KPI, Variance and Performance Analysis
Learn to explain why performance differs from plan, from choosing a fair comparison to splitting a variance into price, volume and mix. No coding: you work on the data of Raket Parts, a fictional maker of rocket parts.
You will be able to
- choose and defend a fair comparison
- recompute a KPI from its definition
- build a price, volume and mix bridge from order lines
- say what a variance does not prove
Compared with what?
Choose and defend a fair comparison for a sales decision, including when last year is misleading.
KPIs by the book
Recompute a delivery or net-sales kpi from its period, inclusions, exclusions and calculation rule.
Price, volume and mix
Build an ordered revenue bridge from actual order lines and state what a price step does not prove.
What drives the total?
Build a driver tree for a sales total, measure each driver's contribution with the interaction stated, and keep a contribution apart from a cause.
Which assumption failed?
Map each difference from plan to the assumption behind it, separate timing and reclassification from failed assumptions, and test the other explanations before you name a failure.
Investigating the residual in a cost variance
Find what is left after the known causes are explained.
Reviewing a colleague's variance pack
Check another analyst's bridge and commentary before it goes out.
Integration clinic: the quarterly business review pack
Bring every method together on one review pack.
Final assessment
A variance analysis on an unseen case.
KPI and Variance Analysis / Module 1
Compared with what?
Teun hears a simple question: was September good? A number becomes useful only when you choose the comparison that serves the decision.
Sections
KPI and Variance Analysis / Module 1 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
KPI and Variance Analysis / Module 2
KPIs by the book
Two people report different on-time delivery rates for the same quarter. The arithmetic may be right, but the groups counted can differ. Learn to use a written definition every time.
This module builds on module 1. If you are new to this, start with module 1.
Sections
KPI and Variance Analysis / Module 2 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
KPI and Variance Analysis / Module 3
Price, volume and mix
The average order line was worth more in August, yet the monthly total went down. Learn to distinguish line counts, the kinds sold and values within each kind before naming a cause.
This module builds on module 2. If you are new to this, start with module 2.
Sections
KPI and Variance Analysis / Module 3 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
KPI and Variance Analysis / Module 4
What drives the total?
Bram sees open order value fall sharply from May to June 2025. Teun breaks the total into parts and tests which differences are contributions, rather than proven causes.
This module builds on module 3. If you are new to this, start with module 3.
Sections
KPI and Variance Analysis / Module 4 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
KPI and Variance Analysis / Module 5
Which assumption failed?
Teun sees people cost and order intake depart from a practice plan at Raket Parts. He tests timing, moves between teams and missing evidence before calling any assumption wrong.
This module builds on module 4. If you are new to this, start with module 4.
Sections
KPI and Variance Analysis / Module 5 / Free practice
Try it on your own question
Run it in the practice editor
The practice editors have the Raket Parts tables loaded. Type a query or a formula, run it and change it as often as you like. Nothing is marked.
KPI and Variance Analysis / Module 1 / 1.1
1.1 A number needs a comparison
Section 1.1 complete
Next: 1.2 Four comparison bases
KPI and Variance Analysis / Module 1 / 1.2
1.2 Four comparison bases
Section 1.2 complete
Next: 1.3 When last year misleads
KPI and Variance Analysis / Module 1 / 1.4
1.4 Choose and justify
Section 1.4 complete
Next: Module 1 check
Continue to the module check Recap this section Back to module 1
KPI and Variance Analysis / Module 1 / Check
Module 1 check
Module 1 complete
Next: Module 2, KPIs by the book
Continue to module 2 Try it on your own question Recap module 1 Back to module 1
KPI and Variance Analysis / Module 2 / 2.2
2.2 Total over total
Section 2.2 complete
Next: 2.3 Exclusions change the answer
KPI and Variance Analysis / Module 2 / 2.4
2.4 Net of credits
Section 2.4 complete
Next: Module 2 check
Continue to the module check Recap this section Back to module 2
KPI and Variance Analysis / Module 2 / Check
Module 2 check
Module 2 complete
Next: Module 3, Price, volume and mix
Continue to module 3 Try it on your own question Recap module 2 Back to module 2
KPI and Variance Analysis / Module 3 / 3.1
3.1 Three reasons totals move
Section 3.1 complete
Next: 3.2 Mix without price change
KPI and Variance Analysis / Module 3 / 3.3
3.3 A real price check
Section 3.3 complete
Next: 3.4 A bridge that closes
KPI and Variance Analysis / Module 3 / 3.4
3.4 A bridge that closes
Section 3.4 complete
Next: Module 3 check
Continue to the module check Recap this section Back to module 3
KPI and Variance Analysis / Module 3 / Check
Module 3 check
Module 3 complete
Next: Module 4, What drives the total?
