> ## Documentation Index
> Fetch the complete documentation index at: https://docs.kleep.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Fit Balance

> Rate how each product runs on a five-point fit scale

**Fit Balance** is the product-centric workflow: instead of mapping personas to sizes, you rate how each product fits overall. It's the quickest way to tell Kleep that a product — or a whole product line — runs small or large.

## The fit scale

Each product is rated on five values:

| Rating         | Meaning                              |
| -------------- | ------------------------------------ |
| **Very Small** | Runs significantly small — size up   |
| **Small**      | Runs slightly small — half size up   |
| **Normal**     | True to size                         |
| **Large**      | Runs slightly large — half size down |
| **Very Large** | Runs significantly large — size down |

## Rating products

<Steps>
  <Step title="Browse the catalog">
    Open **Fit Feedbacks → Fit Balance**. Use search, filters (category, brand, size system), and grid/table view to navigate.
  </Step>

  <Step title="Select a fit value">
    Each product shows the five fit values. Click the one that matches how the product runs. Click again to clear it.
  </Step>

  <Step title="Rate in bulk (optional)">
    Select multiple products with the checkboxes, then apply a single fit value to all of them at once — ideal for a collection that consistently runs small or large.
  </Step>

  <Step title="Save">
    Press <kbd>Cmd</kbd>/<kbd>Ctrl</kbd> + <kbd>S</kbd> or click **Save**.
  </Step>
</Steps>

## Feedback sources

A product's fit rating can come from three sources:

| Source          | Color  | Description                                    |
| --------------- | ------ | ---------------------------------------------- |
| **You** (user)  | Indigo | Ratings you set in the dashboard               |
| **Integration** | Purple | Fit data imported from a third-party source    |
| **Kleep**       | —      | Kleep's own AI assessment of the product's fit |

Sources are prioritized: **your rating > integration data > Kleep's assessment**. Integration values are shown read-only (in a muted color) so you can use them as context — setting your own rating overrides them without deleting them.

<Note>
  Kleep-sourced ratings are currently not displayed in the dashboard. They still exist and are used in the priority chain, but only your own ratings and integration data are visible for now.
</Note>

## Automatic CMS sync (integration source)

If your product catalog already contains fit information (e.g. "runs small", "true to size"), Kleep can pick it up automatically — no manual rating needed. This is where the **Integration** source comes from.

### How it works

* During each scheduled catalog import, Kleep reads a fit hint from your CMS for every product.
* Raw values are matched against a mapping we configure for you, and translated to the five-point fit scale (`very_small`, `small`, `normal`, `large`, `very_large`).
* Matching is case-insensitive and whitespace-tolerant. It can be exact, or substring-based for free-form text (e.g. a sizing-advice sentence that contains "runs small").
* Products with an empty or unrecognized value are simply skipped — they carry no signal, they are not treated as "Normal".
* Synced values appear in the dashboard as read-only **Integration** ratings. Your own dashboard ratings always take priority over them.
* The sync runs at your regular import frequency, so updating the value in your CMS updates the fit signal on the next import.

### Shopify

On Shopify, the fit hint is read from a **product metafield**. To enable the sync, communicate the following to your Kleep contact:

1. **Metafield namespace and key** — e.g. `custom.fit_advice`.
2. **The exhaustive list of values** the metafield can take, and what each one means in terms of fit.

That's all — we configure the mapping on our side.

**Example.** A brand stores sizing advice in the metafield `marketing.size_details_info` and communicates these values:

| Metafield value                        | Meaning          | Mapped to    |
| -------------------------------------- | ---------------- | ------------ |
| `True to size`                         | Fits as labelled | `normal`     |
| `Runs small`                           | Size up          | `small`      |
| `Runs large`                           | Size down        | `large`      |
| `Sizing up 2 sizes`                    | Size up a lot    | `very_small` |
| `Runs large, choose two sizes smaller` | Size down a lot  | `very_large` |

A product whose metafield contains `"Runs small"` will then automatically show a **Small** integration rating in the dashboard after the next import.

<Tip>
  The metafield values don't need to follow any specific format — plain sentences in any language work, as long as the list of possible values is stable and communicated to us.
</Tip>

### Other CMS platforms

Automatic fit sync for other platforms (product feeds, PrestaShop, SFCC, etc.) is handled case by case. [Contact us](mailto:support@kleep.ai) to set it up for your integration.

## Reviewing your work

* **Modified filter** — show only products where you have set a rating.
* **Show only selected** — focus on your current bulk selection.
* Sort by name or product ID to work through the catalog systematically.

<Tip>
  Only rate products that deviate from true-to-size. Leaving a product unrated is fine — a deliberate <strong>Normal</strong> rating is a stronger signal than no rating, so use it when you have confirmed the product is true to size.
</Tip>