Core is what the product is. Contextual is what the market says about it. Custom is the part only you can write: your segments, your thresholds, your naming. We apply it identically to every UPC, re-apply it every run, and keep it on the same record as everything else.
Most custom segmentation lives in one analyst's file and drifts the moment somebody else maintains it. Here it is a rule on the record, applied the same way to every UPC and re-applied every run.
Your department, category and segment naming, mapped onto every product we hold. Reports stop needing a translation layer between our taxonomy and the one your business actually runs on.
Better-for-you, premium, value, single serve. You set where the cut falls, we apply it across the whole category, and a product moves in or out of the set the run its facts change.
Own brand against national brand, a demand space that spans three aisles, a watchlist. Anything a team currently keeps by hand becomes a field on the record that everyone reads the same way.
Most category debates are not about the numbers. They are about whose definition of the set is the right one. This turns that into a setting rather than a standoff.
Run the review on your own segment structure and your own thresholds, against the same product record everyone else in the room is reading. The set stops being the thing under debate.
Cut the category the way the strategy actually frames it, including groupings that span aisles and appear in no standard hierarchy anywhere.
Stop maintaining a mapping table that drifts between refreshes. The definition lives with the record, versions with it, and arrives already applied to every UPC in the feed.
A Fortune 100 CPG was coding its product file by hand, and coverage stalled near 68 percent. A custom catalog now codes the whole file to their own taxonomy every cycle, including the 26,185 products no one had ever reached.
Custom is the third layer. Core holds what the product is, Contextual holds what the market says about it, and Custom holds the structure you define. Every attribute is coded into exactly one of the three, so the layers add up to the whole record and nothing is counted twice.
The same record, delivered three ways. Most teams start with one and add another once the record is in their stack.
Custom attributes are the part of the record only your team can write. Once your definitions live on it, every report, every system and every argument starts from the same set.
One consistent reference for product attributes and language, across every system and category.
Thirty minutes on your categories. How the record gets built, what it holds, and the questions your team could put to it.