Case Study
Automated Replenishment Reports Saved 120+ Hours Monthly
Cutting analysis time from days to minutes — boosting in-stock rates across Walmart Canada.
While at P&G Canada, Arturo automated replenishment reporting for Walmart, replacing manual weekly reviews with company-wide POS and inventory dashboards. Analysts went from spending 3 days on SKU-store reviews to acting on insights in minutes — saving over 120 hours monthly and raising in-stock rates portfolio-wide.
Key Results
The Transformation
The Challenge
P&G's replenishment team at Walmart Canada was drowning in manual work. Each analyst spent roughly 3 days every week pulling SKU-store fill rate data from Retail Link, formatting it into review templates, and scanning for availability issues across their assigned categories. The team had 8 analysts covering hundreds of SKU-store combinations each, and the manual process meant they could only deeply investigate a fraction of potential issues each week.
By the time they identified a stockout pattern — say, a Pantene SKU consistently running empty at 15 stores in Ontario — it was Thursday or Friday, too late to influence the next replenishment cycle. The delay between data availability and analyst action was costing real revenue: products sitting in backrooms weren't generating sales, and by the time the issue was flagged, the promotional window had often passed. Phantom inventory was a particular blind spot: Retail Link would show stock on hand, but the product was sitting in the backroom, not on the shelf — invisible to shoppers and invisible to the analysts reviewing the data. Nobody had a systematic way to distinguish 'low sales because of low demand' from 'low sales because the product isn't on the shelf.'
Our Approach
We automated the entire extraction-to-insight pipeline. First, we built scripts to pull Retail Link POS and inventory data automatically — no more manual downloads and copy-paste into spreadsheets. The extraction ran overnight so data was ready before analysts arrived Monday morning.
The critical piece was the phantom inventory detection logic. We built algorithms that compared POS velocity against on-hand inventory levels: if a store showed 40 units on hand but sold zero in the past 7 days for a product that normally moves 8 units per week, that was flagged as a likely phantom inventory situation. We also built thresholds that adjusted by category — bulky items like paper towels had different phantom patterns than small items like razor blades.
Each analyst received a personalized Monday-ready dashboard showing their category's health: flagged stores ranked by revenue impact, trending stockout patterns, and phantom inventory suspects. The dashboards were designed around the analyst's actual workflow — not a generic BI tool, but a purpose-built view that answered the specific questions they asked every week. Instead of spending 3 days finding problems, they could spend Monday morning reviewing pre-identified issues and take action the same day — contacting store managers, adjusting forecasts, or escalating to the replenishment buyer.
We also built a weekly summary email that went to category leads highlighting the top 10 issues by estimated revenue impact across all analysts' territories. This gave management visibility into the biggest opportunities without requiring them to open dashboards themselves.
The Outcome
The automation saved over 120 analyst hours per month across the team — roughly 3 full-time weeks of capacity redirected from data wrangling to strategic action. In-stock rates improved by 1% across Walmart Canada, which sounds small until you calculate the revenue impact across P&G's entire portfolio at Canada's largest retailer. At P&G's scale with Walmart Canada, a 1% in-stock improvement translated to meaningful incremental revenue.
On-time deliveries improved by 5% because analysts could flag replenishment issues before they became stockouts rather than reacting after shelves were already empty. The phantom inventory detection alone was worth the project — it surfaced availability problems that had been completely invisible in the manual review process, where analysts could only check what Retail Link's on-hand numbers told them.
But the real transformation was behavioral. Analysts shifted from reactive problem-finders to proactive problem-solvers. They could identify a stockout trend on Monday and have corrective action in place before the next weekend sales peak. The dashboards became the standard operating rhythm for the replenishment team — every Monday started with the automated report. New analysts onboarded faster because the system told them where to focus instead of requiring months of experience to know which stores and SKUs to watch.
Want similar results for your organization?
Get in touchYou May Also Like
How High is High: Breaking the Negative Feedback Loop in Automated Replenishment
Identified critical flaw in Walmart's automated replenishment, developed custom OSA algorithm, and drove $3M incremental revenue across two P&G categories in 4 months.
Read case studyCracking Nielsen's ETL: From 20 Days to 3
Faster POS data: cutting Nielsen ETL from 20 days to 3.
Read case studyReady to turn data into decisions?
Let's discuss how Clarivant can help you achieve measurable ROI in months.