Frequently Bought Together PRO
“Customers who bought this also bought” companions from real purchase co-occurrence.
Category: Merchandising · Enable at: Ambikly → Add-ons → Frequently Bought Together
What it does
A scheduled job builds a pairs table from your own order history, keeps pairs above a minimum support and scores them by lift. Product pages render the top companions automatically, or you place them with a shortcode; a public REST route serves the same data. Rebuilds are non-blocking so the storefront never shows an empty gap.
Capabilities
- Rebuild on a custom N-day schedule with a rolling lookback window
- Minimum-support threshold so one-off coincidences don’t surface
- Score is overridable via the ambikly_recommendation_score filter
- Companions filtered to published products that are in stock
- Renders on product pages, via [ambikly_recommendations], and over REST
- Manual rebuild action for store managers
Turn it on
- Make sure Ambikly Pro is active and your license is activated under Ambikly → Settings → License.
- Go to Ambikly → Add-ons, find Frequently Bought Together and switch it on. There is nothing else to install.
- Open its settings panel and set the options listed below.
- Visit your storefront to confirm the new surface appears: [ambikly_recommendations] shortcode.
Settings
Open the add-on's settings panel from Ambikly → Add-ons → Frequently Bought Together.
- Number of recommendations
- Minimum orders for co-occurrence
- Rebuild every N days
- Heading
- Only consider orders from the last N days
Where it appears
On the storefront
- [ambikly_recommendations] shortcode
- Automatic block on product detail pages
In the admin
- Rebuild recommendations action
For developers
REST routes
GET /ambikly-pro/v1/recommendationsPOST /ambikly-pro/v1/recommendations/rebuild
Every route enforces the same capability checks as the rest of the API — see the REST API guide.
Core hooks it listens to
ambikly_pro_reco_rebuildambikly_product_detail_bodyambikly_recommendation_scorerest_api_init
The full list is in the hooks reference.
Questions
Does this use an AI model?
No. It is a market-basket co-occurrence query over your real orders — not a language model. The score filter is the hook for swapping in embedding-based similarity later.
Will out-of-stock products be recommended?
No. Only published, in-stock products are returned.
Works well with
Product Bundles · Product Badges
Turning it off
Switch the add-on off from Ambikly → Add-ons. Its settings and data stay in the database, so turning it back on restores exactly what you had. A lapsed license pauses it the same way — the store keeps selling on the free core, and nothing is deleted.
One license covers all 35 add-ons — $199 a year or $549 once for a single site. Nothing here is sold separately.
See pricing →