Skip to main content

Make Waves ’26 Step into AI’s glass box. Prague, 19–20 Oct.

Register now (opens in a new tab)
Tracira
Tally

Connect Tracira and Tally

Explore ready-made Tracira and Tally templates and start automating in minutes.

Get started free

Create a custom workflow

Integrate Tracira with Tally and automate the work that moves between them — no code required.

Create a custom workflow

Every automation starts with a trigger and an action. Select one from each app to generate a workflow tailored to your needs.

Tally
Trigger: Watch New Responses
Tracira
Action: Select an action
1. Select a trigger

  • Triggers when a new form response is submitted.

2. Select an action

  • Downloads a file stored on a Tracira output and returns it as binary data. Map the File URL or File Key from an attachment on Watch decisions. Use it when a person asked the AI to redo the work and it needs the original document (PDF, image, or audio) again.

  • Flags an already-checked output for human review in Tracira, for example when a customer reports a problem with an AI reply. The output shows up on the Review page again and your notifications go out.

  • Retrieves a single output from Tracira by ID, including verdict, explanation, and human decision.

  • Fetches the current AI instructions (system prompt) stored in Tracira for a project and task. When nothing is stored yet, the Starter Instructions are saved as version 1 and returned - so a scenario that ships with Starter Instructions works with no setup. Leave Starter Instructions empty only if you have already written the instructions on the AI instructions page in Tracira. Use the returned Instructions as the system message of your AI step.

  • Performs an arbitrary authorized API call.

  • Sends an AI output to Tracira and checks it against your rules. With 'Wait for the verdict' it returns the status, verdict (pass, fail, or unsure), confidence score, and explanation.

  • Records a person's decision on an output when they decided outside Tracira, for example in your own form or chat tool: approve, reject, edit, or take over. Watch decisions and any Callback URL fire exactly as if they had decided on the Review page.

  • Adds a suggested change to the AI instructions in Tracira, for feedback your scenario receives outside Tracira (for example a correction typed in your own chat app). It joins the same suggested update as Teach the AI, which a manager reviews and accepts in Tracira. Nothing changes until they do.

  • Updates the details around an output already in Tracira: metadata you only learned afterwards (a ticket number, a category), the Session/Subject/Actor IDs, and the label on each attached file so a reviewer can tell several photos or documents apart. The output itself, its verdict and the human decision are never changed.

  • Saves a new version of the AI instructions in Tracira and makes it active. Use it after a reviewer sends a draft back with a comment: have an AI step rewrite the instructions to follow the feedback, then save the result here. The next Get instructions call returns the improved version.

  • Uploads a binary file (PDF, image, or audio) directly to Tracira storage and returns a key. Use this before Send an output for files too large to send inline (over ~3 MB). Map the returned Attachment Key into an attachment on Send an output (source 'Tracira upload'), and say what the file is there, in the attachment's What is it? field: a file uploaded here does not belong to an output yet, so it cannot be named yet.

  • Removes a change you suggested with Suggest an instructions change, by its reference. Use it when the change was settled in your own app (applied there or turned down), so the manager is not asked to decide it twice. Withdrawn is false when the manager already accepted or dismissed it in Tracira.

Select a trigger and an action

Create workflow (opens in a new tab)

Triggers and actions

Explore what you can automate between Tracira and Tally.

Download a file
Downloads a file stored on a Tracira output and returns it as binary data. Map the File URL or File Key from an attachment on Watch decisions. Use it when a person asked the AI to redo the work and it needs the original document (PDF, image, or audio) again.
Action
Flag an output
Flags an already-checked output for human review in Tracira, for example when a customer reports a problem with an AI reply. The output shows up on the Review page again and your notifications go out.
Action
Get an output
Retrieves a single output from Tracira by ID, including verdict, explanation, and human decision.
Action
Get instructions
Fetches the current AI instructions (system prompt) stored in Tracira for a project and task. When nothing is stored yet, the Starter Instructions are saved as version 1 and returned - so a scenario that ships with Starter Instructions works with no setup. Leave Starter Instructions empty only if you have already written the instructions on the AI instructions page in Tracira. Use the returned Instructions as the system message of your AI step.
Action
List Fields Definitions
Lists all field definitions for the selected form.
Search
Make an API call
Performs an arbitrary authorized API call.
Action
Search outputs
Returns a filtered list of outputs from Tracira. Filter by status, project, task name, or date range.
Search
Send an output
Sends an AI output to Tracira and checks it against your rules. With 'Wait for the verdict' it returns the status, verdict (pass, fail, or unsure), confidence score, and explanation.
Action
Set a decision
Records a person's decision on an output when they decided outside Tracira, for example in your own form or chat tool: approve, reject, edit, or take over. Watch decisions and any Callback URL fire exactly as if they had decided on the Review page.
Action
Load more

Support

FAQs

Everything you need to know about Make. Can't find the answer you're looking for? Please chat to our team.

Start by registering for Make. As a new Make user, you’ll automatically be assigned a free account. Once your account is set up, you can start creating your first scenarios. We recommend starting with our step-by-step video tutorial.

A scenario represents a workflow or a project of your own creation, and it is made up of a series of modules that automate apps and services. Creating a scenario allows you to transfer and transform data between apps and services via these modules to automate anything and improve the way you work.

Modules are the main building blocks of automation in Make. Modules represent actions that Make performs with an app, like creating, updating, or deleting data.

Mapping links the modules in your scenario. When you map an item, you connect the data retrieved by one module to another module to perform the desired action. For example, you can map email addresses and subject lines to create a spreadsheet using this data.