Introduction
When you first connect a product source to DataHub, your raw source data arrives with inconsistencies. Availability values might be spelled differently across rows, descriptions might have leading or trailing whitespace, and brand names might use mixed capitalisation. A feed rule library is a pre-built set of rules that fixes these common issues automatically at export time.
This article walks you through deploying your first rule library. By the end, you will have a normalised data baseline ready for further customisation or export.
Prerequisites
Before you start, you need:
- At least one connected or imported product source. If you do not yet have a source connected, follow connecting-a-shopify-store-for-the-first-sync or from-csv-to-your-first-google-shopping-listing depending on your data origin.
- Editor-level access or above in your DataHub account. Check your account permissions if you are unsure.
Before state: Raw, unstandardised source data
Your product source currently contains data exactly as it arrived. Common issues include:
- Availability fields with values like 'in stock', 'In Stock', 'available', and 'Available' used interchangeably.
- Text fields with unnecessary spaces at the start or end.
- Price and numeric fields formatted inconsistently.
- Missing or empty values that could be standardised.
No rules are applied yet, so every export carries these inconsistencies forward to your destination platforms.
Step 1: Open the Rules area in DataHub
From your DataHub dashboard, locate the main navigation menu on the left. Click on the Rules section. You will see a Rules overview page with options to manage, create, or deploy rules.
Before/after: You are now in the Rules management area where all rule libraries and custom rules are managed.
Step 2: Browse available rule libraries
In the Rules area, look for a section showing available rule libraries or templates. DataHub provides two main types:
- Data normalisation library: Handles whitespace trimming, standardises availability values, normalises boolean fields, and formats numeric data.
- General best-practice library: Covers additional improvements such as ensuring required fields are populated and applying common platform-specific fixes.
Before/after: You can now see which pre-built libraries are available for your account and understand what each one does.
Step 3: Preview the data normalisation library
Before deploying any library, preview what it will change. Click on the data normalisation library to open its details. Look for a preview or test option.
DataHub will show you a sample of products from your connected source and display the before/after state for each field the rules will affect. For example:
- Original availability value: 'in stock' becomes 'in stock' (standardised format).
- Original title: ' Widget A ' becomes 'Widget A' (whitespace trimmed).
- Original price: '10.5' becomes '10.50' (consistent decimal places).
Review the preview to confirm the changes match your expectations.
Before/after: You have validated that the rule library will improve your data without introducing errors.
Step 4: Deploy the library with one click
Once you are satisfied with the preview, click the deploy button. DataHub will apply this rule library to your account. The deployment is immediate and requires no additional configuration.
Before/after: The rule library is now active. Every future export will apply these rules to your product data.
Step 5: Review the deployed rules
After deployment, return to the Rules area. You will see a new entry showing the deployed library with a timestamp and status (typically 'Active'). Click on this entry to expand it and see the individual rules it created.
You will see a list of rules such as:
- Trim whitespace from text fields.
- Standardise availability to predefined values.
- Format prices to two decimal places.
- Remove null values from required fields.
Each rule is now active and can be inspected, disabled, or extended with additional custom rules if needed.
Before/after: You can now see exactly which rules are running on your data and understand the normalisation being applied.
Step 6: Run a test export to verify the effect
To confirm the rules are working, trigger a test export to one of your connected destinations. After the export completes, review the export diff report. This report shows which fields changed compared to the previous export.
You should see changes in fields like:
- availability: Values now consistent and standardised.
- title, description: Whitespace removed.
- price: Formatted uniformly.
The diff confirms that the rules are actively transforming your data at export time.
Before/after: You have verified that the deployed rules are functioning and improving your data quality in real exports.
After state: Deployed rule library actively normalising data
Your product data is now normalised at every export. The rule library is deployed and visible in your Rules section as an active, inspectable set of rules. Your data baseline is clean and consistent, ready for platform export or further customisation with hand-authored rules.
Why start with a rule library?
Building rules from scratch is time-consuming and error-prone. A pre-built rule library encodes fixes for the most common data issues: whitespace trimming, availability standardisation, numeric formatting, and boolean normalisation. Deploying a library gives you an immediate baseline of data quality improvements without manual rule authoring. You can then add custom rules on top if your specific use case requires them.
Checkpoint: What you should see
Once you have completed this walkthrough, verify the following:
- The Rules section shows a deployment record for the data normalisation library with a status of 'Active' or 'Deployed'.
- Clicking into the deployment shows the list of individual rules it created, all marked as active.
- Your next export's diff report shows changes in multiple fields (availability, title, price, description, or similar), confirming the rules are being applied.
- No errors or warnings appear in the deployment status.
If all these elements are present, your first rule library is deployed and working correctly. You can now export with confidence that your data is normalised, or proceed to create additional custom rules to address specific requirements.