Systems & Data
One source of truth. PIM and ERP integration, product feeds, analytics and server-side tracking, multi-channel dashboards and AI automation — the plumbing that decides whether multi-channel selling scales or breaks.
Questions & answers — Systems & Data
What systems do you integrate?
Shopify and Atomstore stores, Baselinker, Comarch ERP XL and other ERPs, warehouse systems, Allegro and Amazon, Klaviyo, Google Merchant Center and price comparison feeds. The goal is always the same: one stock truth, one order queue, one view of revenue.
What is a PIM and do we need one?
A product information management system holds the master data for every SKU: attributes, descriptions, images, EANs, compliance fields, translations. If you sell on more than one channel or have more than a few thousand products, you need one, because channel-specific listings should be generated from it rather than retyped.
Can you fix our tracking? The numbers in GA4, Meta and Shopify never match.
Yes, that is the most common request. We rebuild GA4 e-commerce events against real orders, move tracking server-side with Elevar or a GTM server container so ad blockers and consent do not erase half your data, and reconcile the three sources until the differences are explainable.
What does server-side tracking give us?
First-party data sent to Google, Meta and TikTok from your server rather than the browser. It survives ad blockers and Safari restrictions, respects Consent Mode v2 for EU visitors, and gives the ad platforms cleaner conversion signals, which directly improves their optimisation.
Can you show store and marketplace revenue in one place?
Yes. We build multi-store, multi-channel dashboards that pull Shopify, Amazon, Allegro and the other channels next to each other, so MER, margin and channel mix are calculated from real numbers instead of estimated in a spreadsheet.
Our Baselinker export has broken image links and duplicate SKUs. Can you clean it?
This is routine work for us. Dead image URLs, hotlink-blocked images, duplicate and mismatched SKUs, missing EANs and inconsistent attributes get filtered and fixed before any bulk operation, because a bulk upload of bad data is far more expensive than the clean-up.
How do you handle a catalogue of 30,000 or 40,000 SKUs?
With scripts, not spreadsheets. We have scraped, normalised and structured parts catalogues at that scale, generated search-structured titles automatically and pushed them to store and marketplaces through the API. Manual work is reserved for the exceptions the rules cannot cover.
Do you do AI automation for operations?
Yes, where it earns its place: product content at catalogue scale, draft customer replies, listing quality checks, advertising reports and morning operations briefs. Every automation has approval steps and a tuning period, and we run the same automations on our own operation first.
Do we keep access and ownership of the data and integrations?
Always. Accounts, API keys, dashboards and documentation are in your name. We design integrations so that another engineer can maintain them, and we say so in writing.
How is systems work priced?
Scoped projects with a fixed price, or the Dev Block for smaller jobs. Tool subscriptions such as Elevar or PIM licences are paid by you at cost and never marked up on our invoice.