One source
Product data lives in one place. The site, the PDF catalog, the quoting screen and search all feed from it. Fix one dimension and all four are fixed. Nobody has to ask which one is current again.
Services
Product catalog & data engine
You have hundreds of products. The catalog is a PDF, prices live in Excel, spec sheets sit in a folder. Each one is updated separately.
As the product count grows, the data spreads across four or five places: the description on the site, the catalog PDF, the sales spreadsheet, the production spec sheet, and the English file for export. When one dimension changes, all five should change. Usually two do. Then one day a customer orders from the figure on the site, and production builds to the figure on the spec sheet.
Gain: One source. Catalog, site and quotes finally agree with each other.
single source of truth · multilingual · site + PDF + search from one placeWhat we build
Gathering the scattered data is our job
If your data is in spreadsheets, in the PDF of an old catalog and in a few people's folders, that's fine. Collecting, cleaning and surfacing the contradictions is our work. When the site says one dimension and the catalog says another, we don't guess: the contradictions come to you as a list.
Images matched to products
Archive photos rarely line up with products: the filename says something else, the same product has three different names, one model has no photo at all. We do the matching and separate the certain from the doubtful. What doesn't match isn't hidden; it's listed.
Multilingual, with the gaps visible
The same product data keeps the same shape in every language. A field added in one language doesn't quietly stay empty in another; the gaps come out as a report. You see the blank cell in the export catalog before your customer does.
A new channel doesn't mean retyping
When a dealer portal, a marketplace feed or a new language shows up tomorrow, nobody re-enters the data. The source stays the same and the new channel connects to it.
How the work goes
- 01
Inventory
How many products, how many fields, how many separate sources. The size of the job is set by counting, not guessing.
- 02
Collecting and normalizing
The same field from different files moves into one structure. Units, decimal separators and naming are brought to a single form.
- 03
Contradiction report
Values that disagree are listed, and you make the call. Skipping this round is how a catalog spreads one error across four channels at once.
- 04
Wiring the channels
The site, the PDF catalog, the quoting screen and search are wired to the single source. There's one path for updates.
What comes with it
- A single source for product data, with an admin screen
- Collection and cleanup from scattered files
- A report of contradictions and gaps
- Image matching, with the doubtful ones kept separate
- Wiring to the site, the PDF catalog and search
What we don't do
- We don't invent data you don't have. A missing technical value isn't made up; it stays empty and gets listed.
- The first collection round takes your time too. Only someone who knows the product can settle a contradiction.
- The graphic design of the catalog is separate work. We build the data and the generation; the page design is a separate conversation.
Seven sources into one structure
In our own product, the Real-Estate Toolkit, we brought data from seven separate official public sources into one structure: 389,000 points of interest and 149,000 zoning plans answer a single query. A product catalog is the same job; only the name of the data changes.
Data and security
Your documents stay on your own server and are never sent to train an outside model. We decide together, during setup, who gets access to what.
Let's figure out where to start, together.
A conversation to understand where you are and decide the first step together. We'll talk as long as you need.
Let's talk