Use Zapier to create a custom AI Chatbot for your business

Zapier offers a creative solution with its custom AI chatbots, leveraging OpenAI's advanced models. These chatbots enable users, even those without coding expertise, to develop bots that assist in editing content for conciseness, aligning with specific tones of voice, and identifying areas for improvement. This process occurs prior to the involvement of human editors.

Beyond serving writers, Zapier's chatbots cater to a wide array of functions across different teams. They are employed for tasks such as responding to frequently asked questions, supporting customer events, providing automation advice, simplifying complex jargon, and enhancing employee feedback and training.

The blog post aims to provide a detailed walkthrough on creating a personalized AI chatbot using Zapier, including practical examples and use cases for various professional scenarios.

Overview: Creating a Chatbot with Zapier

Creating a chatbot using Zapier is a streamlined process, ideal for both beginners and experienced users. Here's a brief overview of the steps involved, with more detailed instructions to follow:

• Prepare Accounts: Ensure you have an active Zapier account and an OpenAI account.

• Access Chatbot Creation: Visit Zapier's chatbot page.

• Initiate Creation: Click on +Create to start building your chatbot.

• Start from Scratch: Choose the option to Start from scratch and give your chatbot a unique name.

• Customize Settings: Tailor your chatbot's settings according to your requirements.

• Share Your Chatbot: Distribute your chatbot by sharing a public link, embedding it on your website, or integrating it within an Interface.

These steps provide a foundational approach to creating a functional and customized AI chatbot with Zapier.

Step 1: Building Your Chatbot from Scratch

• Access Chatbot Dashboard: Navigate to your chatbot dashboard and click +Create in the top-right corner.

• Choose Setup Method: Ignore the pre-made templates and select Start from scratch in the top left-hand corner to create a new, blank chatbot.

• Name Your Chatbot: In the left panel, under 'Name', enter a unique name for your chatbot, like "Editor Bot," depending on your specific use case.

• Set Greeting Type: Choose between a 'Static' (constant) or 'Generative' (changing) greeting. In the 'Greeting text' field, type the initial message your chatbot will display to users.

• User Input Placeholder: Fill in the 'User input placeholder' field with instructional text that guides users on how to interact with the bot.

• Write the Directive: This crucial step involves detailing the chatbot's role, objectives, target audience, necessary context, response style, and any rules it should follow. For an editing bot, include specifics on style, tone, grammar preferences, word count, and brand voice.

• Start Simple: Begin with a basic directive and refine it based on testing and feedback in the chatbot builder's right-hand panel.

• Set Up OpenAI Authentication: Default to using GPT-3.5 or add your OpenAI authentication for access to GPT-4 or other models. Select your OpenAI account and model from the dropdown, and adjust creativity levels (default 0.7).

• Advanced Options: Customize your browser title and bot's subdomain under 'Advanced options'.

• Save and Test: Save your changes and test your chatbot. Tweak the creativity temperature or refine the directive for better results. Once satisfied, your chatbot is ready to be shared.

Step 2: Adding Your Own Data Sources to the Chatbot

• Access Data Tab: In your chatbot's dashboard, locate and click on the 'Data' tab in the left-hand panel.

• Add Source: Click on +Add source. By default, the data type will be set to 'File'.

• Supported File Types: Currently, the supported file types include .txt, .csv, .json, .PDF, .doc, and .docx.

• Upload Your File: Click 'Browse files' to select and upload your file. For instance, uploading a Zapier blog and style guide allows the chatbot to understand specific content dynamics and provide relevant feedback.

• Optimize File Content: To facilitate better processing by the AI, organize related information in paragraphs rather than bullet points.

• Confirm Upload: After selecting your file, click 'Add source' to upload it to your chatbot.

• Customize Responses: Once the file is uploaded, you can customize how your chatbot responds to queries not covered in your data source. Choose either to allow the AI to respond using its built-in knowledge or to display a custom message for queries not found in the data source.

• Edit Custom Message: If you opt for a custom message, edit the text to convey the desired information.

• Response Generation: If you allow response generation, the chatbot will use publicly available data in the AI model to answer questions.

• Save Changes: After customizing the responses, click 'Save changes' to update your chatbot settings.]

Following these steps, your chatbot will now be able to pull answers from the added data source, enhancing its ability to provide specific and tailored responses.

Step 3: Adding Actions to Your Chatbot

• Navigate to Actions Tab: In your chatbot interface, locate and click the 'Actions' tab in the top-left corner.

