Shopware 6 with a populated order history if you want to use the „Co-purchases" source. AI scoring needs an API key with the relevant provider. Scheduled analysis requires a running Shopware scheduler or worker.
Note: The plugin writes into native Shopware cross-selling. It adds no blocks of its own to the storefront and replaces no Shopware feature.
Installation
Install and activate the plugin, then clear the Shopware cache. The Cross-Selling area is then available in the admin.
Opening the plugin
You will find it in the admin under Settings, in the Cross-Selling area. From there you reach the three working areas: managing profiles, running an analysis and reviewing suggestions.
What the plugin does
The plugin suggests matching products for cross-selling. It evaluates order history and catalog, scores the pairs with AI if you want, and you decide which ones are accepted. Accepted suggestions end up in a native Shopware cross-selling block on the product page.
The process always has three steps:
Analyse - generates suggestions and does not change the sets in the shop yet.
Accept - adds the product to the cross-selling block.
Live in the shop - the block is visible to customers.
Important: An analysis on its own changes nothing in the shop. Only accepting writes assignments. Accepting is additive: the product is added to the existing block of the same name, never twice, and nothing is ever removed automatically.
The six sources
Suggestions come from these sources. Each can be enabled per profile and combined freely.
Co-purchases
Products frequently bought together, from order history. Needs purchases. Controlled by „Min. co-purchases" and „Min. confidence". Example: camera and memory card, because both were often in the same basket.
Same categories
Products in the same categories. Controlled by „Min. shared categories" (default 2). Example: two items that both sit in „Hiking" and „Backpacks".
Same properties
Matching property values, e.g. brand or series. Controlled by „Min. shared properties" (default 2).
Same manufacturer
Products from the same brand.
Category bestsellers (fallback)
Best-selling products of the category. Ensures coverage even without purchase history, but weighted lower than real matches. Needs maintained sales figures (product.sales).
Viewed together
Products that visitors looked at within the same visit. Needs no orders and works for new products too. Controlled by „Minimum shared views" and „Maximum suggestions per product". Recording has to be switched on in the settings. Example: tent and sleeping bag, because both were viewed in the same visit.
What a profile is
A profile is a complete preset and maps to its own cross-selling group in the shop. Each profile has its own sources, thresholds, scope, AI usage and auto-accept. The profile name is also the title of the block in the shop.
For example: one profile „Frequently bought together" based on co-purchases, and one profile „Matching accessories" based on shared properties. Each creates its own block on the product page.
Creating a profile
Under Settings, Cross-Selling, „Add profile" or „Edit". In the modal you set the name, the sources including their minimum values, the scope, AI and thresholds, and auto-accept.
Do not forget: Profile changes only take effect once you press „Save" at the bottom of the settings tab.
Sources in the profile
Two fields control how strictly the catalog-based sources match.
Min. shared categories Default 2
For „Same categories" only: this many categories must overlap.
Min. shared properties Default 2
For „Same properties" only: this many property values must match.
Thresholds and AI
Min. co-purchases Default 3
How often two products must have been bought together for the pair to count - an absolute number, not a percentage. Example: 3 = the pair appeared together in at least 3 orders. Higher = only robust pairs; lower = rare combinations too.
Min. confidence Default 0.1
Share of A-orders that also contained B (0 to 1). Example: 0.1 = B was present in at least 10% of orders with A. Higher = only strong „buyers of A also buy B" links.
AI scoring Default off
Scores each pair with AI, detects complement vs. similar item, gives a short rationale and a score (0 to 100). Needs an API key for the chosen provider.
Exclude substitutes Default on
Keeps only genuine complements and filters out substitutes (replacements). Example: camera and memory card stays, camera and another camera is dropped. Only works with AI scoring enabled.
Min. shared categories Default 2
For the „Same categories" source only: this many categories must overlap.
Min. shared properties Default 2
For the „Same properties" source only: this many property values must match.
Minimum shared views Default 3
For „Viewed together" only: how often two products must have been viewed together within the same visit.
Maximum suggestions per product Default 5
For „Viewed together" only: how many of the most frequently co-viewed products per item are used.
Scope and output
Time window Default 365 days
How far the order history for „Co-purchases" reaches back. Example: 365 = last year only; shorter emphasises current trends, longer uses more data.
Max. suggestions per product Default 8
Upper cap on good matches per source product. Example: 8 = at most 8 items in the block. On the storefront slider 4 to 8 usually works best.
Minimum per product Default 0
If the sources return fewer, the list is topped up with category bestsellers until the number is reached. Raises the maximum if needed. 0 = no top-up.
Active products only Default on
Ignores inactive or hidden products as source and as suggestions.
Products without cross-selling only Default off
Only suggests for products that have no block yet - good for filling gaps without touching existing ones.
Update existing Default off
Allows adding to already maintained assignments instead of skipping them.
Shop / sales channel Default all
Optionally ties the profile to one shop: only products visible there count and suggestions are assigned to that channel. „All channels" = the choice from the analysis dialog.
