Team — a “Last Login” column for every member
The table on the Team page now shows when each member last signed in, with the date and time. If a user has never signed in, it shows “Never”.
Every entry comes here straight from the production release — with its date, module and type of change.
The table on the Team page now shows when each member last signed in, with the date and time. If a user has never signed in, it shows “Never”.
Background tasks — AI descriptions, AI translations, AI parameter extraction, AI image alt text and bulk actions — now carry a readable name.
Admins can now edit the first name, last name and e-mail address of other team members. The menu next to each team member has a new edit option with a pencil icon, which opens a dialogue for changing their details. You can only edit members of your own company, and each e-mail address must be unique.
You will find the System Overview page as the first item in the Parameters menu, before “Manage Parameters”. The item has a database icon.
The “Extract parameters from feeds” button goes through the stored feed data of all products and adds the parameters it finds to the list of unmapped parameters, or updates them. The data is processed in batches. It recognises parameters in three different formats, such as those used by the Heureka, Davorin, Origos and CTM feeds. The feature requires editor access.
For each product, variant and supplier, only one feed snapshot is now kept — always the latest, which is overwritten on the next sync. Older snapshots were removed once. The stored data therefore no longer grows without limit.
Supplier imports no longer delete and re-save product images on every sync when they have not changed, which is the case in about 90% of imports. The Grandus import also keeps the images created with AI in Photo Studio and places the feed images after them, so their order does not clash.
The new System Overview page shows all parameters in the system in one place: those discovered during imports, mapped parameters and values extracted from products. Each parameter name appears once, with the number of products, the list of suppliers, the mapping status (“Mapped” or “Unmapped”), the target parameter and sample values. You can filter by supplier and status and use full-text search. Results are split into pages, with the total count shown.
Merging products that have variants failed with an error while copying the variant prices. This is fixed, and such products now merge without an error.
You can undo a product merge: the source products are restored as drafts and the merged product is deleted together with all its related data. A paged merge history shows who merged what and when. Only admins can undo a merge.
On the Duplicates page, both “All Products” and “Duplicates” were highlighted as active in the menu. Now only “Duplicates” is highlighted.
The Products section has a new Duplicates page listing automatically detected groups of duplicates. Each card shows the match type, a colour-coded confidence badge, product thumbnails with the supplier label, and the actions “Compare & Merge” and “Dismiss”. You can filter suggestions by match type and status and search them. The “Scan for Duplicates” button starts a new scan.
The source product tabs on the compare and merge screen overflowed outside the card. The primary source selector and the product headers now sit in the header of the field selection table, so they scroll horizontally with it. A new “Product Detail” row at the bottom of the table opens each product in a new browser tab. The header and footer rows have a subtle grey background to set them apart.
By default, the product list now hides source products that have already been merged. Advanced Filters has a new merge status filter: all products, excluding merged sources, merged products only or source products only. Merged products carry a “Merged” badge, and source products a “Merged Source” badge with a link to the merged product. When you select two or more products, the “Merge Selected” bulk action becomes available. The detail of a merged product lists its source products, and the detail of a source product shows a banner saying it has been merged.
Compare two or more products side by side and choose, for each field, which source the value should come from. Merging creates a new combined product that is not tied to any supplier and holds the values you selected from the source products. Variants can be taken from one source or combined from all of them, with SKU conflicts resolved. Parameters can be combined from all sources or taken from one, you pick the images individually including the main image, and categories from all sources are combined automatically. The source products are archived, and the whole merge happens in one go — either completely or not at all.
The compare screen shows the products side by side in a table. For each field — such as title, description, GTIN, MPN, brand or weight — you choose the source of the value with a radio button. In the image gallery you select images with tick boxes and set the main one with a star. Toggles let you take variants from one source or combine them from all, and the same applies to parameters. Before merging, a confirmation shows a summary of how many source products are involved.
The Duplicates item in the Products menu shows an orange badge with the number of pending duplicate suggestions. You can see at a glance that suggestions are waiting for review and resolve them regularly.
Monokaido finds duplicate products across suppliers using five rules, each with its own confidence level: product GTIN (95%), variant GTIN (90%), MPN (80%), manufacturer SKU (75%) and variant SKU across different suppliers (60%). Overlapping groups are combined into one — if A and B match by GTIN and B and C by MPN, they form a single group of A, B and C. Results are saved as suggestions with the status pending, merged or dismissed. You start the scan manually.
A new import connects to the B2B web service of the supplier Vavrys (Benefit CZ), which supplies Orbea bikes and e-bikes and nvii running shoes. The whole catalogue loads in a single request — about 186 products, 228 variants and 202 images. Variants are grouped into products by their stock card and told apart by size and colour. Prices are in euros excluding VAT, categories have three levels (category, type, collection), and stock combines the warehouse and supplier quantities.
Product transfers to ERP no longer load the whole catalogue into memory at once; they process it in chunks of 1,000 products. Transfers of large catalogues, such as 100,000 products or more, therefore no longer crash for lack of memory.
Simple bulk actions — changing the status, category, manual category or brand, and setting or removing a parameter — now update the whole batch of products in one step instead of changing each product separately, so they finish faster.
Scheduled product and availability syncs now reserve a supplier in a single step, so the same supplier can no longer be started twice at once. If a sync cannot be handed over for processing, the supplier is marked as failed instead of staying stuck in the waiting state.
The regular checks that start scheduled product and availability syncs and scheduled bulk actions, as well as the automatic clean-up of stuck statuses, no longer start again until the previous check has finished. Previously they could overlap when a check took longer than the interval between runs.
After a bulk action finished, the products stayed selected because the bulk action window did not report that the job had completed. This is fixed: once a job completes, the selection is cleared as it should be.
The Trek availability import could fail when several imports ran at once and each tried to create the Trek warehouse. The warehouse is now created in a single step, as in the other importers, and is correctly assigned to the supplier.
We'll walk through your feeds, categories and parameters live.