• Initiate Action Setup: Click on Add action to start configuring actions for your chatbot.

• Set Trigger Event: Define the 'User event' as the trigger, which occurs when the user receives a response from the chatbot.

• Configure Action Button: Choose to display a button in the 'Action' field. This button, when clicked, will execute a predefined action.

• Customize Button Text: In the 'Button text' field, input the label for your button, like 'Copy Response'.

• Notification Text: Enter the message to be displayed post-click in the 'Notification text' field.

• Action on Click: In the 'On click' field, choose the action to be performed, like 'Copy response'.

• Create the Action: After setting up the button, click Create action.

• For Advanced Actions (Zaps): To integrate a Zap, follow similar steps, but change the text for your button and notification accordingly.

• Setting Up a Zap: Change the selection in the 'On click' field to 'Run Zap'. After clicking 'Create action and Zap', you will be directed to the Zap editor to complete the setup, such as sending responses to a Slack channel.

Following these steps will allow you to effectively add actionable features to your chatbot, enhancing its functionality and interaction capabilities with users.

Step 4: Adjusting Your Chatbot's Style

Access Style Tab: In your chatbot dashboard, locate and click the 'Style' tab.

Customize Colors: Here, you can change various color settings, including:

• Page background color
• Button color
• Chatbot message background color

Adjust Text Settings: Modify the appearance of text elements, such as:

• Disclosure message text
• User message background

Brand Alignment: Optimize these settings to match your company's brand colors and aesthetic, ensuring a seamless integration of the chatbot into your website or application.

By customizing these elements, you can create a chatbot that not only functions well but also visually aligns with your brand identity.

Step 5: Sharing Your Chatbot

• Share with Team: To share the chatbot with your team, provide them with the public URL, found in the top-right corner of your chatbot's setup page.
• Embed in Your Site: For embedding the chatbot on your website, click 'Share' in the top-right corner.
• Select Embedding Option: In the pop-up window, navigate to the 'Embed' tab. Choose between:

• 'Standard': Embeds the chatbot directly on a web page.
• 'Pop up': Places the chatbot as an overlay fixed to the bottom-right of the page.

• Copy Embed Code: Choose 'Standard' or 'Pop up' and copy the script link or iframe by clicking 'Copy embed'.
• Embed on Your Website: Paste the copied embed code into your site. For example, integrate an editorial bot into a WordPress site using the provided code.

The embedded chatbot will appear on your site as configured, providing an interactive element for users to engage with.

Step 6: Building a Landing Page for Your Chatbots

Access Interfaces Dashboard: Go to the Zapier Interfaces dashboard and click +Create.

Start from Scratch: Select Start from scratch and scroll down to choose Link cards.

Customize Your Page:

• Name your page in the 'Header' field, e.g., "AI for the content team".
• Adjust the header's width and alignment.

Add Link Cards: Click Add link in the right-hand corner to insert new tiles for each chatbot.

Customize Link Cards:

• Click a tile to open the customization panel.
• Add a title, description, and call-to-action (CTA) copy for each chatbot.
• Insert the chatbot's URL in the 'Link URL' field.
• Optionally, add an emoji and choose whether the link opens in a new tab.
• Click Done after customizing each card.

Preview and Share Your Page:

• Preview the landing page using the link in the top-left corner.
• Share the page with your team or others by clicking Share in the top-right corner.
• Set up collaborators, public access, and generate embed code if needed.

This process allows you to create an organized and accessible landing page for various chatbots, enhancing user experience and collaboration.

Chatbot Ideas

Need some inspiration for some use cases where such a chatbot can assist you in your business? Below are some ideas that you can produce:

PR:

• Generate press releases.
• Set up sentiment analyzers for press articles.

Content:

• Improve writing skills in specific areas.

Team Communication:

• Formulate professional feedback for both positive and difficult conversations.

Social Media:

• Automate social promotions.
• Find influencers for campaigns.

Marketing:

• Translate corporate jargon to clear language.
• Deal with legalese, cultural context, or specific buzzwords.

Personality Bots:

• Offer ideas for birthday presents.
• Assist in planning work renovation projects.

The variety of chatbots you can create with Zapier is vast, catering to different professional needs and personal assistance. Whether it's streamlining communication, aiding in content creation, or simplifying complex tasks, chatbots can be tailored to meet a wide range of requirements.