The Chatbots section is the place that allows you to set up automatic messaging in the chat, as well as create chatbots, which will be able to handle the most common inquiries asked by your visitors.
Please note that some of the pre-made Chatbots are available on Shopify only.
The Chatbots Interface
The Chatbots section is divided into one main panel where you can edit your existing bots, set the chatbot launcher and explore preset workflows.
The three dots menu lets you check, edit, turn off or delete your currently added chatbots. It also contains stats on how many times each particular chatbot was triggered or used by the visitors.
Some templates from the list are added through a quick setup guide, which asks what the messages should say.
THE BOTS LAUNCHER
The chat has the option to start with a small introduction from your Bot, allowing the visitor to choose how to proceed. The Bots Launcher panel allows to set custom messages, as well as add some graphics to make it more personalised.
To find out more about how to set up the Bots Launcher, please go to this part of the article.
Creating a new Chatbot
Adding a new chatbot can be done in two ways: you can use a ready template or build your very own chatbot from scratch.
One of our most requested chatbots is the one that sends a custom greeting, e.g. ‘Hello, how can I help you?’ which is triggered only once upon the first visit on the website, and it’s called ‘Welcome new visitors‘. You can find it in the Chatbots section in your Tidio admin panel. To test it out, simply visit your website in a new Incognito window to appear as a fresh visitor.
If you wish to use one of the templates, simply click the add another bot button.
– this will open a list of the most popular and most commonly used scenarios, grouped into categories to help you find them easily.
After choosing a template, simply click on it – this will open a quick setup tour, allowing you to define the text that will be shown to the visitors (if you choose one of the Shopify-specific Bots, clicking the Add button will ask to connect the Bot with your Shopify store).
Creating a Bot manually allows you to build your own scenario using nodes, which represent the actions for a particular moment/step.
This is done on a grid which allows you to add and connect all of the different nodes (triggers, conditions, and actions). It’s where you create your “map” or a “workflow” – simply said, a chain of events that can take place on the visitor’s side.
A new workflow always has to start with a trigger – something the visitor needs to do to start this particular automation.
Once you have the starting point specified, you can add and connect the actions that should take place after the trigger has occurred. Since the grid shows the logical chain of events, the actions need to be connected in a chronological order; the line that’s connecting them needs to be drawn from the action to the reaction that follows.
The basic scenarios usually consist of just a trigger and a following action, but you can also build more complex reactions by using conditions. Conditions filter out the path for the actions, making them target things like the country, operating system, or even specific names. Each condition branches out with a Yes and No path, allowing you to set a different action for each path.
Conditions have their own internal settings, allowing you to specify if they should target the specific element, omit it, or work as a wild card.
The tool also allows creating an interactive scenario, where the visitor needs to choose an appropriate option or provide some details to move on with the inquiry. These can be done using the action nodes ‘Decision (Quick Replies)’ decision’ and ‘Ask a question.’
‘Decision (Quick Replies)’ asks the visitor to select a preset option, allowing you to build, e.g. an FAQ Bot that will give answers to a chosen question. The node will enable you to create a list of selectable options (quick replies), which branch out of the node to connect with appropriate actions.
Decision node supports up to eleven answers!
Below every Decision (Quick Replies), Decision (Buttons) and Decision (Card Messages) nodes editing panel, it is possible to edit the “Transfer to operator message”. The message is set to trigger when the conversation of the bot with a visitor has ended, and it will forward the conversation to a human operator.
The toggle allows disabling the message. Every Decision node will have the option. It will also have the editing panel when you click on “here.”
‘Ask a question’ will allow the visitor to manually type in a reply to a pre-prepared question. The node also allows setting up a validation option to make sure the person types in exactly what you need (e.g. if you request an e-mail address, the validation can check if it’s indeed an e-mail address or just random text).
Now, once you’re happy with your setup but afraid to publish it, you can use the Test it out feature. This will open a new window in which you can test out your chatbot, even before saving it!
Managing existing Chatbots
This panel allows you to quickly add a Bot or an automatic message, which you can choose from a list of the most common scenarios. The templates are grouped thematically, based on what they do and when they’re triggered.
Once you add at least one chatbot, it will be listed in the Manage panel. The panel will group everything according to the used triggers, allowing you to quickly navigate and find the one that’s responsible for a particular action.
Once you click on one of the added Chatbots, the right side of the panel will show its current usage statistics, as well as the options to edit, turn off, or delete the particular workflow.
Each workflow can be turned off temporarily. You can also choose if the workflow should work when none of the operators are online.
The Bots Launcher
To make the chat start off with a short introduction by your Bot, you can set up a launcher that can show all of the possible Bot scenarios, so the visitor can choose the question they want to ask.
The Bots Launcher can consist of a text message, an image/gif, and the automatic scenarios that you currently have added in the Manage panel (they will show as clickable buttons).
In order for the Chatbot to appear as a selectable option in the Launcher, you need to add a ‘Visitor clicks the bots button’ trigger in the editor, making it one of the starting points on the grid.
The trigger also allows you to name the button – this is what the visitor will see in the Launcher’s list.
Another part of our Bots Launcher system is the “Transfer to operator message.” It triggers when the bot has finished its work with the visitor and the conversation was sent to the operator. It can be enabled, disabled and edited here:
Changing the text of the message is made through the clickable “here” button below. Once clicked on, it will open a panel allowing to adjust the message there:
Making chatbots GDPR-friendly 🇪🇺
To make some of the workflows compliant with the GDPR laws in Europe, we suggest you add a step to your workflow to ask visitors and future leads for consent. Here’s how you can do this:
From there, we can use the “Based on contact property” node, to continue the workflow only when a user gave their consent.
In this example, the workflow will continue only for users who gave us the permission. Their answer is saved in the Contacts section, under Contact Properties.
Another option to comply with the GDPR is to use the dedicated field in the Pre-chat survey. Read more about it here.
If a chatbot doesn’t seem to appear or continues to show an old message, please open your website in a new private/incognito window – the chat widget saves the past visits and data in the browser’s local storage, so you’ll need to test the new setup in a clean session.
If you’d like to go even more in-depth you can check out our eBook with over 30 pages covering the most common scenarios and explaining how to build them, clicking here. Furthermore, you might be interested to read more about our Tidio Shopify Chatbots.
If you have any other issues with the chatbots, please reach out to us at [email protected] or use the chat in the bottom-right corner of our website.