuFtqhhWj%2Fimage.png?alt=media\&token=d4932f44-d2f3-4f46-b0a6-b6cdf38647fc) The next step is to customize your inbox. In the **Customize Inbox** tab, you'll see a form-like page where you need to enter the details of your human-agent service. Here are the input fields to be filled: * **Widget Name**: The name of the chatbot widget. E.g., Food expert human-agent service. This widget will get embed at the bottom right-hand side corner of the website. * **Website Domain**: The URL or domain of the website to which you want to integrate your human-agent service. You can also deploy a local HTML file on the web browser such as file:///C:/Users/kashmira/Desktop/Samplewebsite.html. * **Widget Color**: Select the color for your website widget as per the theme of the website. ![Figure 4: Entering inbox details](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQhEOWVVnH8Pi3ti0z%2Fimage.png?alt=media\&token=04063136-2753-4d5c-9943-dc6c29efecd8) Click the **Continue** button. You'll be taken to the **Add Agents** tab where you can add one or more agents to this inbox as per your requirement: ![Figure 5: Adding agents](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQhVzDKUDnUDUBxbUW%2Fimage.png?alt=media\&token=003db283-d178-4ecd-be44-3a4e16212361) Click the **Continue** button. You'll be taken to the **Finish Setup** tab where you can see a **Messenger Script** of the human-agent service: ![Figure 6: Generating the inbox script](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQhiJEeG2L4htGFm9N%2Fimage.png?alt=media\&token=8066f3c3-53f5-42eb-851f-90be20680c37) Now, open the website code in an editor of your choice and paste this script within the `head` tag: ![Figure 7: Adding script to the website code](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MMJcttyZjUMoGL3lLOZ%2F-MMJiFbHHYGB5mts3VWe%2Fimage.png?alt=media\&token=efde810d-7801-4b08-8998-cd8dac6ed874) Save the code and launch it on a web browser. You'll see a chat widget embed on your website on the right-hand side corner of your screen. When you'll click the widget icon, a chat window will pop up and is now ready for your assistance: ![Figure 8: Human-acgent service on board](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MMJcttyZjUMoGL3lLOZ%2F-MMJkvovfrdUiHF-TDy2%2Fimage.png?alt=media\&token=024687dc-01fb-4b6d-8676-b99fccabe208) ## Executing the human-agent service Send a message, say **Hello**, to the chat window. As soon as a message hits the chat window, a notification can be seen on the Conversation Management dashboard under the **Conversations** tab. Select the **Food expert human-agent service** inbox by clicking the conversations icon and click on the user chat to view the conversation. If you still face the issues to see the conversation, select the **All** option from the filter drop-down menu: ![Figure 9: Message from a visitor](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQi12nDJXGTLHMhTG7%2Fimage.png?alt=media\&token=7a98f437-6d07-4106-9ffd-bb9b54c0439e) {% hint style="info" %} Here, the name of the user who has sent a message gets a dummy name i.e. **Spring-Dawn-515**. You can edit this name and other information by clicking the **Edit** icon under the **Personal Information** section. {% endhint %} Now, you or an agent can start the conversation with the user by typing the message in the text field, say ***How may I help you? Please enter your email address for verification**:* ![Figure 10: Message from an agent ](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MkQdMB6ZpjRQY_6Urbq%2F-MkQiICPcrn6hTXLx44Y%2Fimage.png?alt=media\&token=7f09cf24-e367-4681-8ac8-d65dc49964ca) As soon as you send a response, the user will get a message in real-time on the chat window as shown here: ![Figure 11: A fully-executing human-agent service](https://1811093075-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-M34pLt4SmVLxK2J-oXh%2F-MMJrede8pM2hKCPvM6E%2F-MMJxNx3EQD-M94wXqz0%2Fimage.png?alt=media\&token=8fe2c87e-4279-4eee-b78f-30c0a3c5bf92) In this way, a human-agent service by Conversation Management not only helps to streamline the conversation but also looks into lead enrichment as it records the information of each user or visitor reaching you via your web portal.