BACKLANE / PROMPTS

A useful question. A clear next step.

12 curated prompts for your own marketplace data. Choose a task, replace the placeholders and paste the prompt into your AI chat connected to Backlane.

All templates ask for read-only requests. Check the connection, date range and results in your chat. A prompt does not replace technical access controls; available data depends on the marketplace and your permissions.

12 of 12 prompts

OTTO / Listings

Why is my OTTO product not online?

What you need
OTTO connection and SKU
What you get
Status, original messages and next checks for each variation.
Check OTTO SKU [SKU] in [connection] using read-only requests.
Show its marketplace status, active, MOIN, error code and original message, hints, lastModified and any shop link.
Separate errors from optimisation hints. State the source and retrieval time. Missing fields or records are unknown; do not infer a status.
Explain what is evidenced, what remains open and what I should check next. Do not change products or activation settings.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: OTTO-Demo · SKU: DEMO-SHIRT-M

The variation is reported as RESTRICTED. The specific cause remains unknown because the original error message is missing.

Data sources: OTTO status queries for one SKU.

Scroll the table sideways if needed.

Status check for DEMO-SHIRT-M
CheckExample findingNext check
Marketplace status RESTRICTED Retrieve the associated original errors.
Activation active = true Investigate the restriction despite the intended activation.
Error code and original message Not available Obtain the missing evidence before correcting product data.

What this does not establish

active = true does not prove publication. The status alone does not identify an incorrect product field. A sandbox status also does not establish publication in the live shop.

View integration ↗

OTTO / Stock

Which OTTO SKUs are below my stock threshold?

What you need
OTTO connection and minimum quantity
What you get
Reported quantities below the threshold, with data gaps listed separately.
Read all available OTTO quantities in [connection], following every results page. My minimum stock threshold is [quantity] units.
Show SKUs with quantity below this threshold, sorted by quantity ascending: sku, quantity, lastModified. Identify zero quantities separately.
State the source, retrieval time and completeness. An unknown SKU or missing value is not zero. lastModified is not the retrieval time and does not prove a synchronisation failure.
Describe these as reported OTTO quantities, not verified warehouse stock. Use read-only requests only.
Direct link

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: OTTO-Demo · Stock threshold: 5 units

Two SKUs are below 5 units: DEMO-A with zero and DEMO-B with two. DEMO-D has no quantity value.

Data sources: four SKUs, all results pages covered. DEMO-C reports exactly 5 units and is therefore not below the threshold.

Scroll the table sideways if needed.

Quantities below the threshold, with a separate data gap
SKUExample findingNext check
DEMO-A quantity = 0 Compare the zero quantity with the source system.
DEMO-B quantity = 2 Check whether replenishment is planned.
DEMO-D · data gap quantity missing Establish the quantity; do not count it as zero stock.

What this does not establish

These are reported OTTO quantities. They establish neither physical warehouse stock nor a synchronisation failure. Stock cover cannot be calculated without outbound quantity data.

View integration ↗

OTTO / Orders & returns

Which OTTO order items are ready to ship?

What you need
OTTO connection
What you get
PROCESSABLE items, oldest first, without buyer details.
Create a read-only shipping worklist for OTTO [connection]. Query fulfillmentStatus=PROCESSABLE with mode=AT_LEAST_ONE and no lower date limit. Follow all nextcursor pages.
Keep only PROCESSABLE items and count each salesOrderId plus positionItemId once. Show the 20 oldest by processableDate, with order ID, item ID, SKU and expectedDeliveryDate where supplied.
Show the total and number of omitted items. List missing dates separately; do not infer a shipping deadline from expectedDeliveryDate. State source, retrieval time and completeness. Do not output buyer details or mark anything as shipped.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: OTTO-Demo

Three items are PROCESSABLE. Two have a known chronological order; one has no processableDate.

Data sources: all results pages with no lower date limit. Three distinct order/item pairs, all three shown. Times are in CET.

Scroll the table sideways if needed.

