> For the complete documentation index, see [llms.txt](https://examind.gitbook.io/v1/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://examind.gitbook.io/v1/feedback-machines/use.md).

# Use a Feedback Machine

How students submit work to a Feedback Machine for iterative feedback — and how instructors can use the same machine themselves.

A Feedback Machine is a **shared tool**: you and your students use the *same* machine. Students submit their work and get immediate, criteria-aligned feedback — then revise and resubmit as many times as they like — while you can run submissions through that machine yourself and always see the full results. The more feedback-and-revision cycles a student goes through, the more their work tends to improve, and you can see all of a student's submissions to follow their progress.

Once a machine is published and [set up in your Canvas course](/v1/feedback-machines/get-started/using-canvas.md) (or shared via [direct login](/v1/feedback-machines/get-started/without-canvas.md)), it's ready to use.

{% hint style="info" %}
Want to grade a whole class's submissions at once instead? See [Bulk Grading Assist](/v1/feedback-machines/bulk-grading-assist.md).
{% endhint %}

## How students submit

Students open the machine (from your Canvas course, or a direct link) and submit their work in the format you configured:

* **Text** — pasted into the submission box (typing directly into it isn't supported), with a live character count.
* **`.docx` or `.pdf`** — uploaded. What the machine can take into account differs by format — `.pdf` carries images and visual design, `.docx` carries exact formatting details; see [What a Feedback Machine Can See in Submissions](/v1/feedback-machines/build-a-feedback-machine/what-a-feedback-machine-can-see.md). If you allowed **DOCX or PDF**, a single submission can even mix the two formats.
* **Multiple files** — if you set **Files Per Submission** above one; all files in a submission are evaluated together.

Each file is checked the moment the student chooses it — the format and size (up to 30 MB), and that the file actually opens — so a bad upload is caught before submitting rather than after. A file that can't be read is marked, and underneath it the student gets the likely cause and the one step that fixes it:

* A `.doc`, or another Office format, renamed to `.docx` — or a Word file named `.pdf`. *Save or export a real copy in the format the machine asks for.*
* A password-protected document. *Remove the password, or print it to a new PDF.*
* A download or export that didn't finish, leaving the file truncated or page-less. *Export it again and check the pages before uploading.*
* A 0-byte file — usually a OneDrive, iCloud, or Google Drive placeholder that hasn't downloaded to the device yet. *Open it on the device first.*

Anything the check can't place is stated as our problem, not the student's, with the support address to send the file to. **Submit Draft** gives way to **Choose a different file** until every file in the submission can be read.

{% hint style="info" %}
For group assignments, students can check **"Submitting only part of the assignment?"** so the feedback adapts to the portion they contributed.
{% endhint %}

## What students see

After submitting, a student's work opens in a side-by-side workspace — their **Submission** alongside whichever view applies to that submission:

* **Feedback** — narrative feedback for improving the next draft: **What you did well**, **Areas for development**, and a prioritized list of **Things you can do to improve**. Score estimates appear only if you turned them on.
* **Results** — the final grade: the overall score, the key takeaways to carry forward, and each rubric part with its score and summary. This appears once you've approved the submission (or immediately, if you configured it that way).

Students select **New Draft** to revise and resubmit.

{% hint style="info" %}
Which of these a student sees, and when, is entirely up to you — see [What Students See](/v1/feedback-machines/use/what-students-see.md) for the settings, and [Review, Adjust & Approve Results](/v1/feedback-machines/review-adjust-and-approve.md) for how approval releases a final grade.
{% endhint %}

## Use it yourself as an instructor

You can run a submission through any published machine yourself — choose **Use** on the machine, just as a student would, and submit a sample. This is the best way to see exactly what your students will experience before you share it.

{% hint style="success" %}
**You always see the full results — even what's hidden from students.** When you view an evaluation (your own submission or a student's), you see every view and the complete criterion-level detail, no matter what you've made visible to students. So you can keep scores or the detailed breakdown hidden to keep students focused on the feedback, and still review everything yourself. Where a view is hidden from students, a chip tells you so.
{% endhint %}

## Follow your students' progress

Because a Feedback Machine is shared, you can see how your class is using it — both submission by submission and at a glance.

### Submissions

There are two ways to see submissions:

* **Across the whole class** — from your class page, select **View all submissions** to see every student alongside **all** of their drafts from **every** machine in the class, combined into one list.
* **For a single machine** — open that machine's **⋮** menu and choose **Submissions** to see just its submissions.

Either way, each student's drafts are listed newest first with score and status, and you can open any draft to read the full feedback and criteria breakdown — including anything hidden from the student. Because students can resubmit as often as they like, you see the complete history of their revisions, not just the latest one.

To grade and approve rather than just browse, use the machine's [Results page](/v1/feedback-machines/review-adjust-and-approve.md).

You can flip how names read between **First Last** and **Last, First**; the list sorts to match, so it always reads the way it's ordered. (This choice resets each visit rather than being saved.)

### Insights

Open a machine's **⋮** menu and choose **Insights** for a summary of how the class is engaging with it:

* **Active Students** — students who have submitted at least once.
* **Total Submissions** — across everyone using the machine.
* **Avg. Iterations** — average submissions per active student.
* **Avg. Improvement** — the average score gain from a student's first submission to their best one.
* **Avg. Effort** — how much students revised between drafts, based on how much their text changed.

It also charts submission activity over time and the distribution of submissions, so you can see engagement and improvement at a class level — not just per student.

## Tell students how to access it

Copy and adapt one of these for your assignment instructions, a course announcement, or a message to students.

{% tabs %}
{% tab title="Canvas" %}

> **Get Instant, Actionable Assignment Feedback** — We've partnered with Feedback Machines, an AI platform that gives you immediate, constructive feedback on your assignments, helping you refine and improve your work with every submission. To access the instant feedback, click the **Feedback Machines** link in the left menu, then click **\[use]** next to the assignment you're working on.
> {% endtab %}

{% tab title="Direct login" %}

> **Get Instant, Actionable Assignment Feedback** — We've partnered with Feedback Machines, an AI platform that gives you immediate, constructive feedback on your assignments, helping you refine and improve your work with every submission.
>
> **How to use:** Go to [feedbackmachines.com](https://feedbackmachines.com) and click **\[Sign In]**, then enter your university email. The site will email you a link to complete the login. If you don't see the email, check your spam/junk folder. Once logged in, click **\[use]** next to this assignment and follow the on-screen instructions.
> {% endtab %}
> {% endtabs %}