Units Sold rose 64.1% compared with Mar 21, 2016.
The largest factor tied to one place is electronics · DVD Players · AweDVD, adding about 9,634 units a week, followed by Male and jewelry. A broad-based factor also moved the whole business.
From the baseline to now, factor by factor
Average units per week. The axis starts at 38,379, not zero, so the factors are easier to compare. "All other factors" is the sum of the smaller real factors.
The factors, largest first
A factor acts at one place in the business and moves everything inside it. Open one to see its evidence.
1Across the boardThe whole business swung together week to week in early April
Between late March and mid-April, every part of the business rose and fell together — up about 60%, down about half, up again, then down about two-thirds in four straight weeks — before settling. It moved about 10,000 units a week in both directions but left only about 2,100 a week in the net result.
Broad-basedOffsetting waveSee the evidence+2,118units a week · 7% of the change
The whole business swung together week to week in early April
Between late March and mid-April, every part of the business rose and fell together — up about 60%, down about half, up again, then down about two-thirds in four straight weeks — before settling. It moved about 10,000 units a week in both directions but left only about 2,100 a week in the net result.
What it added, week by week
Where it acts
This factor does not sit in one place. It moved the whole business together, so look for something that affects everything at once: pricing, seasonality, a market-wide shift, or a change in how the data is recorded.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- What changed for the whole business in this window (pricing, calendar, channel mix)?
- Did anything change in how this metric is recorded or reported?
Largest week
−31,072 units in the week of Apr 18, 2016
2electronics › DVD Players ›AweDVDA single product launched in mid-April and led the growth
AweDVD, inside DVD Players, went from nothing to about 9,600 units a week after mid-April — a third of all the growth in the window, from one item. It acts on that product everywhere it sells, so a region-by-region or category-by-category cut would have spread it thin and missed it.
LocalizedLaunchBuilding · since Apr 11, 2016See the evidence+9,634units a week · 33% of the change
A single product launched in mid-April and led the growth
AweDVD, inside DVD Players, went from nothing to about 9,600 units a week after mid-April — a third of all the growth in the window, from one item. It acts on that product everywhere it sells, so a region-by-region or category-by-category cut would have spread it thin and missed it.
What it added, week by week
Where it acts
electronics › DVD Players › AweDVDFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at AweDVD: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- Trace the timing of this lift against promotions, pricing, inventory, operations, policy changes, and data-quality events. Confirm the explanation with the owner closest to this part of the business before acting.
- When did AweDVD start, and where?
- Is this new demand, or did it come from something that already existed?
- Is it still building, or has it levelled off?
Largest week
+13,963 units in the week of May 16, 2016
3MaleA factor lifted male customers in every region from mid-April
From the week of April 11, sales to male customers stepped up about 26% and stayed there — roughly 6,000 units a week. It shows up in every region at once, so it acts on the customer group, not on any one place.
LocalizedStep change · since Apr 11, 2016See the evidence+6,043units a week · 21% of the change
A factor lifted male customers in every region from mid-April
From the week of April 11, sales to male customers stepped up about 26% and stayed there — roughly 6,000 units a week. It shows up in every region at once, so it acts on the customer group, not on any one place.
What it added, week by week
Where it acts
MaleFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at Male: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- Trace the timing of this lift against promotions, pricing, inventory, operations, policy changes, and data-quality events. Confirm the explanation with the owner closest to this part of the business before acting.
- What changed at Male (since Apr 11, 2016): pricing, promotion, supply, staffing, or a lost or won account?
- Is it still in effect, and is it expected to continue?
Largest week
+8,411 units in the week of Jun 13, 2016
4South ›jewelryJewelry in the South built steadily from mid-April
A factor on jewelry in the South region built from mid-April to more than double the category's baseline there — about 5,700 units a week. It is specific to that region and category together, not to either alone.
LocalizedBuilding · since Apr 11, 2016See the evidence+5,682units a week · 19% of the change
Jewelry in the South built steadily from mid-April
A factor on jewelry in the South region built from mid-April to more than double the category's baseline there — about 5,700 units a week. It is specific to that region and category together, not to either alone.
What it added, week by week
Where it acts
South › jewelryFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at jewelry: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- Trace the timing of this lift against promotions, pricing, inventory, operations, policy changes, and data-quality events. Confirm the explanation with the owner closest to this part of the business before acting.
- What changed at jewelry (since Apr 11, 2016): pricing, promotion, supply, staffing, or a lost or won account?
- Is it still in effect, and is it expected to continue?
Largest week
+8,265 units in the week of May 16, 2016
5WestA factor kept the whole West region ahead all window
A factor acting on the West held the region about 45% above its baseline throughout the window — roughly 5,500 units a week. It acts at the region level, not at any one state or product inside it.
