Data Sources (aka the rules engine) are the heart of the Rebuy Engine. The rules engine leverages AI and conditional rules (ie. IF this THEN that) rules to make it possible to tailor merchandising widgets across the entire customer journey from homepage to cart, checkout, post-purchase, and beyond. They enable targeted product selection, exclusions, and tailored collections to meet specific business needs such as promoting best sellers or excluding certain product types.

Data Sources rules are made up of two sections, ‘IF’ and ‘THEN’, which is referred to as “RETURN” in Rebuy Data Sources. Each section can have multiple rules leveraging different objects to power personalization. For more information on the various objects in each section and how they can be used, read the Anatomy of Data Sources section of this document.

Data source rules in Rebuy are made up of IF and THEN (Return) conditions. When your IF section contains multiple objects, Rebuy combines them using either AND or OR logic — and the difference has a significant impact on results.

Rebuy evaluates data source rules sequentially, from top to bottom. The order of your rules directly impacts which products get returned.

Data source filters in Rebuy control which products are excluded from your widget's output before any rules are evaluated.

You can filter out specific products by clicking Add Exclusion Filter. Learn more about exclusion filters in this help document.

To exclude specific products from appearing in certain widgets:

This ensures that products can be selectively excluded or included depending on the data source rules applied to the widget.

Within a Data Source there are four different parts that are combined to create the rule. Those four parts are:

The tables below are grouped by object, and outline the corresponding attribute, operator and values. The tables will look something like the following:

Below each table are keys—aka tips and tricks—for each object group. While we tried to cover as many use cases and as much additional helpful information as possible, this is not an exhaustive list.

With that in mind, let’s dive in!

This section of the Data Source is where we set what conditions need to be met in order for the Return portion to be presented or shown. 🔑 For rules at the top of your list, you can enable the "Exit if Matched" option to stop evaluating further rules once a match is found. This is particularly useful when you have specific product replacements or exclusions that should take precedence over general rules.

Use order rules to target customers based on the tags applied to their previous orders.

Use product rules to target or exclude specific products, tags, titles, vendors, types, or handles in your data source IF conditions.

🔑 Remember, when using ‘Equals’ or ‘Does Not Equal’ you need to type the exact word/phrase you want used. With ‘Contains’ or ‘Does Not Contain’, you can type keywords or partial phrases. ‘Equals’ requires an exact match where ‘Contains’ only looks for the words or phrases regardless of if there are other words around it.

Use collection rules to target or exclude products based on the Shopify collections they belong to.

Use customer rules to target or exclude customers based on their tags, total order count, or lifetime spend.

Use cart rules to trigger your data source based on what's currently in the customer's cart, including specific products, subtotal, item count, or line count.

Use date rules to control when your data source is active, based on a specific date and time.

Use URL rules to target your data source to specific pages or traffic sources based on the current page URL or UTM parameters.

Use geolocation rules to target customers based on their physical location, including country, region, city, or postal code. Note that geolocation accuracy depends on IP data quality and is best used as a general targeting guide rather than a precise filter.

Use Klaviyo segment rules to target customers based on the segments defined in your Klaviyo account. Rebuy matches customer emails from Klaviyo segment profiles to Rebuy customer records, keep your segments up to date in Klaviyo to ensure rules reflect current membership.

This section of the Data Source is where we set what will be shown if the conditions of the If portion are met.

Use product rules to target or exclude specific products, tags, titles, vendors, types, or handles in your data source IF conditions.

Here are two different use cases for Product Price to get your wheels turning:

Use product settings to control quantity and discount behavior for products returned by this rule.

Use widget language settings to override the default text displayed on your widget, including titles, button labels, and cart state messaging, on a per-rule basis.

Use these Return options to create custom language on the widget in a variety of places. You can also use metadata variables to personalize widget language dynamically, pulling in real values like the customer's first name or the featured product title at the time the widget renders. For example, {{ metadata.shopify_customer.first_name }} would display as "Molly" for a customer named Molly, producing a title like "Faves Picked For You, Molly!"

The available metadata variables are:

For more detail, see How to Configure Custom Languages Using Data Sources.

Use widget settings to apply a visual discount label to products returned by this rule.

Use custom data to pass additional key-value pairs to your widget for advanced personalization or developer-driven customization.

Yes, you can duplicate a data source. On the Data Sources page, find the data source you want to duplicate, click the three dots next to it, and select Duplicate Ruleset.

No, data sources cannot be exported to another Shopify store. Data sources are tied to specific product IDs within your current store, which makes them incompatible with other store environments.

If a data source's rules aren't met, the widget will not display any content to the customer. This happens automatically — no additional configuration is needed. Note that the widget remains installed on the page; it simply won't show any products until the conditions are satisfied.

Rebuy recommends keeping data source rules to 30 or fewer per data source. You can add up to 50 rules, but exceeding this limit can cause slow widget load times and may result in failed saves. If you're hitting the limit, see the FAQ below on scaling beyond 50 rules.

If you need more than 50 rules, there are three approaches:

To avoid conflicting data source rules, check for overlapping criteria, for example, including and excluding the same product or tag simultaneously. Use logical rule sequencing and position your most specific rules at the top of the list. Broad or catch-all rules (such as "if anything") should always sit at the bottom, since Rebuy evaluates rules top to bottom and a catch-all will override everything below it.

To test a data source, use the Preview tool to verify that the expected products or collections are being returned based on your rules. Manually simulate the conditions your rules target, for example, add the relevant products to a test cart or visit the relevant page and confirm the widget displays the correct output before going live.

Use descriptive, specific names that reflect the data source's purpose and context — for example, "Product Page Cross-sell by Tag — Apparel" rather than "Data Source 12." Clear naming keeps your configurations organized and makes it easier to identify the right data source when building or editing widgets.

Only the Gift With Purchase (GWP) Widget and the Tiered Progress Bar (TPB) support automatically adding a product to the cart. All other Rebuy widget types display product recommendations for the customer to select, they do not add products to the cart on their own.