What's new in Ecombooster?
Follow the development of the platform. Here we publish new features, improvements and important updates.
More than one shipment per order
An order shipped in two rounds — part today, the rest when the backorder lands — is handled as two separate bookings. Each shipment carries its own tracking number, its own documents and its own status, both appear in Pick and Pack, and both labels are fetched automatically. The carrier name and consignment number are stored per shipment, so you can look up exactly which parcel is where when a customer calls. The webshop currently receives the first booking's tracking number.
Read more →One signing sheet per carrier, not one per delivery
A button in Pick and Pack produces one pickup manifest per carrier. The sheet lists every consignment being handed over — consignment number, recipient, parcel count and weight — and prints in two copies: one for the driver and one to keep with the signature. A day of twenty-one booked orders across two carriers becomes two pages instead of twenty-one.
Read more →Additional insurance shown where the shipment is booked
When a shipment is booked, the premium Fraktjakt charges to top up the carrier's own liability to the declared goods value is shown — alongside the cover the carrier already provides, so you can see what the premium actually adds. Where the carrier's own protection already reaches the declared value, it says so; where the carrier supports no additional insurance, it says that too. The choice is made per booking with the company setting as the default, and both the choice and the cost are stored on the order.
Read more →Every shipping document gathered in Pick and Pack
Fraktjakt produces more paperwork than labels — consignment lists the driver signs at pickup, pro forma invoices, CN22 and CN23 forms. A button in Pick and Pack lists them and prints or downloads them. A document covering several orders, such as a carrier manifest, is deduplicated automatically so the driver gets one copy rather than one per order. Documents are fetched and stored whether or not a label printer is configured — downloading an A4 document needs no printer.
Read more →A bundle's gallery composed from its components' images
The structure editor for composite products has gained an image step. It gathers every image from the bundle's components, groups them by part, merges near-duplicates, flags the ones that are too small or too low-resolution and pre-selects one clean image per component. A vision model can rank the candidates and recommend one image per part with a sentence of reasoning — the suggestion pre-selects, you decide what is applied, and the cost is shown before the call runs. Images are linked to the bundle product without copying files. Each part can be given a soft gradient backdrop, a horizon line or a contact shadow, drawn by the application rather than generated, so no product pixels change.
Read more →Bundle descriptions written from the components
Generating a description for a composite product draws its material from the parts it is made of — their specifications, catalogue descriptions and supplier texts. Hidden components are left out, and how many components and supplier texts to include is set in the AI Hub. A step in the structure editor shows exactly what will be sent and what it is estimated to cost, then generates a draft for approval — nothing is written to the product automatically.
Read more →The structure editor shows the pricing and takes manual prices
The preview in the structure editor shows the whole pricing calculation step by step, so you can check the right rules are applied before saving — and it shows the price the customer will actually pay, both excluding and including VAT. A manual markup and a sale price are set directly in the editor, next to the components they are priced against, with paired inc- and ex-VAT fields that update each other as you type. Where a sale price is active, the margin is reported separately for the regular and the sale price, so you see what the customer is really paying against. Saving the structure reprices the product immediately, including stock-proxy children inheriting the same list. The product list also shows which component limits the effective stock level, so the answer is there without opening the editor.
Read more →Alerts when a source stops delivering
Every import source — a competitor price feed or a supplier feed — is monitored against its expected interval. If fresh data fails to arrive, the system raises an alert on the System Alerts page automatically, so you learn that a feed has gone quiet while it can still be fixed, rather than when someone happens to notice the prices are stale.
Read more →Live crawl progress and monitored connections
While a supplier feed is being crawled, the status chip now shows a live product count and — where a previous successful import exists to compare against — a percentage and an estimated time of arrival. If the spider goes quiet for more than fifteen minutes the chip turns amber and says how long contact has been lost; if the silence continues long enough, the run is marked failed rather than left showing as running. Connections to Search Console, GA4, Merchant Center and the advertising services are now tracked on whether they actually work, not merely on whether a credential is stored: a broken token turns the hub badge red within one sync cycle, disables the Sync button with an explanation, and raises a notification once per outage that clears when the connection is restored. A nightly job also looks for four kinds of signal worth acting on — a search term that lost ranking, a page that lost search traffic, Merchant Center rejecting more products, a supplier running low — and sends at most five hints per run, ranked by magnitude.