Exclude manufacturers Default none
Selected manufacturers are never suggested as cross-selling.
Auto-accept
Auto-accept is set per profile. One profile can roll out automatically while another stays manual.
Auto-accept Default off
Accepts suggestions automatically without manual review - set per profile. One profile can roll out automatically while another stays manual.
Above score Default 80
Only suggestions with at least this AI score (0 to 100) are auto-accepted. Recommendation: since cross-selling is customer-facing, set it high (80+).
AI complements only Default on
Auto-accept only for pairs the AI classified as genuine complements.
Activate block automatically Default on
Activates the generated cross-selling block directly in the shop.
Mutual assignment Default on
Assigns A to B and B to A, instead of one direction only.
Write to variants Default off
Also writes assignments to product variants, not just the main product.
Setting up AI scoring
Supported providers are OpenAI (ChatGPT), Anthropic (Claude), Mistral and Gemini. Enter an API key, pick a model and verify it with „Test connection".
The AI scores every pair, classifies it as a complement or a similar item and returns a short rationale plus a score from 0 to 100. With „Exclude substitutes" only genuine complements make it into the suggestions.
Example: A camera and a memory card is a complement and stays. A camera and another camera is a substitute and is filtered out.
Running an analysis
Manually via the „Start analysis" button, where you choose the profile or all active profiles and the sales channel. Or from the console:
Pairs that have already been decided, meaning accepted or rejected, are not suggested again. After the analysis the view jumps to the new open suggestions automatically.
Reviewing and accepting suggestions
In the review pool you filter by status, min. score, sales channel and group. Accept individual suggestions with „Accept" or dismiss them with „Rejected". „Accept all" respects the active filters, so it can act on a single group only.
The badges on each suggestion show source, group and score, plus an „auto" badge for automatically accepted ones. „Remove" takes an assignment back out of the shop.
Automation and cron
Under Settings, „Automation", you enable scheduled analysis and pick the interval. This requires a running Shopware scheduler or worker. The automation status shows the last and next run, the task status and how many assignments were accepted automatically.
Note: An analysis alone does not update the sets, not even via cron. Only accepting does. With auto-accept enabled, cron maintains the sets by itself.
Setting up viewed together
This source evaluates which products visitors looked at within the same visit. It needs neither orders nor well-maintained categories and therefore produces suggestions for new products as well.
Under Settings, Viewed together, switch on Recording active. Without it no data accumulates. The status below shows recorded pairs, counted views, products with data and the last entry.
Products considered per visit Default 6
How many recently viewed products are paired with the current one. Higher values yield more data but looser relationships.
Count variants towards the main product Default on
Views of individual variants roll up to the main product instead of being scattered.
Retention Default 365 days
Older pairs are removed during the scheduled run, as are pairs of deleted products.
Recording is kept separate per sales channel, detects bots and writes a single query per page view. Counters from a previously used table are carried over on update, so the source does not start from zero.
Data protection: Only product pairs with a counter are stored. No personal data and no visitor profiles are created.
Then enable the source in the profile and run an analysis. Expect a few days of lead time until enough views have accumulated.
Cleanup
„Cleanup" removes open suggestions that no longer belong to any current profile, for example after renaming or deleting a profile. Accepted and rejected suggestions are left untouched.
Tips from practice
Cross-selling is customer-facing. For auto-accept, set a high threshold, for example from score 80, and ideally allow AI complements only.
The bestseller fallback needs product.sales to be populated. Shopware counts sold quantities automatically, but a brand-new shop has no bestsellers yet.
Give the fallback its own profile. It then sits in its own group and can be controlled separately.
Troubleshooting
The analysis finds nothing
The message names the reason per source. Common causes are a value set too high for „Min. shared categories", no shared manufacturer, or the „products without cross-selling only" option removing all candidates.
No suggestions despite many orders
Check „Min. co-purchases" and „Min. confidence". Both filter before the AI scores anything. The time window may also be too short.
Viewed together returns nothing
First check whether recording is active under Settings, Viewed together. Without it no data accumulates. If recording was switched on only recently, the counted views may not yet reach the minimum set in the profile.
Scheduled analysis does not run
Without a running scheduler or worker the task never starts. The automation status shows the last and next run.
Support
If you have questions about setup, connecting an AI provider or cross-selling strategy, we are happy to help. Get in touch via the contact page or write to info@stoneandwater.online.
Through the store or directly from us.
Two routes, the same feature set - pick whichever suits your setup.
Directly from Stone & Water
Available right now
Monthly9,99 €
Half-yearly49,99 €17 % cheaper
Best valueYearly89,99 €25 % cheaper
All prices net, plus VAT.
You buy the licence directly from us and get an invoice from us. Install by upload in the admin without SSH, updates through our licensing platform, support straight from the developer.
The plugin will be listed in the official Shopware Community Store. Installation and automatic updates then run through the plugin manager, billing through your Shopware account. We will link it here as soon as it is live.
Better cross-selling?
Back to Cross-Selling.
Find all details on the product page. We are happy to help with AI setup and strategy.