Handling Forecast Exceptions: Real-World Situations
Sometimes demand doesn't follow the forecast. Here's how to handle the most common situations.
Scenario 1: Demand Spike (Unexpected +50%)
What Happened
Normal sales: 100 units/day
Actual sales: 150 units/day (+50%)
Reason: Viral TikTok post about your product
Immediate Actions (Today)
Step 1: Create Emergency Order
├─ Go to: Purchase Orders → New PO
├─ Select: Your fast supplier
├─ Product: The viral item
├─ Quantity: 200-300 units (3 days stock)
├─ Lead time: 3-5 days (express shipping)
├─ Send ASAP
└─ Aim: Arrive in 5 days (cover the spike)
Step 2: Notify Team
├─ Sales: "We're in demand surge, low stock risk"
├─ Marketing: "Pause additional promotion if running"
├─ Customer Service: "Prepare for high volume"
└─ Goal: Manage customer expectations
Short-Term Actions (Next 3-5 Days)
Step 3: Update Forecast
├─ Go to: Demand → Select product
├─ Current month: Increase forecast by 50%
├─ Next month: Keep elevated (spike may last 2-4 weeks)
├─ Reason: "Viral TikTok post [link]"
└─ Supply plan recalculates
Step 4: Create Additional Orders
├─ Check supply plan: What does it recommend?
├─ Create PO for recommended quantity
├─ Use fast suppliers (higher cost OK in shortage)
└─ Goal: Rebuild inventory as fast as possible
Ongoing (Next 2 Weeks)
Step 5: Monitor Trend
├─ Daily: Check actual sales vs. forecast
├─ Question: Is this temporary spike or new normal?
├─ Data point: "Spike lasted 2 weeks, then normal"
└─ Learning: "Viral marketing = predictable 2-week surge"
Step 6: Adjust for Future
├─ If spike fades: Return forecast to normal
├─ If spike sustains: Keep elevated forecast
├─ Document: "Spike pattern" for next viral moment
Real Result
Day 1: Viral post, sales surge to 150/day
Day 2: Emergency order placed (300 units, express)
Day 4: Emergency goods arrive, crisis averted
Day 5-10: Maintain elevated forecast, standard orders arrive
Day 15: Spike ends, sales return to 100/day
Day 16: Forecast adjusted back to normal
Scenario 2: Demand Drop (Unexpected -40%)
What Happened
Normal sales: 100 units/day
Actual sales: 60 units/day (-40%)
Reason: Competitor launched better product, or seasonal transition faster than expected
Immediate Actions (Today)
Step 1: STOP All Orders
├─ Check: Any POs currently outstanding?
├─ Review: What arrives in next 2-4 weeks?
├─ Decision:
│ ├─ If PO not yet shipped: Can you cancel/reduce?
│ ├─ If PO already shipped: Accept incoming inventory
│ └─ Contact supplier TODAY
└─ Goal: Don't order more when demand dropped
Short-Term Actions (Next 3-5 Days)
Step 2: Update Forecast
├─ Go to: Demand → Select product
├─ Current month: Decrease forecast by 40%
├─ Next 2 months: Keep reduced
├─ Reason: "Competitor launched cheaper option"
└─ Supply plan recalculates
Step 3: Assess Overstock Risk
├─ Go to: Supply Plan tab
├─ Look for: "Overstocked" status
├─ If yes: You have problem
│ ├─ On-hand inventory will build
│ ├─ Capital tied up in excess stock
│ └─ May become obsolete
└─ If no: You're OK (forecast drop prevents overstock)
Step 4: Pause Marketing (Optional)
├─ If product declining: Don't add fuel
├─ Pause paid ads (save money)
├─ Let inventory sell down naturally
└─ Goal: Don't chase dead demand
Ongoing (Next 2-4 Weeks)
Step 5: Monitor Trend
├─ Is it temporary dip or permanent drop?
├─ Watch data: 1-2 weeks needed to confirm
├─ Possible scenarios:
│ ├─ Competitor is temporary → Sales recover
│ ├─ Seasonal ending → Expected, plan for off-season
│ └─ Product decline → May need to discontinue
Real Result
Day 1: Sales drop to 60/day, notice issue
Day 2: Cancel 2 pending POs (save 200 units, $10K)
Day 3: Update forecast down 40%
Day 7: Actual sales confirm trend (still 60/day)
Day 14: Inventory stable at 150 days on hand (healthy)
Day 30: Either recover (sales return to 100) or new normal
Scenario 3: New Product Launch (No Historical Data)
What Happened
Launching: New product you've never sold
Challenge: No sales history to forecast from
Question: "How much should I order?"
