Walmart M5 Data Platform Step 12 report · Stakeholder briefing
Step 12 · For leadership

What the numbers say: decisions for the next planning cycle

A non-technical summary of validated findings from the Executive, Store, Product, and Demand dashboards. Numbers below come from the Gold export and the Power BI pages already reviewed in steps 08–11.

No pipeline jargon Findings from 08–11 Planning & decisions
Network sales
$191.58M
67M units · 10 stores
FOODS share
58%
$111.1M of sales
Flagship store
CA_3
17.1% of network · $32.70M
Full-year sales growth
~9%
CAGR 2012→2015

1. Bottom line for the boss

Demand is healthy when you read it on a daily basis. Average daily units rose from about 37.8K in 2015 to about 41.8K in 2016 year-to-date (through 22 May). Comparing the same calendar window (1 Jan–22 May), 2016 sales are about 18% higher than 2015. The business is concentrated: California is ~45% of sales, one store (CA_3) is the clear flagship, and FOODS is still the revenue engine at roughly 58% of sales.

The main planning risk is reading partial years as a crash: 2016 only has 143 days in the training window, so annual totals look small even while per-day demand is strong. Use same-window year-over-year and average daily metrics for leadership reviews.

2. Key findings (from dashboards 08–11)

Network & growth (Executive)

Stores & geography (Store)

Products & assortment (Product)

Demand & calendar (Demand)

3. Outlook: what to expect next

These are planning implications from the observed history — not a separate forecast model. Use them to set expectations and capacity, then refresh after each Gold / Power BI update.

Signal in the dataPlanning implication
Avg daily units 2015 → 2016 YTD: ~37.8K → ~41.8K Baseline demand planning should assume continued strength on a per-day basis, not a 2016 “collapse” from partial-year totals.
Same-window sales +18% (Jan–22 May 2016 vs 2015) Near-term reviews should use comparable calendar windows; full-year cards alone will mislead until a complete year is available.
Full-year sales CAGR ~9% (2012–2015) Reasonable multi-year planning band if assortment and store coverage stay similar — re-check after each annual close.
FOODS ~58% sales; Top 15 all FOODS_3 Most forecast error and stock risk will sit in FOODS_3 head SKUs; protect service levels there first.
HOUSEHOLD share growing while FOODS share eases Keep FOODS as the engine, but allocate some assortment and promo attention to HOUSEHOLD growth stores (e.g. CA_2 mix).
Event days weaker on units/day in this calendar Do not assume holiday labels automatically lift volume; build baselines on normal days and treat Dec 25 / closures as special cases.

4. Decisions and actions to take

  1. Store ops: Treat CA_3 as the benchmark for staffing, replenishment, and assortment. Open a focused review on CA_4 (lowest sales) and CA_2 (highest HOUSEHOLD mix).
  2. Category investment: Put the majority of demand-planning and inventory effort on FOODS, especially department FOODS_3 head SKUs (FOODS_3_090, FOODS_3_586, and the rest of the Top 15).
  3. Leadership metrics: Report growth with average daily units/sales and same-calendar-window YoY. Keep the all-years YoY card (~27%) as context only — it mixes uneven year edges.
  4. Promo & calendar: Plan from non-event baselines. Interpret labeled events with units/day, not stacked totals. Always exclude or specially model 25 Dec in forecasting baselines.
  5. Pricing watch: Use the department price table for assortment reviews — watch HOUSEHOLD_2 high-price outliers while remembering they are not the volume engine.
  6. Operating cadence: After each Gold refresh, re-open the four Power BI pages and this brief; the detailed validation pages (08–11) remain the audit trail if numbers move.

5. Suggested 90-day plan

HorizonOwner focusSuccess look
Next 30 days Confirm FOODS_3 head-SKU availability and CA_3 replenishment rules; brief leadership on same-window YoY only. No surprise stockouts on Top 15; leadership pack uses daily / comparable YoY language.
30–60 days CA_4 and CA_2 deep-dives (volume gap vs mix shift); align promo calendar to non-event baselines. Written store actions; event plans do not inflate expected holiday units without evidence.
60–90 days Re-forecast FOODS capacity using latest rolling 7D/28D; review HOUSEHOLD growth share store by store. Updated demand plan for next season; HOUSEHOLD opportunities listed where mix already supports them.

6. Where to dig deeper

This page stays at business language. Technical validation, DAX, and screenshots live on the analysis pages:

08 Executive · 09 Store · 10 Product · 11 Demand · Overview

Live dashboard: Open Power BI ↗