Read more →Export products with cost and margin
A new export includes SKU, name, product type, selling price, sale price, primary purchasing source with supplier and supplier article number, purchase cost, total unit cost, margin in kronor and percent, contribution margin in kronor and percent, own stock and supplier stock. Use it to feed Google Ads and Merchant Center, or for margin analysis outside the platform. From the modern product list the export follows your current filters; from the classic list you can export a hand-picked selection. CSV downloads immediately; Excel is built in the background and arrives as a download link in the notification bar.
Read more →AI usage in kronor, by feature and by day
A new usage page in the AI Hub shows your company's AI activity for any month: number of generations, tokens used, cost in kronor and average quality — broken down by feature, by day and by model. Functions you have not used are listed as exactly that rather than omitted, so a zero is not mistaken for something free. Every AI cost figure across the platform is converted to kronor at the ECB rate for the relevant period, so the number you see is the one you actually pay. AI content generated automatically when supplier content is imported is now reported separately instead of being mixed in with product texts.
Read more →Competitor Discovery — find the shops you are not watching
A new page ranks every shop found selling your products by how many of them it actually carries, and shows whether you already import prices from there. Use it to spot competitors you are not yet watching — shops appearing on hundreds of your articles but missing from your price comparison today. Products where a market source found a possible match without being confident enough to decide land in a dedicated review list. Each row shows two signals to judge on: how distinctive the matched term is — a model number is decisive, a generic size description is not — and the competitor's price against yours. Rows are ordered strongest-evidence-first, so stopping halfway still means you have reviewed the most likely ones. Approving a listing adds it to your price comparison; rejecting it prevents the same listing being suggested again. The list can be sorted by what the product earns over a chosen window, by margin in kronor or in percent, with a header line showing what share of the period's revenue is already covered. A market lookup starts directly from the interface — you see how many products will be searched and the estimated cost before anything is written. Market data is a separate add-on with a free trial.
Read more →Pricing lens: unit cost and margin in kronor
The Pricing view now shows the full unit cost per product — including packaging and shipping where applicable — alongside margin in kronor and a summary strip across the whole filtered selection. Eight sort options, two range filters and three health filters, among them "loses money", turn the view into a worklist rather than a report. Companies where packaging and shipping costs apply also get contribution margin in a column of its own, separate from the standard margin.
Read more →WooCommerce: more fields and Rank Math SEO
Descriptions, excerpts, product names, backorder policy, brand, dimensions and weight can be pushed to WooCommerce through Platform Comparison. Attributes, brands and EAN are included when products are created or updated. Stores running Rank Math get meta titles and descriptions compared and pushed, detected automatically. VAT is converted using the product's own rate and the store's own incl/excl-VAT setting, read directly from the store, so the price that lands in the store is exactly the one you set. The comparison lists only fields WooCommerce can actually accept.
Read more →Duplicate values and decimal formats cleaned up in the Attribute Manager
A new panel shows which attribute values across your catalogue are really the same thing — `0.006kg` and `0.006 kg`, or `Blue` and `blue`. Values are grouped by the kind of difference (casing, number formatting, unit spelling), each group showing how many products are affected and which value is recommended as the target. Pure formatting differences can be merged in one background action; groups where the displayed text would genuinely change are listed separately and always require a manual decision. A second tab proposes consistent decimal formatting within each attribute — padding `1` to `1.0` where the attribute's other values carry one decimal, but never truncating a value with more. Merges go through the safe path, so procurement links, markup rules and synonyms are carried across rather than deleted.
All the evidence, in the attribute grid
The Attribute Workbench now shows the whole basis for a value next to the values themselves: your own and the supplier's images (with a built-in viewer), your own and the supplier's descriptions, the supplier's raw attribute rows, attached documents with their titles, and related products — what the product is a spare part for, what it replaces and what it fits, each relation labelled in the direction it reads. The seven separate buttons have been replaced by a single Show panel presenting every kind of material against the Catalogue and Procurement columns, with a count so you can see whether anything is on. Your choices are now stored on your user account rather than in the browser, so they follow you between machines and survive a cleared cache.
Read more →Competitor content: original descriptions composed from extracted facts
For sources marked as competitors, only the image URL and a checksum of the description text are now stored — not the text itself. That is enough to detect change without retaining anyone else's wording, and existing competitor descriptions are replaced by their checksum the next time each product is crawled. Where you need your own description, the structured facts stated on the page are extracted and the text is composed from those — never from the source's phrasing. A near-duplicate check confirms the result shares no substantial phrasing with the source before it is stored, and model, prompt version, timestamp and source URL are recorded with every generated text. The text is only recomposed when the source page has actually changed.
Read more →Bulk editing and cross-page selection in the Attribute Workbench
"Set values" has gained a mode that replaces what is already there instead of adding alongside it, so a correction leaves no duplicates behind. A value you type is also treated as a review decision: if an identical value was flagged for review, it is confirmed automatically. The header checkbox offers "select all N products matching the filter", so a bulk action can hit the entire result set at once rather than only the rows on the visible page. Changing the filter clears the selection. Variants are now grouped by family with the parent first, and each child shows its own short name instead of repeating the full family name on every row.
Read more →A cost estimate before every AI run
Translating product descriptions, generating alt text, running SEO analysis and generating AI content for categories now all show an estimated cost and a confirmation step before anything is charged. The estimate is calculated from your own previous runs for that specific function — where there is no history, it says so plainly rather than inventing a number — and it updates live as you change the selection, for instance by ticking an extra language. Bulk translation now runs as a background job: progress appears in the notification bar, and the completion message names which descriptions were translated, which were skipped and which failed, with reasons. The run continues in the background even if you close the tab.
Read more →A warning for purchase orders that were never sent
The purchase order list now flags orders created but never placed with the supplier. Ordinary drafts get a grey marker; a red "Not ordered · N waiting" appears when customer orders are waiting on goods that were never actually ordered — a problem otherwise invisible until the customer calls. The draft count in the list header is now a link that filters straight to them, and a toggle lets you combine that with supplier, search and sorting. Overdue delivery dates are coloured by how late they are, amber under 30 days and red beyond, with the number of days shown — because a list where half the rows are late has to tell "late since Monday" apart from "late since last year".
Read more →Competitor alerts and a freshness window
You can now be notified when a matched competitor goes out of stock, comes back in, drops their price by more than a threshold you set, or moves below your own price. Alerts are off by default and switched on per company. Your own price increases never trigger one — a repricing engine that just raised your price should not then alert you about what it did. At the same time, competitor-based pricing rules now act only on observations inside your company's freshness window, 14 days by default. Older observations still appear on the comparison page with their date, so you keep full visibility without the engine acting on a stale market, and the price breakdown explains when a rule declines and why. The competitor lens in the product list says "Out of date — seen [date]" instead of asserting a market position.
Read more →Model Report: measure the AI models against your own catalogue
A new page in the AI Hub shows which model we recommend per function, what the recommendation rests on, and — where we have measured against your own catalogue — how that compares to the cross-customer baseline. The report distinguishes recommendations backed by reproducible measurements from those reconstructed from earlier notes, and says plainly when a function cannot be measured at all rather than implying an answer is coming. Cost and quality are shown together in one chart per function, where a model both more expensive and lower quality than another is greyed out as dominated. You can order a measurement against your own products directly from the report — either chosen products or a random sample, which is the honest default since hand-picked products tend to be the well-documented ones — and you see the full cost before anything is charged.
Read more →Taxonomy browser and images in the Attribute Workbench
Settings → Taxonomies & Feeds now links to a browse page for Google (5,595 categories) and Facebook (2,967 categories). A top-level branch picker narrows the list instantly, a per-row toggle adds or removes a category without ticking rows and choosing an action, and the counts update as you click. Search now accepts a taxonomy ID as well. In the Attribute Workbench, a new toggle shows a thumbnail beside each product row — useful when filling in colour, shape or material — with the supplier's own image alongside, since a supplier photo is often the better evidence. A click opens an image viewer directly on the page with keyboard navigation.
Review AI results in one place
After a bulk generation run, the completion notice links to a new review page. Every generated product is shown side by side with its content before and after, and attribute conflicts — where two supplier sources disagree on a value — are lifted to the top. You approve or reject per field and choose between competing candidates without opening each product separately. Every proposed value is also labelled with the source that claimed it, so it reads "12.5 A (BooSpa)" rather than just a number. Attributes extracted during a bulk run land on the products directly, so the result of a run is available straight away.
Read more →Close order lines without pretending the goods arrived
Order lines that will never be delivered — because the supplier cancelled, or through stocktaking or write-off — can now be closed without being booked as received. A reason is required, the full trail is kept, and no stock movement is written. The action is available both on the delivery registration page, per line with a review step before saving, and on the order detail page for orders where the delivery page is no longer reachable. Closed lines are visually distinct from delivered ones and can be reopened if needed. Customer order reservations are not silently released — the confirmation dialog warns when a line being closed is reserved.
Read more →Stock valuation as of any date, with cost basis and exports
Stock Valuation has gained a date field, so you can see your stock balance and value as of any past date — for year-end closing and audits. The page states clearly that the date reflects when transactions were registered, not when goods physically moved. You can choose whether stock is valued at today's purchase cost or at the cost recorded on the purchase order line at stock-in; every row shows when its cost was recorded, and rows where the cost cannot be reliably determined are flagged and counted in the report header. Exports to PDF, Excel and CSV reproduce exactly what is shown — same date, cost basis, filters and sort order. The Excel workbook has editable quantity and unit-cost columns with running formula totals; the PDF repeats column headings on every page and states the totals once, at the end.
Read more →Competitor lens in the product list
The product list can now be filtered and sorted by competitor price position — show only products where you sit above a competitor who actually has the item in stock, or sort by the largest price gap. The lens is hidden automatically for companies with no matched competitors. A match that looks implausible, where the price ratio is unusually large, is flagged as "check this match" rather than presented as a pricing opportunity — you are sent to review the link instead of cutting a price that may be perfectly correct.
Read more →Attribute Workbench: category filter, purchase sources and half the load time
The workbench grid can now be filtered by category, with the full path spelled out in the picker (e.g. "Spas > Filters > Filters 0–99 mm diameter") so nodes sharing a name can be told apart — search matches any word in the path, in any order. Every product row has quick links to the catalog and procurement lists for that article number, and a new Procurement sources panel expands each row to show supplier, source article number and a direct link to the supplier's own product page. The compliance view is now computed in the database rather than in memory, roughly halving load time for a 3,800-product catalogue, and open filter panels remember where you left them.
More precise matching for brand-heavy catalogues
The matching scorer requires more than a brand name to pair two products. A token appearing across most of a brand's range contributes to the score but is not decisive by itself. The difference shows most in catalogues whose products carry no model or part numbers, where the brand is otherwise the strongest clue available.
Price suggestions by default — you switch automation on yourself
Competitor-based pricing rules now suggest prices by default and only apply them automatically once you explicitly switch that on. Until then the rule runs as usual and shows you the price it would set — visible in the product's price breakdown — while leaving your actual prices untouched. The setting lives per company under Settings → Pricing, and every change to it is recorded with the name of the user who made it. The reason is competition law: algorithmic pricing should be an active choice, documented, with an accountable person behind it.
Read more →Putaway sheets, returns sheets and count sheets
After registering a goods receipt you can now print an A4 putaway sheet straight from the stock-in page. Rows are grouped and sorted by inventory location — the order you walk the warehouse — with reservation notes so you can see what goes to the packing bench instead of the shelf. Registering returns offers the same sheet, covering the whole batch in one document regardless of how many customer orders the returns came from, built from the stock transactions actually written rather than the browser's own summary. The stocktaking view gains a count sheet with columns for SKU, product name, the system's expected quantity and a blank box for the counted figure — for whichever shelves your filter is showing. If a print agent is online everything goes straight to the A4 printer; otherwise it downloads as a PDF.
Read more →Copying to the catalog runs in the background — with bulk placement
Copying products from procurement into the catalog now returns immediately and continues in the background. A progress bar shows which products have been processed, which succeeded and which were refused and why. The dialog can be closed and reopened while the copy is still running, and a page reload does not lose your progress. You reach the dialog straight from the procurement product list: select any mix of products, decide per row, submit once. Several rows can also be ticked and given a category and classification in a single action — the fill does not overwrite choices already made on individual rows, and a later per-row edit still wins. If the category or classification does not exist yet, it can be created directly from the picker.
Competitor price comparison & competitor-based pricing rules
A new page shows every product matched to at least one competitor — your price against the lowest and the median, the gap in percent, and what the product earned over a chosen period. The list is sorted by what the gap is *worth*, so a 4% gap on your best seller ranks above a 40% gap on something that sells twice a year. The "X% below/above competitor" pricing rule is now live: choose lowest or median (median is less sensitive to a single clearance listing), and the price breakdown names the competitor behind the price. A new rule type sets a margin floor in kronor, and all clamp rules now apply last regardless of priority — a floor is always the last word.
Read more →Sales data where the decisions are made
A new Sales lens in the product list shows units sold, revenue excl. VAT, share of period revenue and last sold date — with a period selector (1/3/6/12/18 months) driving both columns and sorting. A "sold / not sold in period" filter turns the list into a worklist: "sold in the last 12 months and still missing a supplier link" is now one query. The product detail Stock tab gets a matching Sales card, and the matching worklist can be sorted by revenue with a coverage figure showing what share of revenue already has a supplier and competitor link — the measure that tells you when the matching effort is done.
AI proposes classifications, attributes and values
From a classification's name, the AI proposes which of your products belong in it, in a ranked review panel where nothing is linked until you press Link. From the linked products it then proposes the attributes the classification should declare, the allowed values for each, and a suggested value per product where the product's own text supports one — all as proposals with a confidence score and an "only high confidence" shortcut. You can now also choose an AI model per function (content, translation, attribute extraction, matching) instead of one model for everything, and AI usage is metered in tokens with cost broken down per function.
Read more →Attribute-driven category placement
Declare that an attribute drives category placement and the system handles the rest. Mark e.g. "Pipe dimension" as a category axis under "Pipe fittings" — every product carrying the value "2 inch" lands in the matching subcategory automatically. No rule per value, no manual step per product. The mapping table is editable, shows which values still have no category, and includes an explicit "create missing categories" action that lists what it will create before doing anything. Axes never set a product's primary category — they only add secondary links.
Read more →Waybills print automatically after payment
A background task watches booked shipments and queues the waybill on the label printer the moment Fraktjakt confirms payment. The button remains for manual printing and reprints. Nothing is fetched for unpaid shipments, and nothing is fetched twice.
Read more →Printing from any device — including the warehouse tablet
Labels, picking slips, packing slips and waybills now go straight to the right printer without anyone downloading a PDF or having DYMO Connect installed on that particular computer. A lightweight on-premises print agent connects to Ecombooster over outbound HTTPS and claims its jobs, so every print button works from warehouse tablets and any device on the network. Printers are resolved by purpose (product label, address label, waybill) rather than device name, jobs are queued with automatic retries, and the system health page shows whether each agent is online, offline or never started.
Read more →Duplicate detection: faster, safer, and a shared-source signal
Catalog products linked to the same supplier item now appear as duplicate candidates on their own tab — a signal based on a direct link rather than name similarity, so it works regardless of how products are named. Detection runs in the background, so the review page responds in under two seconds even for the largest catalogs. Groups above 25 members are withheld rather than offered with an irreversible Merge button, and name-based scoring requires more than a short shared name to chain products together.
Naming rules with exclusions, coverage analysis and quality warnings
Every supplier has a visible default naming rule, so what governs a product's name is always on the record. Every filter on a naming rule has gained a matching exclusion, so "all suppliers except one" can be expressed in a single rule instead of one rule per supplier. A new coverage analysis shows, per rule, how many products it matches, how many it actually names (a rule can match thousands but name none if a higher-priority rule takes them first), and flags rules that are dead or shadowed. The product page shows which rule governs its name, and configurable quality thresholds warn when a composed name comes out too short or meaningless.
Stock-source ladder for storefront delivery labels
Products now resolve to one of four rungs — own warehouse stock, supplier warehouse stock, backorder with a known arrival date, and backorder with no date — and the storefront receives the label matching the actual situation. Own stock and supplier stock are kept apart, so a product sitting in your own warehouse shows your own lead time rather than the supplier's. The current rung and the label the storefront will receive are now visible directly in the product list.
Read more →Edit brand, classification, categories and part numbers inline
A pencil icon in the product list opens a modal for editing brand, product classification and categories — without opening the full product form and saving the whole record. Categories show their full path, can be added and removed via typeahead, and exactly one must always be primary. Part numbers (MPN, EAN, RSK, SKU and others) are edited the same way from every product surface. Scanning a barcode with Enter keeps focus in the field, so an entire bag of parts can be registered without touching the mouse.
Review Queue — one workspace for all matching
A modern workspace now covers all three matching workflows: linking catalog products to procurement sources, spotting duplicates, and identifying procurement products that should be promoted into the catalog. All three are reached from one place with a subject switcher showing how many items wait in each area, and the evidence for each suggestion appears inline — including signals that cast doubt on a match, so reviewers can decide without leaving the page. Dismissing a suggestion is reversible, as is promoting a procurement product; merging duplicates is irreversible and the interface says so explicitly.
Seasonal planning and season remnants
Define recurring seasons (Christmas, Easter) with a calendar window, a sell-off deadline and a buy-in lead time. Products can be tagged to one or more seasons, and a supplier profile can carry a default season so an entire new series is labelled in bulk when copied in. A Christmas article sitting in the warehouse in May is now recognised as out of season rather than a non-seller, so its ABC/XYZ class, days of stock and reorder point are not distorted by months of zeroes. The restock engine also compares the days left in a season against the measured lead time and flags suggestions that cannot land in time ("arrives week 51, season ends week 52"). A new worklist shows what is still in stock per season ahead of its sell-off deadline, ordered by capital at risk.
Read more →Image health, AI-generated alt text and image standardization
Ecombooster now automatically checks every product image for common quality problems — too small, low resolution, or an unusual aspect ratio — and gives each product an Image Health score. A worklist makes it quick to find and fix products with weak or missing images before they affect your storefront. Alt text can be generated straight from an image: the AI looks at the actual photo, grounded in the product name and brand, and stores the result as a review candidate you choose to apply — it never silently overwrites alt text you wrote yourself. A new action pads and resizes a batch of images to a consistent square format in the background, keeping the originals.
Read more →Publish readiness, supplier change awareness and cost-change guard
Every product now gets a unified publish-readiness score (0–100) combining name, description, image, alt text, price, category, required attributes and SEO health into one status, with a breakdown showing exactly what is blocking publication. Filter the list to "one field from publishing" for a focused worklist. When a supplier feed changes the name, description or excerpt of a product you have already copied in, your curated content is never overwritten automatically — the change goes to the review queue with old and new values alongside your own, to pull in with one click or dismiss. And a cost-change guard holds the repricing when a supplier changes a purchase cost beyond your threshold: raw data always updates, but the selling price waits for approval.
Read more →Assortment analysis: ABC/XYZ, movement and recommended actions
Your assortment is now classified every night on three axes: ABC ranks products by annual usage value, XYZ measures how predictable demand is, and movement class (F/S/N) plus days of stock shows how long current stock will last. A new matrix report shows the full 3×3 grid with count, annual value and share per cell — and the stocking policy each cell implies (AX = run lean, AZ = carry a real buffer, CZ = order on demand). The report opens with a ranked list of recommended actions drawn from your own data: which high-value products are out of stock right now, how much capital is tied up in excess stock, which products are valuable and predictable enough to run lean. The classes also appear in the Low Stock report, the product list and on the product page, and the report exports as CSV or PDF to take into a purchasing meeting.
Read more →Setup profiles and learned rule suggestions
Save a named collection of setup choices — default classification, categories, markup rule, description template and naming rule — and attach it to one or more suppliers. When you copy products in from that supplier, the profile fills in what you left blank without overwriting anything you chose explicitly. A supplier-specific profile always wins over a catch-all, and a profile can also be re-applied to a batch of existing products to bring older imports up to standard. On top of that: when operators repeatedly make the same manual edit (removing "SALE!" from every title, say), the system can detect the pattern and suggest a permanent cleanup rule — with before/after samples, and nothing ever applied without explicit approval.
Price calculator in the top bar and returns putaway sheet
A calculator icon on every admin page looks up sell price and margin instantly — no more opening pricing rules and calculating by hand. Pick a procurement product to price it through the pricing engine, or enter a purchase price manually: both modes show the price excl. and incl. VAT, margin in kronor and percent, the markup factor, every matching markup rule with the price it would produce, and customer-group agreement prices with their margins. And after registering a batch of returns, a putaway sheet now appears automatically, grouped by inventory location and sorted by location code so shelves can be walked in order — with a print button.
Bulk label printing and a per-product label factor
A "Print labels" button on each stock-in delivery opens a modal with every product and an editable label quantity, pre-filled from the amount received — set them all at once, reset to zero, or adjust individual rows and print everything in one click. Each product now also has a label factor controlling how many labels print per received unit: factor 3 × 4 units gives 12 labels, while factor 0 means "one label for the whole row" — useful for bagged goods such as gaskets and o-rings that go into a single bag anyway. The factor can be edited straight from the modal while receiving.
Demand-driven restock suggestions and supplier stock alerts
Ecombooster now calculates a suggested minimum stock level and reorder quantity per regularly-selling product, based on actual sales velocity and the *measured* delivery lead time of that product's supplier. Suggestions appear read-only alongside your manual levels and can be bulk-applied with a before/after preview — your own levels are never touched by the nightly run. The safety buffer is configurable per company, per product and per demand class, and companies with steady high-volume demand can switch to a service-level-based statistical method. A new report also lists supplier sources trending toward empty regardless of your own sales volume, and restock suggestions automatically raise the quantity to cover the gap before that source runs out.