Continue to module 4 Try it on your own question Recap module 3 Back to module 3
KPI and Variance Analysis / Module 4 / 4.2
4.2 Build the tree
Section 4.2 complete
Next: 4.3 One at a time, then in a chain
KPI and Variance Analysis / Module 4 / 4.3
4.3 One at a time, then in a chain
Section 4.3 complete
Next: 4.4 Biggest is not always the lever
KPI and Variance Analysis / Module 4 / 4.4
4.4 Biggest is not always the lever
Section 4.4 complete
Next: Module 4 check
Continue to the module check Recap this section Back to module 4
KPI and Variance Analysis / Module 4 / Check
Module 4 check
Module 4 complete
Next: Module 5, Which assumption failed?
Continue to module 5 Try it on your own question Recap module 4 Back to module 4
KPI and Variance Analysis / Module 5 / 5.1
5.1 Not every miss is a failure
Section 5.1 complete
Next: 5.2 Map each difference to its assumption
KPI and Variance Analysis / Module 5 / 5.2
5.2 Map each difference to its assumption
Section 5.2 complete
Next: 5.3 Quantify one failed assumption, then test the others
KPI and Variance Analysis / Module 5 / 5.3
5.3 Quantify one failed assumption, then test the others
Section 5.3 complete
Next: 5.4 Conflicts, confidentiality and the review
KPI and Variance Analysis / Module 5 / 5.4
5.4 Conflicts, confidentiality and the review
Section 5.4 complete
Next: Module 5 check
Continue to the module check Recap this section Back to module 5
KPI and Variance Analysis / Module 5 / Check
Module 5 check
Module 5 complete
Next: Module 6, coming later
Back to the Path Try it on your own question Recap module 5 Back to module 5
KPI and Variance Analysis / Module 1 / Recap
Recap card
1.1 A number needs a comparison
A number needs a comparison
- Same measure: compare invoiced sales with invoiced sales.
- Starting point: an increase in per cent is measured from the earlier total.
Common mistake
The trap. A monthly rise does not, on its own, tell you why it happened.
KPI and Variance Analysis / Module 1 / Recap
Recap card
1.2 Four comparison bases
Four comparison bases
- Previous month: recent movement. Last year: comparable season.
- Plan: delivery against an approved plan. Target: progress to a goal.
Common mistake
The trap. A stretch target is not an approved budget.
KPI and Variance Analysis / Module 1 / Recap
Recap card
1.3 When last year misleads
When last year misleads
- Same month is only the start of fair comparison.
- Source history matters as much as the arithmetic.
Common mistake
The trap. Calling a very large calculated change clean growth without checking the older source.
KPI and Variance Analysis / Module 1 / Recap
Recap card
1.4 Choose and justify
Choose and justify
- Choose: the basis for the decision.
- Defend: dates, definition, source quality and uncertainty.
Common mistake
The trap. Selecting whichever number makes performance look best.
KPI and Variance Analysis / Module 1 / Recap
Recap card
Module 1: Compared with what?
A number needs a comparison
- Same measure: compare invoiced sales with invoiced sales.
- Starting point: an increase in per cent is measured from the earlier total.
Common mistake
The trap. A monthly rise does not, on its own, tell you why it happened.
Four comparison bases
- Previous month: recent movement. Last year: comparable season.
- Plan: delivery against an approved plan. Target: progress to a goal.
Common mistake
The trap. A stretch target is not an approved budget.
When last year misleads
- Same month is only the start of fair comparison.
- Source history matters as much as the arithmetic.
Common mistake
The trap. Calling a very large calculated change clean growth without checking the older source.
Choose and justify
- Choose: the basis for the decision.
- Defend: dates, definition, source quality and uncertainty.
Common mistake
The trap. Selecting whichever number makes performance look best.
KPI and Variance Analysis / Module 2 / Recap
Recap card
2.1 The parts of a KPI
The parts of a KPI
- Group: promised-quarter order lines with confirmed delivery.
- Share: count ON_TIME over the count of ON_TIME plus LATE.
Common mistake
The trap. Using every line because it happens to be in the table.
