Creating your first bot on Web2Chat is a straightforward process that involves naming your bot and choosing sources for its training. This guide will walk you through the steps to set up your first bot and will detail how to add a website as a source.

1. After logging in, start by entering a name for your bot. This name is internal and used to help you identify your bot within the platform.

1. Once your bot has a name, you'll be redirected to the 'Sources' page. This is where you'll add training sources for your bot.

Website as a Source

1. On the 'Sources' page, click on 'Scrape new domain'. A dialog box will appear with two options: enter a website URL directly, or enter the sitemap URL of the website.
2. Enter the website URL or the sitemap URL and click submit. The platform will begin to scrape all the links of the website or sitemap.

<aside> 💡 For optimal performance, it is recommended to add server-side websites as these are currently better supported by the platform. The platform also supports GitBook, Intercom articles, and various resource centers

</aside>

1. After the scraping process, all links from the website or sitemap will be displayed in a table. This table will show details like page URL, character count, the last trained on, and status.

<aside> 💡 Note: Scraping process might take some time depending on the size of the website or the number of links present in the sitemap. The larger the site or the more extensive the sitemap, the longer it will take to fully scrape the domain.

</aside>

Once scraping done, the platform will list down the links in a table. This table includes columns for the page URL, character count, the last trained on, and status.

• Page URL: The URL of the scraped page
• Character count: The total number of characters scraped from the page
• Last trained on: This will be 'Empty' for a new domain, but will show the date when the bot was last trained using this page for existing domains.
• Status: Indicates the current state of each individual source and its involvement in the bot training process.
• Actions: These actions are used to manage the training status of the pages from the domain
  • Enable/Disable:
    • 'Enable' will appear if the page status is 'Skipped', clicking this will change the status to 'Draft'. '
    • Disable' will be shown if the page is in 'Draft' mode or 'Trained' mode. By clicking this, you can remove the page from the bot's training process.
  • Refresh: This action will appear if the bot status is 'Trained'. By clicking 'Refresh', the platform will rescrape the specific page and retrain the bot with any new content from that page. This allows your bot to stay updated with the latest information from the source.

<aside> 💡 Remember, you can add as many domains as you like until you reach the page limit set by your chosen plan. With each new domain, your bot gains a wider knowledge base to pull from, enhancing its ability to assist your users

</aside>

Files as a Source

1. In the 'Sources' section, select 'Files' as the source type you want to add. This will take you to the file upload interface.

2. Click on the 'Upload Files' button. This will open a file explorer where you can browse and select the files you wish to upload. Select the files you want and click 'Open' or 'Choose' in the file explorer.

   <aside> 💡 Web2Chat currently accepts files in .txt, .docx, .pdf and .md formats

   </aside>

3. The selected files will now appear in the upload interface. Review the files to ensure you've selected the correct ones.

4. Once you're happy with your selection, click the 'Save Files' button. This will initiate the upload process. Depending on the size and number of files, this might take a few moments.

5. After successful upload, you'll receive a confirmation message, and the newly uploaded files will be listed in the 'Files' section. If you've deleted any files during this process, they will no longer appear in the list.

Once you've uploaded files on Web2Chat, you can view and manage them in the File Management Table.

• File URL: This column displays the URL of the file, which usually includes the file name. This URL is generated by Web2Chat when the file is uploaded.
• Characters Count: The number of characters in the file is shown in this column. This helps keep track of your plan's characters limit.
• Last Trained On: This column shows the date when your bot was last trained using the file. If the file hasn't been used for training yet, this will be marked as 'Empty'.
• Status: The status column indicates whether the file has been used for training ('Trained'), is currently being used for training ('Training/Retraining'), or hasn't been used yet ('Draft').
• Action: This column contains actions you can take on the file. Currently, only the 'Remove' action is available. Clicking this will delete the file from your bot's sources, and the bot will no longer use the information from this file for training.

<aside> 💡 Remember, you can upload as many files as you want, keeping in mind the character count limit of your chosen plan.

</aside>

Notion as a File Source

If you utilize Notion for maintaining their knowledge base, Web2Chat provides an easy way to incorporate that content. Notion allows you to export pages as markdown files (.md). You can directly import these markdown files in Web2Chat as a file source.

1. Export the desired pages from your Notion workspace as markdown files (.md). If you need assistance with this, you can follow the instructions provided by Notion on their help page.
2. Follow the above steps to upload these markdown files to Web2Chat.
3. Once the files are 'Processed', they can be included in your bot's training set.

<aside> 💡 This integration allows for a seamless transition from your Notion workspace to the bot training, ensuring your bot is well-versed with your company's knowledge base.

</aside>

Free text as a Source

Free Text Box feature on Web2Chat provides you with a flexible option to directly input content as a source for training your bot

1. In the 'Sources' section, select 'Free Text' as the source type you want to add. This will bring up the Free Text Box.
2. In the Free Text Box, start typing or paste in the content you want your bot to learn from.
3. Once you're done, click on 'Save Text'. This action will save the content you've entered and it will appear in the Free Text sources list.

Like other sources, the Free Text source has a status that can be 'Draft', 'Training/Retraining', or 'Trained'. This status, along with the 'Last Trained On' date, helps you manage the training of your bot effectively.

<aside> 💡 Remember, you can freely add as much content as you want, as long as you stay within the character limit of your chosen plan.

</aside>

QnA as a Source

QnA feature allows you to directly input information in a Question and Answer format, making it a convenient method to train your bot with specific dialogues

1. In the 'Sources' section, select 'QnA' as the source type you want to add. This will bring up the QnA interface.
2. In the QnA interface, you can enter pairs of questions and answers. Start by typing your question in the 'Question' field and the corresponding answer in the 'Answer' field.
3. Once you've entered a QnA pair, click on 'Save'. This action will save the QnA pair, and it will appear in the QnA sources list.

Like other sources, each QnA pair has a status that can be 'Draft', 'Training/Retraining', or 'Trained'. This status, along with the 'Last Trained On' date, helps you track the training progress of your bot effectively.

Each QnA pair also comes with a 'Delete' action. You can use this to remove any existing QnA pair from your sources if needed.

<aside> 💡

Remember, you can add as many QnA pairs as you want, as long as you stay within the character limit of your chosen plan.

</aside>