How to generate a response proposal with AI and notify it on Slack when a chat starts

The template will be copied to My Project. You can access My Project from the left menu bar.

‍

‍

Step 3: App Trigger "When a new customer chat is created"

Let's click on the first step of the template.

‍

‍

Here, you will set up to retrieve ticket information issued by Channel Talk.

(1/2) Select Integration Account and Action

This item is already pre-configured.

• App Trigger Title
• App
• Account information integrated with Channel Talk
• Trigger Action

→ Click Next.

(2/3) API Detailed Settings for App Trigger

‍

‍

• Webhook Name
  • Enter any value that suits your purpose. Example: When chat starts
• Access Key
  • Enter the access key obtained from Channel Talk.

(3/3) API Detailed Settings for App Trigger

‍

‍

• Webhook Event
  • Start a new chat in Channel Talk and click Test.

→ Click Test. → Successful. The following items can be retrieved. The retrieved items will be used in Dify's prompt (instruction text).

‍

‍

→ Click Save.

Step 4: Integrate with the App "Send Chat Message"

Click on the second step of the template.

‍

‍

Here, you will set up a prompt (instruction text) in Dify.

(1/2) Select Integration Account and Action

This is already pre-configured.

• App Trigger Title
• App
• Account information integrated with Dify
• Trigger Action

→ Click Next.

(2/2) API Connection Settings

• query
  • Quote the inquiry content from the customer.
  • Click the edit field to display the output obtained from Channel Talk. The code {{details}} is embedded in the pre-settings, but customization by quoting other codes is possible.

‍

• user
  • Set an arbitrary identifier using alphanumeric characters.
• Variables are not mandatory, but utilizing them allows for more flexible and efficient prompt and notification settings. For example, dynamic information such as "name" or "date" can be automatically reflected in the prompt, reducing setup effort and enabling optimal response to users.

→ Click Test. → Successful. You can obtain Dify's response proposal (response content).

‍

‍

→ Click Save.

Step 5: Integrate with the App "Send Message"

Click on the third step of the template.

‍

‍

Here, you will set up a template message for Slack notifications.

(1/2) Select Integration Account and Action

This is already pre-configured.

• App Trigger Title
• App
• Account information integrated with Slack
• Trigger Action

→ Click Next.

(2/2) API Connection Settings

• Channel ID
  • Click the edit field to display a list of channel candidates from the integrated Slack account. When you select the appropriate channel, the ID is automatically quoted.

‍

‍

• Message Content

• Create a template message for the sent message. Click the edit field to open a dropdown, displaying the output obtained in the previous step.

‍

‍

• When you select an item to send to the channel, the embedded code is automatically quoted. Enter the item name before the quoted code to make it easier to understand when viewing.
• This time, we created a template message to notify the chat content from the user and Dify's response proposal.

→ Click Test. → Successful
→ Click Save. This concludes the steps!

Step 6: Turn ON the App Trigger

Once all steps are set, a completion screen indicating the setup is complete will be displayed.

‍

‍

<span class="mark-yellow">The flow bot setup was completed easily without programming.</span> The template used this time is here.

Examples of Automation Using Channel Talk, Dify, and Slack

In addition to the flow introduced this time, you can achieve various business automations by utilizing the APIs of Channel Talk, Dify, and Slack.
All of these can be used with simple settings, so if you're interested, please give them a try!

Examples of Automation Using Channel Talk

Add When a Chat is Created in Channel Talk

This can eliminate the hassle of entering chat content, helping to prevent human errors such as omissions or input mistakes.

Notify the contents of forms and emails to the internal chat of Channel Talk

Since all members can centrally grasp the information, it eliminates the need to individually check the contents, leading to improved work efficiency.

Automation Example Using Dify

When there is an inquiry in HubSpot, create a response with Dify and send a notification

Since the inquiry details and response proposal are notified together, the response speed is likely to improve.

Create a response in Dify and notify when an inquiry email is received in Outlook

By being notified in the communication tool after a prompt response draft is created, the response speed can be improved.

Automation Example Using Slack

Notify Slack when audio data is stored

As soon as audio data is uploaded, it will be automatically transcribed and the results will be notified, making it easier to understand the content.

Read document data posted on Slack with OCR, update the file name, and store it

By reducing the hassle of manually organizing documents and changing file names, you may be able to increase the time you can focus on other tasks.

Notify Slack When a Lead is Registered

When lead information is added, a notification is automatically sent to Slack, facilitating quicker information awareness.

Benefits of Integrating Channel Talk, Dify, and Slack

1. System for Quick Response Even During Busy Periods

AI (Dify) reads the inquiry content, creates a draft response, and notifies Slack.
<span class="mark-yellow">This saves the person in charge from having to create a response from scratch, making the post-inquiry response process smoother.</span>
For example, handling frequently asked questions occurs in many situations, but efficient responses are especially required during busy periods.
By focusing on the procedure of modifying or adding to the draft response generated by Dify as needed, it is expected that the time to create responses can be reduced.
Additionally, there may be cases where the need to research information from scratch is eliminated, further reducing the burden.

2. System for New Staff to Respond with Confidence

AI (Dify) generates draft responses based on the company's rules and tone & manner by learning them.
Therefore, <span class="mark-yellow">even if different staff members handle the response, it is possible to maintain consistency in the reply content by organizing the content based on Dify's draft response.</span>
This may reduce the likelihood of inconsistencies in customer responses.
For example, there may be situations where inexperienced staff spend time researching to handle irregular content or consult with a supervisor multiple times.
By utilizing Dify's suggestions in such cases, it should be possible to proceed with stable, high-quality responses smoothly.

3. System for Everyone to Share the Workload

<span class="mark-yellow">By sharing standardized draft responses with everyone, it may be possible to prevent situations where work is concentrated on specific staff members.</span>
For example, complex and high-difficulty inquiries tend to concentrate on some experienced staff or supervisors, leading to an uneven burden.
Even in such situations, by proceeding with responses based on AI-generated draft responses, it may be possible to standardize the work.
This creates an environment where anyone can respond at a certain level, potentially reducing the burden on each staff member.

Conclusion

In this article, we introduced the method of implementing and the benefits of using a flow bot that integrates Channel Talk, Dify, and Slack.
By utilizing this flow bot, you can expect faster customer responses, improved response quality, and prevention of work becoming dependent on specific individuals.

Furthermore, by utilizing Yoom, this automation can be easily implemented.
<span class="mark-yellow">With the flow bot template, you can achieve app integration with intuitive operations, allowing you to start immediately without programming skills.</span>
Please realize a system for smooth customer responses with this flow bot.

See you next time!