CatalystRetailCPGAssortment Optimization Use Case
Eight weeks out from the season, the merchandising planner is staring at a buy sheet built on store-group averages and last year's sell-through.
Ranked in dollars
·What-if validation
·Push to ops
Diwo Playbook · Catalyst
Catalyst PlaybookVol. 2375
Assortment Optimization Use Case
RetailCPG
Sample prompts
- 01How many units of this SKU should I buy for each store and channel this season?
- 02Where is my planned assortment diverging from True Demand, and by how much?
- 03What's the expected lift if I shift units from store group A to group B?
- + 3 more inside
Signed · Diwosha256:assortmentop
01The Problem
Today’s workflow is the bottleneck.
Eight weeks out from the season, the merchandising planner is staring at a buy sheet built on store-group averages and last year's sell-through. Finance already signed off on the top-line dollars. But she knows the coastal stores don't behave like the interior ones, and marketing's new campaign calendar just landed in her inbox this morning — too late to factor in. The PO goes out Friday anyway, with the same blunt allocation that left her marked down in two regions last fall.
The status quo· typical decision cycle
signal decays- Day 1Signal capturedModels score. Data is fresh.
- Day 2–3Dashboard builtAnalyst pulls CSVs, joins sources
- Day 4Review meetingStakeholders ask for context, re-pull
- Day 5+Window has closedSignal stale, action wasted
By the time the action is ready, the window has closed.
5 days between signal and action. The data science team did their job. The operator is still waiting.
02The Approach
How Catalyst handles it.
Catalyst doesn't hand the planner another top-down forecast. It lets buyers work at the SKU-Store/Channel level, determining optimal purchase quantities and prices needed to satisfy True Demand before the season begins. The Recommendation First System compares planned assortment against demand and surfaces the divergences that matter, with marketing inputs folded in before the commitment is made. A conversational interface lets the planner run quantified what-ifs — shift units from group A to B, flex price on this SKU — and see the revenue impact. A Semantic Knowledge Graph ties structured sales data to unstructured context, and continuous learning sharpens next season's recommendations. The buy goes out at SKU-Store precision, not store-group averages.
Catalyst · Conversation
A
How many units of this SKU should I buy for each store and channel this season?
Catalyst · 320ms · sources: 4
Here’s what I found — three drivers explain most of the signal, and I’ve ranked them by impact.
RECOMMENDATION
Take action on the top-ranked driver first.
Expected lift: +9.5% · 2-week window.
See this live
Watch Catalyst solve assortment optimization use case on your retail stack.
45-minute working demo. Your data, your question, a real answer — not a pre-recorded walkthrough.
03Ask
Questions you can ask.
Every case ships with a set of high-leverage prompts — the shortlist operators reach for every week. Here are the ones we see working for assortment optimization use case.
Anchor question
01 · start here
How many units of this SKU should I buy for each store and channel this season?
This is where most
operators start.
02
Where is my planned assortment diverging from True Demand, and by how much?
03
What's the expected lift if I shift units from store group A to group B?
04
Which SKUs am I at risk of over- or under-buying before the PO goes out?
05
How should next month's marketing calendar change my pre-season buy plan?
06
If I flex price or quantity on this SKU, what happens to revenue?
Related cases
Same playbook, other shapes.
Catalyst
Campaign Optimization Use Case
The marketing lead at a mid-size card issuer just got the monthly life-events refresh — thousands of cardholders who relocated, married, or changed jobs in the last 30 days.
Financial Services·Retail·CPG·Cross-industry
Catalyst
Payment Processing Optimization
Monday morning. Your regional finance head opens the weekend PoS report and sees a decline-and-latency spike concentrated in one state.
Retail·Financial Services
Catalyst
Procurement Optimization — Purchasing Harmonization Savings
Tuesday afternoon. Your sourcing lead has three vendor spec PDFs open side-by-side, trying to confirm whether the bracket from Supplier A is functionally the same part as the one Supplier B ships at a different price.
Manufacturing·Retail·Healthcare·Cross-industry
See it on your data
Bring a real Retail question. We’ll show you the decision.
We’ll run Catalystagainst a slice of your own data during the demo — no slideware, no prerecorded mock. You leave with a working decision and a line of sight to the next one.