How to Estimate
Method 1: Use Comparable Product
├─ Similar product you already sell?
├─ Example: "New Blue T-Shirt variant"
├─ Use: That product's historical data
├─ Adjust for differences:
│ ├─ New product cheaper? Might sell +20%
│ ├─ New product premium? Might sell -20%
│ └─ Same category? Use baseline
└─ Result: Estimated forecast
Synplex Setup
Step 1: Manually Set Initial Forecast
├─ Go to: Demand tab → New product
├─ Month 1-3: Enter conservative estimate
├─ Example: 30 units/month (conservative for similar product)
├─ Reason: "New product, estimated based on [similar product]"
└─ Supply plan generates orders for 30 units
Step 2: Launch Product
├─ Order 30 units
├─ Arrive, sell in store
├─ Actual sales: Let's say 25 units (better than feared!)
Step 3: Adjust After 4 Weeks
├─ Actual data: 25 units sold
├─ Revised forecast:
│ ├─ Month 2: 25 units (actual proved estimate)
│ ├─ Month 3: 30 units (slight growth expected)
│ ├─ Month 4: 35 units (momentum building)
│ └─ Reason: "Launch data, 25 units actual, growth assumed"
└─ Supply plan now recommends 25 units
Step 4: Rinse & Repeat
├─ Each month: Compare forecast to actual
├─ Adjust next month based on real data
├─ After 3-6 months: Have real pattern
└─ Forecast becomes accurate
Real Timeline
Month 1: Forecast 30, Actual 25 (83% accuracy, OK)
Month 2: Forecast 25, Actual 28 (112%, higher than expected!)
Month 3: Forecast 30, Actual 32 (107%, growing)
Month 4: Forecast 35, Actual 36 (103%, consistent)
Month 5: Forecast 38, Actual 37 (97%, stabilized)
By Month 6: Forecast accurate within 5%
Scenario 4: Seasonal Product (Winter Jacket)
The Challenge
Winter Jacket seasonal pattern:
├─ Oct-Feb: 2-3x normal (peak season)
├─ Mar-May: 1x (transition, declining)
├─ Jun-Aug: 0.3x (off-season, minimal)
├─ Sep: 0.5x (transition, rising)
Synplex baseline forecast: 100 units/month (ignores seasonality)
Problem: Wrong quantity for every month!
Solution: Seasonal Adjustments
Step 1: Define Your Seasonal Pattern
├─ Using past 2 years data:
│ ├─ Jan-Feb average: 500 units/month (2x baseline 250)
│ ├─ Mar-May average: 250 units/month (1x baseline)
│ ├─ Jun-Aug average: 75 units/month (0.3x baseline)
│ └─ Sep-Dec average: 300 units/month (1.2x baseline)
└─ Baseline: 250 units/month (annual average)
Step 2: Manual Adjustment in Synplex
├─ Go to: Demand tab → Winter Jacket
├─ Jan: 500 units (2x baseline)
├─ Feb: 500 units (2x baseline)
├─ Mar: 300 units (1.2x baseline)
├─ Apr: 250 units (1x baseline)
├─ May: 250 units (1x baseline)
├─ Jun: 75 units (0.3x baseline)
├─ Jul: 75 units (0.3x baseline)
├─ Aug: 75 units (0.3x baseline)
├─ Sep: 125 units (0.5x baseline)
├─ Oct: 300 units (1.2x baseline)
├─ Nov: 400 units (1.6x baseline)
└─ Dec: 400 units (1.6x baseline)
└─ Reason: "Seasonal pattern, 2+ year history"
Step 3: Supply Plan Now Shows Reality
├─ Jan: 500 units = big order in Dec (arrive by Jan 1)
├─ Jun: 75 units = small order in May (arrive by Jun 1)
├─ Oct: 300 units = medium order in Sep (arrive by Oct 1)
└─ Much better than 100 units every month!
Step 4: Review Annually
├─ Each January: Did the pattern hold?
├─ If yes: Keep pattern, minor tweaks
├─ If no: Market changed (competitor, trend), adjust
└─ Goal: Continuous refinement
Real Supply Impact
WITHOUT seasonal adjustments:
├─ Order 100 units every month
├─ Jan: Crisis! Need 500, only have 100
├─ Jun: Disaster! Order 100, sell only 75
├─ Result: Stockouts + Overstock = Chaos
WITH seasonal adjustments:
├─ Jan: Order 500 units (arrive by Jan 1, perfect!)
