AI lifts sales in sari-sari stores: what sales pros can learn from Cebu's micro-retail surge
AI is starting to change neighborhood retail. New data from Packworks.io shows sari-sari stores using its AI-driven insights posted a 46% bump in daily GMV within two weeks (after September 2025 data collection), and a 17% rise in total sales.
The punchline: small stores used data to make fewer mistakes. Better inventory choices, smarter product mixes, and focused demand planning delivered more sales in less time.
Key numbers sales teams should care about
- +46% daily GMV after two weeks of acting on AI insights.
- +17% total sales in the same window.
- 20% fewer selling days (from five to four days over two weeks) yet higher earnings.
- Gains came from underperforming SKUs flagged by the system-owners moved slow stock and improved turnover.
How the system worked
Packworks' Store Insighting Project (SIP) turns each store's transaction history into store-specific actions. Think: reorder rules, product mix adjustments, and micro-promotions based on what actually sells.
The tech stack behind it was funded by DOST-PCIEERD's Startup Grant Fund in 2024, trained at ST Telemedia Global Data Centres Philippines' AI Synergy Lab, and supported by Ateneo's Business Insights Laboratory for Development for data warehousing and BI.
"Even at this early stage, we've recorded increased sales and enhanced operational efficiency from stores using our AI tools," said Andoy Montiel, Packworks' chief data officer. "As stores learn to leverage SIP's recommendations, micro-retailers can make smarter decisions that translate into higher sales and more efficient operations."
Why this matters for sales
Sari-sari stores operate on thin margins and unpredictable demand-the toughest testbed. If AI-backed decisions can lift GMV while cutting selling days, bigger teams with more data should see equal or better gains.
The takeaway for sales leaders: stop guessing. Swap anecdotal bets for SKU-level actions driven by transaction data.
Steal this playbook
- Prioritize underperformers with potential. Identify SKUs with decent margins, stable supply, and low sell-through. Push visibility, trial promos, and bundles to move them.
- Focus on "open-hour ROI." Concentrate staff and promos on peak windows. Trim low-yield selling hours or days to reduce costs without losing sales.
- Set simple reorder logic. Reorder when days-of-stock drops below a threshold tied to lead time and demand volatility. Prevent both stockouts and dead stock.
- Run micro-promotions. Use small, time-bound offers to trigger repeat purchases on SKUs the model marks as lift-ready.
- Bundle for velocity. Pair a hero SKU with a flagged slow-mover at a slight discount. Track basket size and margin contribution, not just top-line.
- Refresh shelf placement. Move priority SKUs to eye-level or first-touch zones. Measure before/after conversion, not just foot traffic.
- Weekly review cadence. Every 7 days: review sell-through, stockouts, and promo lift. Keep what works, drop what stalls.
Metrics to track (and act on)
- GMV per open hour/day
- Sell-through rate by SKU/category
- Stockout rate and days of inventory
- Contribution margin by SKU and bundle
- Promo lift (incremental units and margin)
- Repeat purchase rate on targeted SKUs
Context: adoption gap = opportunity
Government programs push AI into MSMEs, yet adoption remains low. Only 14.9% of local firms use AI tools, even as estimates suggest trillions of pesos in value by 2030. That gap is your edge if you move first.
For background on local AI adoption, see PIDS. If you want practical upskilling paths by job function, browse curated options here: Complete AI Training.
Bottom line
Data-backed recommendations beat gut calls. Start with underperforming SKUs, tighten open-hour ROI, and audit results every week. The stores that execute simple, measurable changes win-regardless of size.
Your membership also unlocks:
