Product data and PIM
Product data that is the same everywhere
The shop says one thing, the marketplace says another, and the catalogue has a third version. That is not a tidiness problem, it is an ownership problem: nobody can say which source is right. I build the source that is right.
How you know the product data is the problem
An item is called one thing in the shop and another in the marketplace export. A measurement is in centimetres here and millimetres there. A product has not been updated in English for two years. A new model takes three weeks to appear everywhere, because three people type it into three systems.
The common factor is not carelessness. Nothing owns the data. Every system has its own version of the truth, and whoever makes a change makes it wherever they happen to be working.
More discipline does not fix that. One place owning the product data, and everything else taking it from there, does. That system is called a PIM, for product information management.
The tool I reach for is Akeneo, an open-source PIM and the one I have worked with longest. It is a recommendation, not a condition.
What that looks like in practice
A PIM as the one source
Product data lives in one place: attributes, copy, images, translations. A change happens there, and the shop takes it from there. No spreadsheet travelling by email.
Connected to shop and ERP
The PIM connects to the shop and to your inventory system. Prices and stock still come from the ERP, product information from the PIM. Every figure has exactly one owner.
Channels and languages
One product, several outputs: a short text for the marketplace, a long one for the shop, its own images per country. In the PIM that is a setting, not a second record.
Completeness you can see
You can see which products are ready and which are not, before they go live. Missing images and empty attributes are a list, not a discovery in the shop.
How a project starts
-
Take stock
We look at where product data lives today and who changes it. That usually takes half a day and regularly turns up systems nobody had thought about in years.
-
The data model
Which attributes actually exist, which are legacy, what belongs together. This is the work that decides whether the project succeeds, and it happens before anything is installed.
-
Import and connect
Existing data moves into the PIM and the shop takes it from there. From this point there is one source.
-
Running it
New attributes, new channels, new languages. A PIM is not something you set up once and forget.
From projects
-
Several brand shops under one roof, product data in Akeneo, the ERP connected, and the shop joined to the existing CMS over GraphQL. An ongoing partnership since 2016.
Frequently Asked Questions
Do I even need a PIM?
If you have one channel, one language and a few hundred items, no. A PIM earns its place when product data is maintained in more than one place, or when several channels need different versions of it. Tell me your numbers and I will tell you honestly whether it pays.
Does it have to be Akeneo?
No. Akeneo is the system I have worked with longest and know best, so it is my default recommendation. If you already run a different PIM, connecting it to the shop is the same work.
What happens to our existing data?
It comes with you. The import is rarely the problem; the data model is. That is why it sits before the installation in the sequence above and not after it.
One Person, From the Question to the Code
I work out what is needed, write the requirement, organise whoever else works on it, and then build it myself. Usually that is three companies, and something is lost at every handover between them. Here there is nobody between your question and the running system.
Related services
Let us look at your product data
30 minutes, no strings attached. Describe where the data lives today and I will tell you whether a PIM is the answer or whether something smaller will do.