LocalizedSustained · all windowSee the evidence+5,451units a week · 19% of the change
A factor kept the whole West region ahead all window
A factor acting on the West held the region about 45% above its baseline throughout the window — roughly 5,500 units a week. It acts at the region level, not at any one state or product inside it.
What it added, week by week
Where it acts
WestFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at West: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- What changed at West: pricing, promotion, supply, staffing, or a lost or won account?
- Is it still in effect, and is it expected to continue?
Largest week
+8,477 units in the week of Mar 28, 2016
6Northeast ›New EnglandA factor at New England held it far below baseline
A factor specific to the New England division kept it well below its baseline for the whole window — the one sustained headwind, costing about 4,500 units a week.
LocalizedSustained · all windowSee the evidence−4,481units a week · -15% of the change
A factor at New England held it far below baseline
A factor specific to the New England division kept it well below its baseline for the whole window — the one sustained headwind, costing about 4,500 units a week.
What it added, week by week
Where it acts
Northeast › New EnglandFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at New England: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- What changed at New England: pricing, promotion, supply, staffing, or a lost or won account?
- Is it still in effect, and is it expected to continue?
Largest week
−5,592 units in the week of Jun 13, 2016
7Midwest ›electronicselectronics in Midwest added about 1,205 units a week
This factor rose, peaked around Apr 11, 2016, then returned to baseline, a temporary force, not a lasting shift.
LocalizedRose then fadedSee the evidence+1,205units a week · 4% of the change
electronics in Midwest added about 1,205 units a week
This factor rose, peaked around Apr 11, 2016, then returned to baseline, a temporary force, not a lasting shift.
What it added, week by week
Where it acts
Midwest › electronicsFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at electronics: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- What changed at electronics: pricing, promotion, supply, staffing, or a lost or won account?
- Is it still in effect, and is it expected to continue?
Largest week
+4,127 units in the week of Apr 11, 2016
8West ›MaleMale in West added about 1,076 units a week
This factor has been steadily lifting the metric across the whole window, present from the start, no single onset.
LocalizedSustained · all windowSee the evidence+1,076units a week · 4% of the change
Male in West added about 1,076 units a week
This factor has been steadily lifting the metric across the whole window, present from the start, no single onset.
What it added, week by week
Where it acts
West › MaleFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at Male: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- What changed at Male: pricing, promotion, supply, staffing, or a lost or won account?
- Is it still in effect, and is it expected to continue?
Largest week
+1,636 units in the week of Jun 13, 2016
9South › jewelry ›MaleMale in South · jewelry added about 894 units a week
This factor's influence has been growing since Apr 11, 2016 and is strongest now, a gain that is still developing.
LocalizedBuilding · since Apr 11, 2016See the evidence+894units a week · 3% of the change
Male in South · jewelry added about 894 units a week
This factor's influence has been growing since Apr 11, 2016 and is strongest now, a gain that is still developing.
What it added, week by week
Where it acts
South › jewelry › MaleFactorPrism tested every level of Geography, Product, Gender, and their combinations. This movement fits best at Male: a factor any higher would also move parts of the business that did not change, and splitting it further does not explain it any better.
Other places that explain it equally well
None. No other place in the data accounts for this movement as well, so there is one reading of it.
Questions for the owner
- What changed at Male (since Apr 11, 2016): pricing, promotion, supply, staffing, or a lost or won account?
- Is it still in effect, and is it expected to continue?
Largest week
+1,315 units in the week of May 9, 2016
When each factor was at work
Darker means a larger effect. Blue adds, orange subtracts.
What to check next
Start with electronics · DVD Players · AweDVD, the largest factor tied to a specific part of the business, then validate the next-ranked locations in order. These steps are derived from the measured evidence and avoid assuming a business cause that the source data does not prove.
- Validate what changed at electronics · DVD Players · AweDVD. Trace the timing of this lift against promotions, pricing, inventory, operations, policy changes, and data-quality events. Confirm the explanation with the owner closest to this part of the business before acting.
- Validate what changed at Male. Trace the timing of this lift against promotions, pricing, inventory, operations, policy changes, and data-quality events. Confirm the explanation with the owner closest to this part of the business before acting.
- Validate what changed at South · jewelry. Trace the timing of this lift against promotions, pricing, inventory, operations, policy changes, and data-quality events. Confirm the explanation with the owner closest to this part of the business before acting.
How to read this
Each factor is placed at the level where it acts. A factor at a region acts on that whole region, not on any one state inside it.
The factors add up exactly to the change, so the bridge always closes. "All other factors" is the sum of the smaller real factors, not a gap.
This shows where and when the change came from. Why it happened is for the people who run that part of the business to confirm.