KPI and Variance Analysis / Module 2 / Recap
Recap card
2.2 Total over total
Total over total
- Right: add every eligible on-time line; divide by all eligible confirmed lines.
- Not right: give every line kind equal weight regardless of size.
Common mistake
The trap. A neat average of rates can ignore that group sizes differ.
KPI and Variance Analysis / Module 2 / Recap
Recap card
2.3 Exclusions change the answer
Exclusions change the answer
- Exclusion: an outcome or row the written rule leaves out.
- Difference note: name the changed clause, group and effect.
Common mistake
The trap. Using a broader bottom and calling it the same KPI.
KPI and Variance Analysis / Module 2 / Recap
Recap card
2.4 Net of credits
Net of credits
- Invoice amount: sales by invoice month.
- Net of issued credits: invoice amount less notes issued in the month.
Common mistake
The trap. Ignoring a small credit or silently changing its date rule.
KPI and Variance Analysis / Module 2 / Recap
Recap card
Module 2: KPIs by the book
The parts of a KPI
- Group: promised-quarter order lines with confirmed delivery.
- Share: count ON_TIME over the count of ON_TIME plus LATE.
Common mistake
The trap. Using every line because it happens to be in the table.
Total over total
- Right: add every eligible on-time line; divide by all eligible confirmed lines.
- Not right: give every line kind equal weight regardless of size.
Common mistake
The trap. A neat average of rates can ignore that group sizes differ.
Exclusions change the answer
- Exclusion: an outcome or row the written rule leaves out.
- Difference note: name the changed clause, group and effect.
Common mistake
The trap. Using a broader bottom and calling it the same KPI.
Net of credits
- Invoice amount: sales by invoice month.
- Net of issued credits: invoice amount less notes issued in the month.
Common mistake
The trap. Ignoring a small credit or silently changing its date rule.
KPI and Variance Analysis / Module 3 / Recap
Recap card
3.1 Three reasons totals move
Three reasons totals move
- Volume: how many. Mix: which kinds.
- Price: how much for a comparable unit under stated terms.
Common mistake
The trap. Treating higher line value as proof of higher unit price.
KPI and Variance Analysis / Module 3 / Recap
Recap card
3.2 Mix without price change
Mix without price change
- Share: FC order lines divided by all order lines in the same month.
- Observation: average line value increased while total line value fell.
Common mistake
The trap. Relabelling a mix change as an observed unit-price rise.
KPI and Variance Analysis / Module 3 / Recap
Recap card
3.3 A real price check
A real price check
- List versus net: an observed gap on the same order line.
- Review flag: an identifier for records requiring price inspection.
Common mistake
The trap. Calling a list-to-net gap an across-time price increase.
KPI and Variance Analysis / Module 3 / Recap
Recap card
3.4 A bridge that closes
A bridge that closes
- Order: start, line count, kind mix, within-kind average value, residual, end.
- Residual: disclose anything the named steps do not account for.
Common mistake
The trap. Silently plugging a remaining balance or treating a proxy as matched unit prices.
KPI and Variance Analysis / Module 3 / Recap
Recap card
Module 3: Price, volume and mix
Three reasons totals move
- Volume: how many. Mix: which kinds.
- Price: how much for a comparable unit under stated terms.
Common mistake
The trap. Treating higher line value as proof of higher unit price.
Mix without price change
- Share: FC order lines divided by all order lines in the same month.
- Observation: average line value increased while total line value fell.
Common mistake
The trap. Relabelling a mix change as an observed unit-price rise.
A real price check
- List versus net: an observed gap on the same order line.
- Review flag: an identifier for records requiring price inspection.
Common mistake
The trap. Calling a list-to-net gap an across-time price increase.
A bridge that closes
- Order: start, line count, kind mix, within-kind average value, residual, end.
- Residual: disclose anything the named steps do not account for.
Common mistake
The trap. Silently plugging a remaining balance or treating a proxy as matched unit prices.
KPI and Variance Analysis / Module 4 / Recap
Recap card
4.1 Do the parts add up?
Do the parts add up?
- The order-count and average-order-value branches multiply.
- The two separate contributions need an interaction to close.
Common mistake
The mistake. Adding one-at-a-time effects without interaction leaves €19,137.46 unexplained.
KPI and Variance Analysis / Module 4 / Recap
Recap card
4.2 Build the tree
Build the tree
- Four multiplying branches describe the total.
- Channel values are additive but distinct customer counts overlap.
- Working days did not change between May and June.
