Enterprise Intelligence Group
Category managers don't have an information shortage. They have a retrieval problem: the answers exist somewhere in the business, but finding them takes longer than using them. Here's the fix.
Category managers don't have an information shortage. They have a retrieval problem: the answers they need exist somewhere in the business — in last year's range review deck, a supplier email thread, a planner's spreadsheet — but finding them takes longer than using them.
Watch a category team prepare for a review and you'll see it. The performance data comes out of the system in minutes. Then the real work starts: digging for why the range changed last time, what was agreed with the supplier in March, which store formats got the exception, what the promotion actually returned. Hours of hunting for knowledge the business already paid to create.
Where the time actually goes
The information a category decision needs is scattered by default:
- Decks — the reasoning behind past decisions, one review per file, named inconsistently. - Spreadsheets — the working models, living in personal folders, versioned by initials. - Email and Teams — supplier agreements, exception approvals, "as discussed" confirmations. - People — the judgment and history that never made it into any document at all.
None of this is anyone's fault. It is what happens when knowledge has no home, no owner and no structure — every function creates intelligence daily, and almost none of it lands anywhere findable.
The visible cost is time: senior commercial people doing document archaeology instead of trade strategy. The bigger costs are quieter. Decisions get made without the history ("we tried that in 2023") because the history couldn't be found in time. New category managers take years to get productive because the knowledge they need is in other people's heads and inboxes. And every AI initiative inherits the mess: point an assistant at fragmented, conflicting files and it retrieves fragments and conflicts.
The fix is structural, not another folder
Better folder discipline doesn't survive contact with a trading week. What works is treating category knowledge as a governed asset:
1. Map the decisions the category team makes repeatedly — range, delist, promote, exception. 2. Consolidate the knowledge each decision needs — out of decks, threads and personal drives. 3. Validate and assign owners — one authoritative version, someone accountable for keeping it true. 4. Then add retrieval — search and AI assistants over the governed foundation, so the answer to "what did we agree with this supplier?" takes seconds, with the source cited.
That last step is where AI genuinely earns its keep: a category assistant grounded in validated knowledge turns preparation from hours of hunting into minutes of checking. But it only works because of the three steps before it.
Frequently asked questions
How much time do category managers spend searching for information?
It varies by retailer, but the pattern is consistent: in judgment-heavy work, preparation is dominated by finding and reconciling information rather than analyzing it. A simple time-audit of one review cycle gives you your own number in a week — and it is usually the most persuasive slide in the building.
Won't better search software fix this on its own?
The EIG Library · Insights · AI Maturity Assessment · Contact