Lumière Skin Malaysia
Synthetic demo data
Six moments: detect, explain, predict, recommend, act, then close the loop.
Date range
Region
Branch
SKU
Campaign
Compare

Forecast and Campaigns

Connect future demand to promotion plans, always with a range and a model version.

Comparison
Model v2.4.1 · run 17 Aug 2026 06:00

Actual versus forecast · SK-002

i
v2.4.1 · run 2026-08-17 06:00
Forecast, next 30 days
8,099 units
range 6,741 – 9,457
The band widens with horizon because an estimate thirty days out is not as tight as one made for tomorrow. Mean absolute error across the last eight runs is 7.8% of actual units.
Actual units (M-02)Forecast point estimateLower–upper boundCampaign window

Campaign results

i
CampaignOfferUpliftIncr. GPCannibalisation
Payday 15% off Serum, Aug
in flight · Healthy across the network
Percentage discount+21.2%−MYR 12,480None detected
Cleanser + Moisturiser bundle, Aug
Healthy across the network
Bundle+20.7%+MYR 26,300−15.2% SK-003
Cleanser + Moisturiser bundle, Jul 2026
Healthy across the network
Bundle−12.4%−MYR 42,760−4.1% SK-003
Flash 12% off Cleanser, Jul 2026
Healthy across the network
Percentage discount+5.8%−MYR 18,006None detected
Daily ritual bundle, Jun 2026
Healthy across the network
Bundle+8.7%+MYR 3,358−41.2% SK-003
Payday 20% off Serum, Jun 2026
Healthy across the network
Percentage discount+4.4%−MYR 20,275None detected
Merdeka SPF push, May 2026
Healthy across the network
Gift with purchase+23.1%+MYR 4,498None detected
Mid-month 18% off Night Cream, May 2026
Healthy across the network
Percentage discount+6.9%−MYR 26,997None detected
Cannibalisation is measured at receipt level: units of the moisturiser bought on a basket that does not also contain the cleanser. Counting total units would show the line growing and miss the effect entirely.

Scenario

i

The projection applies the price elasticity observed on prior discount campaigns for this SKU, then re-runs the stock position across the campaign window.

Forecast accuracy

i
5.8%
Jun w1
8.5%
Jun w3
6.7%
Jun w5
9.9%
Jul w1
7.7%
Jul w3
5.0%
Jul w4
11.6%
Aug w1
7.1%
Aug w2
Mean absolute forecast error across the last 8 runs: 7.8% of actual units. Model v2.4.1, trained on 24 months of daily history.