Read more →Low stock and stale purchase orders surface in Pick & Pack
The procurement sources block now expands automatically when a product's stock will fall below its minimum *after* the current order is picked — not only when the order itself is short. A badge shows current stock against the threshold, and if an open purchase order already covers the shortfall the row stays collapsed but the badge names it. Lines with an open purchase order older than 30 days that still has undelivered quantity get a warning with the PO number and its age — a long-outstanding PO is more often a forgotten remnant than real incoming stock. Markup rules have gained exclusions at the same time, so a general shipping surcharge can skip products that already have a supplier-specific rule.
Flag returns before the parcel arrives
When a customer calls or emails to say an order is coming back, staff can flag it as an expected return straight away — before the parcel physically arrives. The order appears in the Pick & Pack list with an "On the way back" status badge and a Register return button, so the warehouse can book the return the moment the goods land. The flag records who set it and when, can be applied to the whole order or individual lines, and clears itself automatically once the real return is registered.
Read more →Automatic shipping method mapping for Fraktjakt
The shipping method the customer chose in your webshop is now matched to the corresponding option in Fraktjakt — "DB Schenker Ombud" finds the right parcel-shop service automatically. The matched method appears at the top of the shipping list, is pre-selected and tagged as the customer's choice, ready to book in one click. If it is more than 20% more expensive than the cheapest alternative, a price warning is shown. When no mapping exists yet, a banner prompts you to pick the right option and tick "remember this choice" — the mapping is then saved automatically and applied to future orders.
Read more →Automatic push on change (push-on-change)
Keep your store in sync without manual runs: Ecombooster can push stock, prices and backorder status to the platform the moment they change in the backend. Opt-in per field, the option to push available stock (on-hand minus reserved), safety guards that hold oversized price changes or too-low margins for review, and a Platform Push Queue for queued, held and failed pushes. Available for Nyehandel; more platforms to follow.
Read more →Field-change log & change history — full traceability
Every change to price, cost, stock and lifecycle is now logged with what changed, why, and by whom or which platform — linked back to the cause (e.g. a feed import or a push to the store). A "Change history" button on the product page shows the full timeline with old and new values. Governance for teams editing thousands of products.
Read more →Returns workspace — process returns in batches
A new returns workspace where you search multiple customer orders at once, add them to a working list, set return quantities and putaway locations per line, and register everything in a single action. Returnable quantity is based on what was actually picked, and stock is booked correctly even for proxy products.
Attribute Workbench: bulk review & AI conflict detection
Review and set attribute values across many products at once with three clear bulk actions. When AI extracts different values for the same attribute from different sources, all candidates are now kept and flagged as a conflict — instead of silently guessing the first one. On-demand filters, frozen columns and horizontal scroll make wide attribute matrices easy to work in.
Read more →Price guardrails: outlier & margin warnings before publishing
Platform Comparison now shows the actual value difference and percentage for price, cost and stock — and blocks accidental large price changes (e.g. 190 kr when the right price is 6,390 kr) until you approve them. Products below your minimum gross margin are flagged automatically so you catch broken pricing rules or wrong costs before they reach the store.
Modernized product interface & Product Classification
The product detail page and Product Classification now have a fully modernized interface — search, sort, and manage categories, attributes and linked products right on the page without reloading. A Modern/Classic toggle remains if you prefer the previous view.
Advanced PIM governance: approval workflows, audit and rollback
Add-on for teams managing product data seriously — multi-step approval workflows, field-level audit log, mass-edit preview and safe rollback, plus custom quality rules. For when multiple people edit thousands of products and governance becomes critical.
Read more →Price Intelligence — competitor price monitoring (closed beta)
Automatic competitor monitoring + dynamic pricing rules. Match the lowest competitor but not below minimum margin, beat by X%, or stay within Y% of a specific competitor. Per product, category or brand. Closed beta — contact us for early access.
Read more →AI Search Optimization (AIO) for generative search engines
Optimize your product catalog to be cited by ChatGPT, Gemini and Google AI Overviews. AI citability checks, auto-generated and validated schema.org JSON-LD, plus cross-provider second opinion between Claude and OpenAI.
Read more →SEO & Marketing Hub — one dashboard for all SEO and advertising
Search Console, GA4, Merchant Center, Google Ads, Tracked Keywords, Google Trends and DataForSEO (6 markets: SE, NO, DK, FI, DE, UK) gathered in one unified view. Striking distance, cannibalization and CTR anomalies are identified automatically.