Common mistake
The mistake. Channel customer totals overlap even though channel euro totals add correctly.
KPI and Variance Analysis / Module 4 / Recap
Recap card
4.3 One at a time, then in a chain
One at a time, then in a chain
- One-at-a-time effects need an explicit interaction.
- A chain closes exactly but depends on its order.
- Average line value already contains changes in line-kind mix.
Common mistake
The mistake. A mix step beside average line value double counts the same underlying shift.
KPI and Variance Analysis / Module 4 / Recap
Recap card
4.4 Biggest is not always the lever
Biggest is not always the lever
- Six ROCKET_BUILDER customers appear in both months.
- That segment contributed most of the value fall.
- Campaign attribution is not causal evidence.
Common mistake
The mistake. A credited touch or a large numeric contribution does not prove why behaviour changed.
KPI and Variance Analysis / Module 4 / Recap
Recap card
Module 4: What drives the total?
Do the parts add up?
- The order-count and average-order-value branches multiply.
- The two separate contributions need an interaction to close.
Common mistake
The mistake. Adding one-at-a-time effects without interaction leaves €19,137.46 unexplained.
Build the tree
- Four multiplying branches describe the total.
- Channel values are additive but distinct customer counts overlap.
- Working days did not change between May and June.
Common mistake
The mistake. Channel customer totals overlap even though channel euro totals add correctly.
One at a time, then in a chain
- One-at-a-time effects need an explicit interaction.
- A chain closes exactly but depends on its order.
- Average line value already contains changes in line-kind mix.
Common mistake
The mistake. A mix step beside average line value double counts the same underlying shift.
Biggest is not always the lever
- Six ROCKET_BUILDER customers appear in both months.
- That segment contributed most of the value fall.
- Campaign attribution is not causal evidence.
Common mistake
The mistake. A credited touch or a large numeric contribution does not prove why behaviour changed.
KPI and Variance Analysis / Module 5 / Recap
Recap card
5.1 Not every miss is a failure
Not every miss is a failure
- Six people transferred into X-ERP.
- The unit pay move did not add company-wide pay for those people.
- July orders still require further diagnosis.
Common mistake
The mistake. A bad-looking unit line must not automatically be called a plan failure.
KPI and Variance Analysis / Module 5 / Recap
Recap card
5.2 Map each difference to its assumption
Map each difference to its assumption
- A plan can miss in date and rate.
- Headcount above plan needs a staffing explanation.
- Payroll and weekly overtime use different measures.
Common mistake
The mistake. Comparing hours from different sources is not a valid overtime assessment.
KPI and Variance Analysis / Module 5 / Recap
Recap card
5.3 Quantify one failed assumption, then test the others
Quantify one failed assumption, then test the others
- The H1 daily rate supplies a checkable practice plan.
- July is 185 orders below that rounded practice plan.
- Test workdays, incidents and missing seasonal history separately.
Common mistake
The mistake. An incident date or seasonal story alone is not proof of the entire shortfall.
KPI and Variance Analysis / Module 5 / Recap
Recap card
5.4 Conflicts, confidentiality and the review
Conflicts, confidentiality and the review
- RL02 price assumptions conflict with the agreement.
- One-person overtime subtotals should not be shown.
- The provider must confirm or reject back-pay.
Common mistake
The mistake. A missing back-pay field does not prove that back-pay did or did not occur.
KPI and Variance Analysis / Module 5 / Recap
Recap card
Module 5: Which assumption failed?
Not every miss is a failure
- Six people transferred into X-ERP.
- The unit pay move did not add company-wide pay for those people.
- July orders still require further diagnosis.
Common mistake
The mistake. A bad-looking unit line must not automatically be called a plan failure.
Map each difference to its assumption
- A plan can miss in date and rate.
- Headcount above plan needs a staffing explanation.
- Payroll and weekly overtime use different measures.
Common mistake
The mistake. Comparing hours from different sources is not a valid overtime assessment.
Quantify one failed assumption, then test the others
- The H1 daily rate supplies a checkable practice plan.
- July is 185 orders below that rounded practice plan.
- Test workdays, incidents and missing seasonal history separately.
Common mistake
The mistake. An incident date or seasonal story alone is not proof of the entire shortfall.
Conflicts, confidentiality and the review
- RL02 price assumptions conflict with the agreement.
- One-person overtime subtotals should not be shown.
- The provider must confirm or reject back-pay.
Common mistake
The mistake. A missing back-pay field does not prove that back-pay did or did not occur.