├─ Jun: Order 75 units (arrive by Jun 1, no overstock!)
├─ Result: Healthy inventory year-round
Scenario 5: Emergency Order (Fast Supplier Only)
What Happened
Your main supplier: 45-day lead time
Problem: Inventory hits critical level in 10 days
Question: "We'll stockout before main order arrives. What now?"
Emergency Supplier Strategy
Step 1: Identify Emergency Supplier
├─ Who's your secondary/backup supplier?
├─ Lead time: 5-7 days? (vs. main supplier 45 days)
├─ Cost: Higher (expedite fee, express shipping)
├─ Quality: Same or acceptable?
└─ Keep list ready for moments like this
Step 2: Create Emergency PO TODAY
├─ Go to: Purchase Orders → New PO
├─ Supplier: Emergency supplier
├─ Product: Critical item (what's running out?)
├─ Quantity: 100-150 units (3-5 days supply)
├─ Lead time: 5-7 days
├─ Cost: Accept premium (maybe 15% higher)
└─ Send immediately
Step 3: Create Main Order (Backup)
├─ Don't cancel main supplier order!
├─ Why? Delivery times vary
├─ Two orders arriving in different times is OK
├─ One fails? Other covers you
└─ Both arrive? You have excess (temporary, but manageable)
Step 4: Update Forecast
├─ Go to: Demand → Product
├─ Emergency was temporary measure
├─ Don't change baseline forecast
├─ Reason: Forecast driven by actual demand, not supply crisis
└─ Supply plan reflects new orders (will show healthy inventory)
Step 5: Prevent Next Time
├─ After crisis resolves:
│ ├─ Lead time setting: Was it accurate?
│ ├─ Safety stock: Was it too low?
│ ├─ Supplier: Should we move faster supplier to primary?
│ └─ Action: Fix so it doesn't happen again
Cost Calculation
Emergency order cost:
├─ Quantity: 150 units
├─ Normal cost: $20/unit = $3,000
├─ Emergency premium: +15% = $450 extra
├─ Total cost: $3,450
├─ Extra carrying: $225 (inventory buildup)
├─ Total emergency cost: $675
Cost of NOT ordering (stockout):
├─ Lost sales: 3 days × 50 units/day = 150 units
├─ Revenue lost: 150 × $50 = $7,500
├─ Profit lost (40% margin): $3,000
└─ Cost of stockout: $3,000
Decision: $675 emergency cost << $3,000 stockout cost
Result: Emergency order was the right call
Common Question: When to Use Each Strategy
| Situation | What to Do |
|---|---|
| Sales +50%, expected to last 2 weeks | Emergency order + adjust forecast up 2 weeks |
| Sales -40%, permanent competitor threat | Stop orders immediately + adjust forecast down |
| New product launching | Estimate conservatively, adjust weekly based on data |
| Winter Jacket Dec spike | Seasonal adjustment, orders placed months in advance |
| Inventory hits critical in 5 days | Emergency supplier, fast shipping |
| Promotion planned for next month | Adjust forecast up now, orders reflect recommendation |
General Rules
Rule 1: Act Fast
When exception happens:
├─ Hour 1: Identify the issue
├─ Hour 2: Decide on action
├─ Hour 3: Execute (order, adjust forecast, notify team)
└─ Delay = Worse outcomes
Rule 2: Communicate
Tell your team:
├─ Sales: "Inventory situation"
├─ Suppliers: "Adjusting orders, here's why"
├─ Finance: "Budget impact"
└─ Everyone: Better decisions when informed
Rule 3: Document
In Synplex, always write reason:
├─ "Viral post, demand spike"
├─ "Competitor launched, sales -40%"
├─ "Seasonal peak Oct-Dec"
├─ Why? Learning for next time
Rule 4: Review & Learn
After exception resolved:
├─ What went right?
├─ What could be better?
├─ How to prevent next time?
├─ Example: "Add 2-week buffer before peak season"
Next Steps
- Identify your exceptions: What usually surprises you?
- Prepare responses: Have backup suppliers ready?
- Set up alerts: Does Synplex notify you of issues?
- Test: Create a small emergency order to practice process
- Document: What's your backup plan?
Related
- Adjusting Forecasts — How to override predictions
- Supply Plan Overview — Understanding recommendations
- Folder 04: Creating POs — How to execute orders
Questions? Contact support@synplex.io