Most teams working with forms may require filling in multiple fields, which can make the process repetitive or time-consuming. Instead of entering values manually each time, parts of the form can be filled automatically based on available inputs or logic, reducing the effort required and making the process more efficient.
Clappia allows you to simplify this by using AI to take a single input and automatically fill the relevant fields in the form. Instead of asking users to enter data in multiple places, they can provide it once, and the system can handle the rest.
In this guide, we will walk through how to set this up in Clappia by taking a customer support example, where a complaint message is submitted and the system automatically fills in necessary details in the form itself.
Suppose you receive customer queries or complaints and need to break down the information they provide into useful pieces so it is easier to work on resolving them. Instead of manually reading through the entire message and filling out different fields one by one, you can set up an app where all the input is captured in a single place and the rest is handled automatically.
In this setup, your app can have a field where you enter the customer’s message — this could be a note, a complaint, or even a full conversation pasted into the form. Once you submit it, Clappia can automatically fill in the remaining fields in the app with the required information. This can involve extracting specific details or even performing calculations, depending on your requirement.
In this guide, we will focus on extracting the required information. The app used here will consist of a Multi Line Text field to capture the input, along with other fields that are not filled by the user but are automatically populated by the system after submission.
This system uses the following Clappia features. Each links to its dedicated help article for detailed configuration guidance.
This system uses a single app. The workflow chain runs automatically after a submission is made, extracts values from the complaint text using AI, and populates the structured ticket fields without any manual input from the agent.
From your Clappia dashboard, click Create App. Go to Design App and name it something like "Customer Support Ticket".
Click Add Section and name it Complaint Submission (or according to your requirement) on the right panel. This is where the raw message is entered. Click on ‘Add Field’ to create the form.
| Field | Type |
|---|---|
| Customer Name | Single line Text |
| Customer Email | Email Input |
| Customer Phone Number | Phone number |
| Complaint Message | Multi-Line Text |
These are the only fields the user needs to fill in. The next section gets populated by the workflow that runs after submission.
In the same app, add another section and name it something like Ticket Details. The fields in this section will be populated automatically using AI. They are left empty at the time of submission and filled in after the user submits (which is when the workflow runs in this scenario).
Click on ‘Add Field’ and add the required fields.
| Field | Type |
|---|---|
| Ticket ID | Unique Numbering |
| Issue Type | Single Line Text |
| Order Number | Single Line Text |
| Summary | Single Line Text |
| Priority | Single Selector |
Here the Ticket ID will automatically generate a unique ID for the customer complaint. This is not done by the workflow but the feature itself. The remaining fields will be filled in by the system.
Add another section and name it as Review. This section is optional in this scenario but can be used by the agent after the AI has populated the ticket. These fields will be filled in later by the person by editing the submission in the Submissions tab.
| Field | Type |
|---|---|
| Assigned To | Single Line Text |
| Resolution Notes | Multi Line Text |
| Resolution Date | Date |
| Attachments | File Upload |
Go to Design App > Workflows and open the New Submission Flow (opened by default).
Click Add Step and select AI Workflow.
In the configuration panel on the right, select your preferred AI provider and model. Then write your prompt in the instructions field. The prompt is what tells the AI what to read and what to return.
You can reference fields from your app by typing @ followed by the field name and select it. This is useful if you need the AI to analyse inputs submitted by the user in specific fields.
A well-structured prompt for this use case looks like this:
You are a customer support assistant. Read the complaint message in {message} and extract the following fields. Return your response as a valid JSON object only, with no explanation, no markdown, and no additional text.
Fields to extract:
To insert the Complaint Message field into the prompt, type @ and select the field. It will appear as a variable like {message} in the prompt text.
The AI will return a response like:
{
"issue_type": "Damaged Delivery",
"priority": "High",
"order_number": "4578",
"summary": "Customer received a damaged shipment with two missing items and reports this is a recurring issue."
}
Once you save this step, a variable gets generated for this AI workflow step which will be used in the Code workflow.
After the AI workflow add a Code Workflow. This step takes the raw JSON string output from the AI Workflow and breaks it into individual values that can be mapped to submission fields.
In the code editor, write the following:
function main() {
output = {};
var raw = {ai};
var parsed = JSON.parse(raw);
output['issue_type'] = parsed.issue_type || "";
output['priority'] = parsed.priority || "";
output['order_number'] = parsed.order_number || "";
output['summary'] = parsed.summary || "";
return output;
}
The ai workflow variable {ai} is used as the input.
Under Output Fields, define the four keys: issue_type, priority, order_number, and summary. These output keys are now available as variables for the next step.
After the Code workflow, add the Edit Submission workflow.
Set the target app as the same Customer Support Ticket app.
For the record to edit the correct submission, the Filter used can be the Submission id. These are unique to every submission by default which you can use. Simply select Submission id from the drop down and under it in the input box, type @ and select submission id. This tells the workflow to update the submission that just triggered it.
In Set Field Values, you will be mapping the fields that need to be edited in the app. Map the fields as follows:
| Submission Field | Value |
|---|---|
| Issue Type | {code#issue_type} (from Code Workflow output) |
| Priority | {code#priority} (from Code Workflow output) |
| Order Number | {code#order_number} (from Code Workflow output) |
| Summary | {code#summary} (from Code Workflow output) |
To insert the Code Workflow output variables, type @ in each field mapping and select the corresponding output key.
Once this step runs, the submission is updated with all four extracted values. The user can now open the submission and see a fully classified, prioritised, and summarised ticket.
If you want high-priority tickets to be handled differently, for example, notified to a team lead immediately, add an IF Workflow after the Edit Submission step.
Set the condition as:
{priority} = "High"
If the condition is true, add an Email Workflow step to notify the relevant person. Include the summary, issue type, order number, and a link to the submission in the email body using field variables (type @ to insert them).
Once the core system is working, you can extend it based on your team's needs.
For SLA tracking, you can add a Date field called Response Deadline and use a Calculation block to set it as the submission date plus a fixed number of hours depending on the priority level returned by the AI.
For analytics, go to the Analytics tab inside Design App. You can build a bar chart showing ticket volume by Issue Type, a breakdown of tickets by Priority, or a line chart tracking how many tickets come in per day. These dashboards give team leads visibility into complaint patterns without opening individual submissions.
For setting statuses for your submissions, you can head over to Configuration under Design App. Click on Status and on the right panel add the necessary statuses for your requirement like Open, In Progress, Resolved, Escalated. This is useful for users to know if a ticket has been resolved or not instantly instead of going through all the details of the Submission.
For user access, go to Workplace Settings > Users to add your team. Then open the app and go to Design App > Distribute to assign access. Customers or frontline staff who submit complaints may only need access to fill in the Complaint Message field. Agents who review and act on tickets need access to the full submission including the Ticket Details and Review sections. You can use field-level permissions to prevent the AI-populated fields from being edited accidentally by submitters while still allowing agents to update the Status and Agent Notes.
Once everything is set up, the process runs as follows.
A user submits a complaint message through the app; either on the web or on the Clappia mobile app available on Android and iOS. They type the message in the Complaint Message field and submit. For example:
My order #4578 arrived damaged and two items were missing. This is the second time this has happened.
That is all they need to enter apart from the Customer details (which could also be used to fill in automatically but here we have not taken it that way).
Immediately after submission, the workflow chain runs automatically in the background.
The AI Workflow reads the complaint message and returns a structured JSON response identifying the issue type as Damaged Delivery, the priority as High, the order number as 4578, and generating a one-sentence summary of the complaint.
The Code Workflow then parses that JSON and extracts each value into a separate output.
The Edit Submission workflow then writes those values into the corresponding fields on the ticket — Issue Type, Priority, Order Number, and Summary — updating the submission without any agent input.
The user opens the submission from the Submissions tab and sees a ticket that is already classified and ready to act on. If the AI made an error — for example, it set the wrong priority or misread the issue type — the user can correct it directly in the submission.
The Status field is set to Open by default and the agent updates it as the ticket progresses.
If the ticket is High priority and the optional notification step is configured, the team lead receives an email the moment the classification runs, with the key details included.
On using the Clappia mobile app, the experience is identical. Users can open submitted tickets on the Clappia app, review the AI-populated fields, add notes, update the status, and escalate if needed, all from their phone.
This system eliminates the manual classification step from your support workflow, reduces response time on high-priority issues, and gives your team a structured view of every complaint from the moment it is submitted. As your support operations grow, you can extend the same app with additional fields, more nuanced AI prompts, integration with external tools through Clappia's API capabilities, or a multi-app structure where tickets feed into a resolution tracking system.
To get started, log in to your Clappia workspace, create the app, and configure the workflow chain. If you need help with any of the steps, the Clappia support team is available and the linked help articles cover each feature in detail.
L374, 1st Floor, 5th Main Rd, Sector 6, HSR Layout, Bengaluru, Karnataka 560102, India
3500 S DuPont Hwy, Dover,
Kent 19901, Delaware, USA
.svg){kind=small}
3500 S DuPont Hwy, Dover,
Kent 19901, Delaware, USA
L374, 1st Floor, 5th Main Rd, Sector 6, HSR Layout, Bengaluru, Karnataka 560102, India