Read more →Supplier comparison & negotiation report (closed beta)
Automatic report comparing prices, stock and delivery times across your suppliers per item — and showing where you're paying more than necessary. Concrete leverage for negotiations. Currently closed beta — contact us for early access.
Read more →Composite products (BOM) with per-component pricing
Build bundles, kits and split supplier packages with per-component pricing. Automatic stock and webshop integration so bundles are marked unavailable when a component runs out.
Read more →B2B customer pricing with customer groups and tiered discounts
Customer groups with automatic price lists (Reseller, Volume customer, Standard), tiered volume discounts, multi-currency with daily ECB rates, and contract-negotiated prices with time periods.
Read more →Multi-tenant architecture for agencies
Run multiple clients' e-commerce from one interface. Separate workspaces with full data isolation, per-client branding and pricing rules, role-based access and centralized billing.
Read more →Open REST API for custom integrations
Need to connect Ecombooster to an ERP, CRM or your own tool? Our REST API is open and documented — contact us for access.
Read more →New glossary for PIM, e-commerce and supplier management
30+ terms explained in Swedish and English — from PIM and AIO to BOM and multi-sourcing. Searchable and categorized per domain.
Read more →AI Attribute Extraction
Extract structured product attributes from unstructured supplier descriptions using AI. Includes unit conversion, conflict detection and fuzzy matching.
Google Trends Integration
Track keyword trend direction — rising, declining or stable — with weekly automated sync. Trend scores visible in keyword tables.
Content Freshness Tracking
Automatic detection of products with stale content. Configurable threshold with priority escalation to keep search rankings fresh.
Stock Proxy Products
Link products with unit and price conversion factors. Inventory, pricing and order fulfillment adjust automatically.
AI Hub — Unified AI Command Center
Six focused views bring together all AI functionality: Content Studio, Quality Center, SEO Intelligence, settings and cost tracking.
SEO & Performance Hub
Unified dashboard for all SEO and advertising data with seven sub-pages: Dashboard, Search Console, Analytics (GA4), Merchant Center, Advertising, Keywords and SEO Health.
AI-Powered SEO Analysis
One-click analysis with Claude AI that generates actionable recommendations based on actual data from Search Console, GA4 and ad spend.
Google Analytics 4 Integration
Import organic traffic metrics, conversions and revenue attribution. Cross-source analysis correlates GA4 data with Search Console and advertising spend.
Scope-Based AI Content Generation
One AI call generates all outputs — description, excerpt, meta title and meta description — with consistent tone and up to 30% lower cost.
Kodmyran REST API Integration
Upgraded to full REST API integration with support for products, categories, attributes, multilingual data, images, prices and stock status.
SEO Health Score
Per-product SEO scores (0–100) based on 6 weighted components. Color-coded in the product list with history tracking.
Stock Status Mapping
Define customer-facing stock labels and delivery timeframes per classification with automatic sync to ecommerce platform.
AI Search Optimization (AIO)
Optimize product content for AI-powered search engines like ChatGPT and Google Gemini. Three intensity levels and specialized writing strategies ensure your products are cited by AI assistants.
Automatic SEO Keyword Integration
Auto-fetch and rank keywords from SEO database, Google Search Console and content opportunities. Keywords are integrated directly into the AI prompt.
Schema.org Structured Data
Auto-generate Product JSON-LD and FAQPage schema for rich search results directly from AI content generation.
Advertising Analytics — Google Ads & Meta Ads
Connect ad platforms with actual customer orders to see real ROAS. Per-product profitability, basket analysis and four-quadrant ad efficiency analysis.
AI Content Generation with Claude & OpenAI
Generate unique product descriptions with full context — attributes, images, supplier data. 7 quality dimensions, 11 writing strategies, versioning and scheduled batch generation.
Google Shopping Custom Labels
Rule-based automatic assignment of custom labels (0–4) for bidding strategies. Segment by margin tier, stock level, brand or any product attribute.
Data Quality Monitoring
Field reports with coverage scoring per supplier. Regression detection with automatic alerts when data quality drops.
Modernized Pick & Pack
New card-based warehouse interface with real-time stock indicators. Create backorders and inventory adjustments directly from picking view.
Stocktaking & Stock Valuation
Complete stocktaking workflow with warehouse location listings and automatic inventory transactions. Stock valuation report with cost breakdown per supplier.
Attribute Workbench
Visual workspace for attribute management with pipeline-based transformations. Automatic extraction and normalization of attribute values from supplier data.
30+ Product Filters
Filter options including data quality, attribute compliance, margin range, image count, sync status and duplicate detection.