The best next-product automation starts with a customer’s most recent purchase, not with a generic “You may also like” block. A WooCommerce store can use order data to identify the product family, use WPSOLR to retrieve relevant candidates, and then apply business rules before presenting the recommendation.
Define the next-purchase decision
Before configuring search or email automation, decide what “next” means for the store. Common objectives include:
- Consumables: Recommend a refill or replacement after the expected usage period.
- Accessories: Suggest products that extend or improve the original purchase.
- Upgrades: Present a higher-capacity or more advanced version when the customer is likely to need one.
- Complements: Recommend products that are commonly used with the purchased item.
- Replenishment: Remind customers to reorder products with a predictable consumption cycle.
These objectives require different timing and ranking rules. A customer who bought a coffee machine should see compatible accessories soon after delivery, while a customer who bought coffee beans may need a replenishment message several weeks later.
Use the order as the recommendation context
At the time an order is completed, capture the purchased product IDs, categories, attributes, variations, and quantities. The order should provide context for a later search rather than becoming the recommendation itself.
For example, an agency might define this rule:
Purchased category: running shoes<br>Candidate categories: socks, insoles, running lights<br>Exclude: the purchased product and out-of-stock items<br>Price range: 20% below to 80% above the original product price<br>Delivery delay: 7 days after the order is completed
The exact values should be configurable by product category. A single global rule usually produces poor recommendations because the purchase cycle for a subscription-like product is different from the purchase cycle for durable equipment.
Use WPSOLR for candidate retrieval
WPSOLR is useful when the store has a large catalog and recommendation candidates need to respect the same indexed product data as site search. Index product titles, descriptions, taxonomies, attributes, brands, compatibility fields, stock status, and other fields that help establish relevance.
A practical workflow is:
- Read the customer’s recent order and identify the purchased product or product group.
- Build a candidate query from the product’s categories, attributes, brand, compatibility terms, or curated relationships.
- Send that query through the store’s WPSOLR-powered search configuration.
- Apply exclusions for the purchased product, unavailable products, restricted products, and products already owned when that data is available.
- Rank the remaining candidates using commercial rules such as margin, popularity, inventory, price, or merchandising priority.
- Render the final products in the relevant channel, such as a post-purchase page, account dashboard, email, or remarketing audience.
WPSOLR should be treated as the candidate retrieval and relevance layer. The purchase-history logic, timing, consent checks, and final eligibility rules should remain in the automation or application layer.
Build useful product relationships
Search relevance alone does not guarantee a useful next-product recommendation. Agencies should create structured relationships wherever possible. Product attributes such as compatible_with, replacement_interval, use_case, and product_family are more reliable than trying to infer every relationship from free-text descriptions.
For example, a replacement filter can contain an indexed compatibility value matching the original appliance model. The automation can then query for products with that compatibility value instead of recommending every product in the same broad category.
Where the catalog is too small or the relationship is highly commercial, use curated cross-sells or product bundles. WPSOLR can still help find fallback candidates when the curated products are unavailable.
Choose the right delivery moment
The recommendation should arrive when it is useful, not immediately after payment by default. A common sequence is:
- Order confirmation: Show only low-friction accessories that do not distract from order information.
- After fulfillment: Recommend setup items, compatible accessories, or educational products.
- After the expected usage period: Recommend refills or replacements.
- After engagement: If the customer views or searches for related products, increase their priority in the next message.
Use the order status and fulfillment date rather than relying only on the order creation date. A delayed shipment should not trigger a setup-accessory message before the original product arrives.
Example automation architecture
A maintainable implementation can separate the process into four services or responsibilities:
- Event capture: Listen for a suitable WooCommerce order event and store a minimal recommendation record containing the customer reference, product IDs, and scheduled evaluation date.
- Eligibility evaluation: Confirm that the order is paid, the customer has marketing consent where required, the recommendation has not already been sent, and the product is still eligible.
- Candidate retrieval: Query the WPSOLR-powered catalog using the product’s structured relationships and apply stock, visibility, category, and exclusion filters.
- Delivery and measurement: Send or display the recommendation, record impressions and clicks, and attribute subsequent purchases without storing unnecessary personal data.
Keep recommendation records idempotent. If an order hook is retried, it should not create duplicate messages or duplicate scheduled jobs. Use the WooCommerce order ID and recommendation rule as a unique key where appropriate.
Prevent poor recommendations
Several safeguards have a direct effect on conversion quality:
- Exclude the exact product purchased unless a repeat purchase is the intended outcome.
- Exclude products that are hidden, backordered, discontinued, or unavailable to the customer’s shipping region.
- Do not recommend an accessory that is already included in the purchased bundle.
- Respect catalog visibility, customer role restrictions, price rules, and minimum advertised pricing requirements.
- Limit the number of recommendations so the customer can understand the choice quickly.
- Use a fallback category or curated list when the WPSOLR query returns too few candidates.
- Do not expose sensitive purchase history in shared devices, public account areas, or messages without appropriate authentication.
Measure incremental revenue
Track more than clicks. Useful events include recommendation eligibility, candidate count, impression, click, add to cart, purchase, revenue, margin, and time from original order to recommended order.
Run a control group that receives the normal post-purchase experience. Compare incremental conversion and contribution margin against that group rather than attributing every later purchase to the recommendation. For replenishment products, also measure whether the automation changes reorder timing or simply captures a purchase that would have happened anyway.
For agencies managing multiple stores, expose the rule inputs as configuration: delay, candidate taxonomies, compatibility fields, exclusion windows, maximum products, fallback behavior, and attribution window. This keeps the implementation reusable while allowing each merchant to tune the automation to its catalog and buying cycle.