Shipping worklist with an undated item kept separate
Order / item · SKUExample findingNext check
DEMO-100 / 1 · DEMO-A PROCESSABLE; processableDate: 27 October 2026, 09:00 Check the oldest known item with the shipping team first.
DEMO-101 / 1 · DEMO-B PROCESSABLE; processableDate: 28 October 2026, 14:00 Check the remaining shipping work.
DEMO-102 / 1 · DEMO-C PROCESSABLE; processableDate missing Establish the timestamp; do not rank the item as the newest.

What this does not establish

PROCESSABLE is not evidence of shipment. A missing timestamp prevents an age-based ordering. expectedDeliveryDate describes expected delivery and does not establish a shipping deadline.

View integration ↗

OTTO / Buybox & advertising

Which OTTO Ads campaigns are below my target ROAS?

What you need
Connection, date range and target ROAS
What you get
Campaigns below target, sorted by advertising cost.
Check OTTO Ads in [connection] read-only for [start date] to [end date] in CET/CEST. State the precise boundaries and supply both dates. My target ROAS is [target].
Read all campaign pages. Show each campaign's ID, name, status, totalCosts, totalSales, ROAS, dateRange and lastCalculatedAt. Check monetary units in the current schema and normalise to euros.
Recalculate ROAS when costs are positive. Sort campaigns below target by costs descending. List missing data and zero-cost cases separately as not assessable. ROAS is not profit. State retrieval time and completeness. Do not change budgets or campaigns.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: OTTO-Demo · Period: 1–7 October 2026 inclusive, CEST · Target ROAS: 3

Demo A and Demo B are below the target ROAS of 3. Demo A comes first in the review list because its cost is higher. Demo C cannot be assessed with zero cost.

Data sources: three campaigns, all results pages covered; the same reporting period for each campaign. Monetary values are already normalised to euros; revenue is attributed to advertising.

Scroll the table sideways if needed.

Recalculate ROAS and prioritise by cost
CampaignExample findingNext check
101 · Demo A €400 cost; €800 attributed revenue; ROAS = 2.0 Check the products driving revenue and costs in this campaign.
102 · Demo B €100 cost; €50 attributed revenue; ROAS = 0.5 Review product performance and the campaign period.
103 · Demo C €0 cost; €0 attributed revenue; ROAS cannot be calculated Check activity and data coverage; do not assume a ROAS of 0.

What this does not establish

ROAS is attributed revenue divided by positive advertising cost, not profit. These figures do not establish incremental revenue from advertising or justify an automatic budget change on their own.

View integration ↗

KAUFLAND / Stock

Which Kaufland offers report zero stock?

What you need
Connection, storefront and fulfilment type
What you get
Zero quantities, visibility and missing data kept separate.
Read my Kaufland offers in [connection], storefront [country code], fulfillment_type [fulfilment type], following all limit/offset pages.
Show offers with amount=0: id_unit, id_offer, id_product, amount, status and is_live where supplied. List missing quantities separately; missing is not zero. Put positive quantities with is_live=false in a separate review list.
State source, retrieval time, filters, total count and completeness. Newly created offers may appear with a delay. Do not infer visibility or physical warehouse stock from quantity or status alone. Do not change anything.
Direct link

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: Kaufland-Demo · Storefront: de · Fulfilment type: fulfilled_by_merchant

One offer reports zero stock. Another reports 12 units but is not live. The third has no quantity value; it does not count as zero stock.

Data sources: three Kaufland offers, with all results pages covered.

Scroll the table sideways if needed.

Check reported quantities and visibility separately
Offer (id_offer)Example findingNext check
DEMO-A amount = 0; is_live = false Compare the reported quantity with the source system.
DEMO-B amount = 12; is_live = false Check the reasons why the offer is not live.
DEMO-C amount missing; is_live = false Resolve the data gap; the quantity remains unknown.

What this does not establish

Reported offer quantities are not verified warehouse stock. A positive quantity does not guarantee visibility. The table establishes neither the reason for is_live = false nor a synchronisation failure.

View integration ↗

KAUFLAND / Buybox & advertising

How does my Kaufland offer compare in the Buybox?

What you need
Connection, storefront, EAN/product ID, condition and own offer reference
What you get
Reported rank, price including shipping and delivery-time comparison.
Check the regular Kaufland Buybox read-only for [EAN or id_product], storefront [country code], condition [condition], in [connection]. My id_offer is [reference]. Resolve an EAN to its product ID first and confirm the title.
Show up to ten available offers with reported rank, seller, condition, item price, shipping costs, combined price, delivery range and fulfilment type. Verify units before converting values. Calculate combined prices only from confirmed item and shipping costs; missing shipping costs are not zero.
Identify my offer only with a confirmed match. State retrieval time, returned offer count and num_units. Separate observations from possible explanations. Do not promise a rank improvement or change prices.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: Kaufland-Demo · Storefront: de · id_product: 123450, Demo-Tasse · Condition: NEW · Own id_offer: DEMO-ME

Your confirmed offer is ranked second. Its combined price matches rank 1, but the reported delivery time is longer. Shipping cost is missing for rank 3.

Data sources: the regular Buybox for one confirmed product; 3 offers returned, num_units = 3. Prices are already normalised to euros and delivery times to days.

Scroll the table sideways if needed.

Read reported ranks and comparable prices
Offer / rankExample findingNext check
Seller A · rank 1 €19.90 + €2.90 shipping = €22.80; delivery: 2–3 days Compare delivery time with your own offer.
Your offer DEMO-ME · rank 2 €20.90 + €1.90 shipping = €22.80; delivery: 3–4 days Check your delivery and shipping information.
Seller C · rank 3 €21.90; shipping cost missing; combined price and delivery time unknown Obtain the missing comparison values.

What this does not establish

The difference in delivery time does not establish the cause of the ranking. A single query provides no history or guarantee of a better rank. Missing shipping cost is not zero; the endpoint returns at most ten offers.

View integration ↗

KAUFLAND / Orders & returns

Which Kaufland products get ordered and cancelled most?

What you need
Connection, storefront, fulfilment type and date range
What you get
Ordered and cancelled units for each matched product.
Analyse Kaufland [connection], storefront [country code], fulfillment_type [fulfilment type], read-only. Use creation time from [start date] 00:00 inclusive to [end date] 00:00 exclusive, Europe/Berlin.
Read all order-unit pages and all statuses. Count each id_order_unit once. Group only by confirmed SKU mapping and condition.
Show ordered units, cancelled units, units excluding cancellations and cancellation share. Calculate shares only with a positive denominator. Separate unknown statuses and mapping gaps; affected rankings remain provisional. Returns are not additionally deducted. State filters, retrieval time and completeness. No buyer details, revenue or profit estimates, or changes.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: Kaufland-Demo · Storefront: de · Fulfilment type: fulfilled_by_merchant · Created from 1 October 2026, 00:00 to 8 October 2026, 00:00 exclusive, Europe/Berlin

Among the matched products, DEMO-A has more ordered units and a higher cancellation share: 4 out of 20 units, or 20%. Two further units without a SKU match remain separate.

Data sources: 32 distinct order units, all pages and statuses covered. 30 units are matched to two SKUs in NEW condition. Returns are not additionally deducted.

Scroll the table sideways if needed.

Ordered and cancelled units by product
Product / mappingExample findingNext check
DEMO-A · NEW 20 ordered; 4 cancelled; 16 excluding cancellations; cancellation share: 20% Review the reported cancellation reasons for this product.
DEMO-B · NEW 10 ordered; 1 cancelled; 9 excluding cancellations; cancellation share: 10% Review the difference over a comparable period.
Unresolved mapping 2 units without a confirmed SKU Resolve the product mapping before drawing final conclusions from product metrics.

What this does not establish

Ordered units are neither shipped units nor revenue. The cancellation share establishes no cause. Unmatched units may change product metrics; returns need a separate analysis.

View integration ↗

KAUFLAND / Orders & returns

Which Kaufland returns need review?

What you need
Connection, storefront and fulfilment type
What you get
Open return items grouped by processing status.
Check Kaufland [connection], storefront [country code], fulfillment_type [fulfilment type], read-only.
Read return-units with need_to_be_returned, return_arrived, return_in_clarification and return_in_repair, with no lower date limit and across all pages. Count each id_return_unit once.
Add the return-level status for each id_return. Show item ID, return ID, item status, return status and creation date. Group by processing status, oldest first. Suggest next checks without inferring a binding deadline from age.
State source, retrieval time and gaps. No customer or address details or free-text notes. Do not change anything or send messages.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: Kaufland-Demo · Storefront: de · Fulfilment type: fulfilled_by_merchant

Three return items need different checks: one awaits return, one has arrived and one is under clarification.

Data sources: all pages for the four requested open item statuses, with no lower date limit. Three distinct items in three returns; none in return_in_repair.

Scroll the table sideways if needed.

Read item status and return status separately
Item / return · createdExample findingNext check
501 / 701 · 20 October 2026 Item: need_to_be_returned; return: label_generated Check whether shipping or receipt evidence is already available.
502 / 702 · 21 October 2026 Item: return_arrived; return: package_received Reconcile warehouse receipt with the pending item review.
503 / 703 · 22 October 2026 Item: return_in_clarification; return: package_received Check the unresolved question and who is handling it.

What this does not establish

Package receipt does not establish acceptance or refund of every item. The creation date does not show when the current status began or establish a processing deadline on its own.

View integration ↗

MIRAKL / Stock

Which Mirakl offers are low on stock or inactive?

What you need
Marketplace connection, shop and minimum quantity
What you get
Offer quantities and activity reported as separate findings.
Read all offer pages using OF21 in [Mirakl connection] and [shop], read-only and respecting request limits. My stock threshold is [quantity].
Show offer_id, shop_sku, product_sku, quantity, active and any returned inactivity_reasons. Separate quantities below the threshold, quantity=0 and inactive offers with positive quantities. List missing values separately.
Do not infer activation status from state_code. If an existing OF26 check returns zero, this does not establish whether quantity is zero or the offer is unavailable.
State source, shop, retrieval time and completeness. These values do not prove physical warehouse stock. Do not change offers.
Direct link

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: Mirakl-Demo · Shop: 2000 · Stock threshold: 5 units

Two offers are below 5 units. A third reports 12 units but is inactive. Quantities and activation lead to different next checks.

Data sources: three OF21 offers, all pages covered. The table is an excerpt with confirmed offer_id and shop_sku mappings; no inactivity reason was supplied.

Scroll the table sideways if needed.

Distinguish low quantities from inactive offers
offer_id / shop_skuExample findingNext check
801 / DEMO-A quantity = 0; active = false Check the zero quantity and activation status in the source system.
802 / DEMO-B quantity = 2; active = true Review replenishment for the low-stock offer.
803 / DEMO-C quantity = 12; active = false; reason missing Check the inactivity reason with the marketplace.

What this does not establish

Positive stock does not explain inactivity. state_code is not an activation status. Reported offer quantities do not establish physical warehouse stock.

View integration ↗

MIRAKL / Listings

Why did my Mirakl product import fail?

What you need
Marketplace connection, shop and known import ID
What you get
Processing status, original errors and next checks.
Check known product import [import ID] on [Mirakl connection, shop] read-only.
Show import_status, reason_status, creation time, transformation counters, available integration details and report availability. State if processing is still running.
Review available transformation and integration error reports. Group by stage, field and original message; show message counts separately from distinct affected products. Include product references, evidence locations and next checks. Do not invent error codes or required fields.
Missing or unreadable reports do not prove success. State retrieval time and gaps. Do not start an import or change data.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: Mirakl-Demo · Shop: 2000 · Known product import ID: 64021

The import is COMPLETE. Three messages in the transformation report concern two products. The integration report cannot be read, so the overall assessment remains incomplete.

Data sources: P42 status and P47 report. P44 could not be analysed. Error messages are summarised here.

Scroll the table sideways if needed.

Distinguish message counts from affected products
Stage / evidenceExample findingNext check
Transformation · report rows 2 and 3 2 messages for DEMO-VASE-01: check category mapping and an attribute value. Compare both original messages with the same product row in the source file.
Transformation · report row 4 1 message for DEMO-VASE-02: check category mapping. Check the mapping to this marketplace operator’s category.
Integration · P44 report Unreadable; number of affected products unknown. Provide the report in a readable format and assess it separately.

What this does not establish

COMPLETE does not mean every product was integrated successfully. Three messages here concern two distinct products in one stage, not an overall failure count. Required fields and report formats depend on the operator.

View integration ↗

MIRAKL / Orders & returns

Which Mirakl orders await acceptance or shipping?

What you need
Marketplace connection and shop
What you get
Separate worklists for WAITING_ACCEPTANCE and SHIPPING.
Check [Mirakl connection] and [shop] read-only. Read WAITING_ACCEPTANCE and SHIPPING separately using order_state_codes, with no lower date limit, across all pages and respecting request limits.
Count each order_id once per connection and shop, and each order_line_id once within it. Show order and line status separately, offer_sku, created_date and shipping_deadline where supplied.
Sort SHIPPING by available shipping_deadline; list missing deadlines separately. Sort WAITING_ACCEPTANCE by creation date without inferring an acceptance deadline.
State source, retrieval window and completeness. No buyer details. Do not accept orders, confirm shipping or send messages.

View example result

Here is what a response using sample data can look like.

Example inputs
Connection: Mirakl-Demo · Shop: 2000

One order awaits acceptance and two are SHIPPING. One shipping order has no shipping deadline and remains a separate review item.

Data sources: separate WAITING_ACCEPTANCE and SHIPPING queries, all pages with no lower date limit. Three distinct orders with one line each; times are in CET.

Scroll the table sideways if needed.

Prioritise acceptance and shipping separately
Order / line · SKUExample findingNext check
DEMO-201 / 1 · DEMO-A Order and line: WAITING_ACCEPTANCE; created: 27 October 2026, 09:00 Check the requirements for acceptance.
DEMO-202 / 1 · DEMO-B Order and line: SHIPPING; shipping_deadline: 30 October 2026, 16:00 Review shipment against the reported deadline.
DEMO-203 / 1 · DEMO-C Order and line: SHIPPING; shipping_deadline missing Clarify the deadline with the operator; do not assume low urgency.

What this does not establish

SHIPPING establishes neither completed shipment nor payment received. The creation date does not establish an acceptance deadline. Orders with multiple lines need each line's status checked separately.

View integration ↗

Across marketplaces / Stock

Where do my marketplace stock figures differ?

What you need
Connections, storefronts/shops and confirmed product mapping
What you get
Reported channel quantities side by side, without summing warehouse stock.
Compare reported quantities read-only in [connections and storefronts/shops] using this confirmed mapping: [internal SKU, channel SKU/offer ID, variation, pack size, condition].
Read all required pages. Show channel quantities, status or availability, sources and retrieval times side by side for each product. Separate missing records, missing quantities, zero and inactive offers.
Flag quantity differences for review. Do not add channel quantities into total stock; several offers may share one warehouse pool. Do not infer synchronisation failures from differences alone.
List unresolved mappings separately. State completeness and retrieval time differences. Do not change stock quantities.
Direct link

View example result

Here is what a response using sample data can look like.

Example inputs
OTTO-Demo · Kaufland-Demo, storefront de · Mirakl-Demo, shop 2000 · confirmed mapping for internal SKU DEMO-TASSE, blue, single unit, new

For the same product, OTTO reports 8 units and Kaufland 5. The Mirakl offer was found but its quantity is missing. The difference of three units calls for review.

Data sources: fully retrieved channel data with confirmed product mapping. The largest gap between retrievals is two minutes.

Scroll the table sideways if needed.

Channel quantities side by side, without a stock total
Channel / referenceExample findingNext check
OTTO · SKU DEMO-TASSE quantity = 8; marketplace status: ONLINE Reconcile the reported quantity with channel allocation and the source system.
Kaufland de · id_offer DEMO-TASSE amount = 5; is_live = true Review the difference, taking allocation and retrieval timing into account.
Mirakl shop 2000 · offer_id 804 Offer found; quantity missing; active = false Resolve the missing quantity and inactivity reason separately.

What this does not establish

8 + 5 does not establish total stock: the offers may share one warehouse pool. The difference does not prove a synchronisation failure. The missing Mirakl quantity is not zero.