← Size matrix

Satin Bias Skirt

🟡WatchSize set breaking

ST-1012 · Skirts · Bottle green · Satin · Bias cut · Milano Silk · £89 retail at 71% margin

🔥 Trend momentum
59/100

Flat — no clear direction

93% confidence · Medium competition

📊 Demand coverage
102%

Stock matches predicted demand

911 units against 896 predicted over 30 days

🚀 Opportunity score
38/100

Weak — do not add stock

35% demand size · 25% momentum · 20% margin · 20% stock gap − competition

⚠️ Stock-out risk
53%

Watch core sizes

4.4 weeks of cover · sales up 10%

🤖 AI recommendation
🟡 Watch

Signals are early and confidence is 93%. Hold the current buy and re-check next week.

4 actions queued worth £18,439, starting with: transfer 6 units between locations before buying more.

An AI-generated decision-support signal from external trend data and this merchant's own sell-through — not a guaranteed forecast. Nothing is ordered without approval.

Momentum · measured and projected
425772W−5W−4W−3W−2W−1NowW+1W+2W+3W+4
measured AI projection

Signals are steady — no clear peak in the next month.

What drives the opportunity score

The six factors behind the number, each one readable on its own.

✓
Search interest
Rising 13%
✓
Category demand
Increasing 8%
✓
Customer interest
Social mentions up 9%
✓
Competition
Medium — contested
✓
Margin
71% — healthy
–
Stock gap
Covered 102%

Signals behind the momentum score

Weighted inputs with their source, so the score can be audited rather than trusted.

SignalChangeWeightSource
Search interest▲ 13%30%Tavily web search
Social mentions▲ 9%20%Tavily social crawl
Category sales▲ 8%15%Internal — category
Competitor listings▲ 9%15%Tavily competitor scan
Own sales velocity▲ 10%20%Internal — sales
Back-test

Six weeks ago the model scored this style 55/100; it now reads 59/100.

Sales over the same window moved ▲ 10%, so the signal led the sales curve in the right direction.

Coverage by size

The same four metrics resolved to the level a merchandiser actually buys at.

SizeStockDaily velocityPredicted 30dCoverageDays of coverRiskReturns uplift
XS1231.756220%662%+3%
S1424.415691%2756%+8%
M1667.425964%1970%+9%
L1666.522773%2266%+8%
XL1663.9137121%3641%+8%
XXL1481.861241%722%+6%

Agent plan for this style

Ranked, costed, and reversible. Approving writes a real record.

Step 1 · Transfer stock£221

Transfer 6 units between locations before buying more

  • ·1 size/location pair is mismatched: one site is over 150% coverage while another is under 88%.
  • ·Rebalancing satisfies predicted demand with stock already owned — no cash outlay and no 28-day lead time.
  • ·Avoids an estimated 6 units ending the season as residual stock at markdown.
SManchester→Shoreditch6 units
Confidence 75%
Step 2 · Purchase order£10,614

Raise PO for 234 units on a corrected size curve

  • ·Trend momentum 59/100 at 93% confidence, and coverage is 102% of predicted 30-day demand.
  • ·M should be 51% of the order, not the 19% in the supplier prepack.
  • ·Ordering the prepack curve instead would leave the core sizes short while adding residual tail stock.
Corrected curve vs supplier prepack · Milano Silk · 28d lead time
XS
0
0% vs 12%
S
30
13% vs 17%
M
119
51% vs 19%
L
85
36% vs 19%
XL
0
0% vs 18%
XXL
0
0% vs 15%
Confidence 93%
Step 3 · Markdown ladder£4,426

Start a markdown ladder on XS, XXL

  • ·142 units in XS, XXL will not clear at full price before season end.
  • ·Momentum is 59/100, so waiting will not rescue these sizes.
  • ·A staged ladder clears them near 25% off rather than the usual 60% clearance.
Ladder · 142 units residual if nothing changes
Week 3
−10%
£80
102 left
Week 6
−25%
£67
85 left
Week 9
−40%
£53
68 left
Confidence 60%
Step 4 · Supplier / fit£3,178

Flag Milano Silk — pattern runs small

  • ·33% of returns on this style are "Too small", well above the 20% baseline.
  • ·That demand did not disappear, it landed one size up — so sales history under-forecasts that size.
  • ·Correcting the graded pattern and the size guide reduces returns and fixes the forecast at source.
Finding · Milano Silk grading runs half a size small on bias cut skirts
Too small 33%Changed mind 20%Too large 14%
Confidence 88%

Returns as a demand signal

Reason mix over the last 120 days.

Too small33%
Changed mind20%
Too large14%
Fit in shoulders13%
Colour not as pictured12%
Fabric quality8%