How an information architectural overhaul improved visual clarity and guided automated processes, reducing error and cutting down on production time.

Introduction

Stock catalogues are a vital part of the B2B sector, presenting to our customers the styles for a given season. However, producing them historically involved a workflow that was likely to produce errors.

This workflow involved transforming internal documents generated by sales to produce the final booklet for customers. These often came in parts of multiple documents, and with an additional list of unlisted styles to be pulled from previous documents. This process took weeks and, more glaringly, was subject to error as it was easy for items to get lost, causing further delays in production. This is on top of the fact that we were (and still are) managing two brand names nationwide, for both English and French markets.

When it came time for the next season, our team decided to introduce automations into the production cycle, and a complete overhaul of the look and feel of the stock catalogues.

The Information Architecture

Before even beginning to think about the process, we first had to start with an information architecture. This process involved studying what the current tiles are communicating, and simplifying it into something the end consumer will understand.

Studying the layout of the existing document

The product tiles displayed a lot of internal information, encoded with cryptic codes and without any indication of what said codes stood for. It also was missing some information we felt was important, such as product features, something we pride our brand on. Much of the existing information, we determined, would be irrelevant to the customer, thus every field was either simplified or tossed entirely.

This new information architecture was designed for clarity. It clearly laid out all the technical features, and reduced the number of labels on the product tiles. It also more clearly indicated what styles were carryovers, and what was new. With this information, we were now ready to start working.

Automation

Once we determined our IA, we set out to build our automation pipeline. The key here is that all required metadata is pulled from a single master file and a premade list of styles, and with our product shots originating from our digital asset library.

We built an automation workflow that can be run completely on a local PC utilizing InDesign data merge with images, and further Adobe CC scripts. A freely available Photoshop script was used to resize all images to their correct dimensions. A product tile template was built once, then all data and images merged via Data Merge. Another freely available script handled translation of common phrases into French for a French-language booklet, further cutting down on production time.

This had a few effects on the workflow. First, all product information came from a single source of truth. The new data merge-based file preserved the necessary information from the master file, keeping only the columns that were relevant for the merge operation. Part of the schema’s design also contained fields that would be used purely for filtering data. This meant that none of the data was directly manipulated.

Second, the new automation allowed for more verification checks between steps, ensuring no items get lost, and allowing for sudden changes in the product lineup while minimizing impact on production time. For instance, if an error was caught where products were missing in the metadata file, the list would simply be regenerated with the new filters. If an error was caught in booklet production for similar reasons, the impact would be that the affected section is corrected and remerged, and only would take an hour at minimum to correct. This is in contrast to the older workflow, where stock lists would have to be re-rendered and the entire booklet redone, delaying publication by a day.

Booklet Design

The end result of this project was a much clearer catalogue for our customers. For us, it was a faster and more robust system. What would have taken some weeks for our catalogues can now be done in the span of a few days.

Left: old stock catalogue. Right: new stock catalogue