Sign in

Blog · Board and management reporting · Supply chain

Supply chain KPI scorecard: layout, owners and thresholds

Build a supply chain KPI scorecard with actuals, approved targets, trends, owners and actions. A worked site comparison shows the layout, threshold rules and checks that keep the recurring review consistent.

The short answerA supply chain KPI scorecard puts each measure's actual result, approved target, trend, status, owner and action on the same recurring report. Keep the period and denominator visible, define whether higher, lower or a range is better, and calculate company results from source totals. Use thresholds to direct review rather than replace judgment.

A supply chain KPI scorecard compares actual performance with approved targets and turns exceptions into owned actions. Put the current result, prior result, threshold rule and next step on the same row so the recurring meeting can decide what needs attention.

The scorecard owns the review layout. The existing KPIs for supply chain teams guide covers the wider measure list and formulas; choose from it rather than filling the report with every available number.

Start with a clear row definition

Use one row per measure, site and reporting period. The measure name alone is insufficient: an on-time rate needs an event population, a date condition and a cutoff. An inventory measure needs a stock definition and a demand assumption.

Column What to record Example entry
Measure and scope Business definition, unit, site Customer on-time delivery, Dayton
Period and cutoff Due-date period and export cutoff September 2026; September 30 cutoff
Current actual Result computed from accepted inputs Current qualifying orders / current eligible orders
Prior actual Comparable earlier result Prior qualifying orders / prior eligible orders
Change Current minus prior, with unit (Current rate − prior rate) × 100 percentage points
Target and direction Approved minimum, maximum or range Approved minimum service rate
Status rule Exact boundaries and missing-data treatment Approved amber lower bound through below target
Owner Person accountable for review Distribution manager
Action and due date Specific next step and deadline Review late orders by October 8, 2026
Evidence Supporting rows and review notes Due-order exception schedule

Keep numerator and denominator columns in the supporting worksheet even if the executive view hides them. They explain how a rate moved when workload changed.

NIST's performance measurement guidance emphasizes meaningful current levels, trends, comparisons and measures that support decisions. A status color contributes little unless the row supplies enough context for someone to act.

Work the scorecard from source counts

Fictional US distributor Harbor Parts has sites in Dayton, Ohio, and Reno, Nevada. The scorecard compares August and September 2026 orders using their original committed delivery month. Open overdue orders remain in the denominator. The following are source counts, before any status is assigned.

Site August orders due August on time September orders due September on time September supplier commitments due Supplier commitments received on time
Dayton 100 92 120 108 80 72
Reno 50 45 60 57 20 18
Total 150 137 180 165 100 90

The inventory export below covers the same stocked part at both sites. Available units exclude blocked stock. Daily demand and unit cost are approved inputs; cover uses that stated demand assumption rather than claiming a forecast.

Site Available units Daily demand (units) Unit cost (USD) Inventory value (USD) Days of cover
Dayton 600 30 40 24,000 20.0
Reno 450 15 40 18,000 30.0
Total 42,000

Dayton inventory value is 600 × 40 = $24,000, and its cover is 600 / 30 = 20.0 days. Reno value is 450 × 40 = $18,000, with cover of 450 / 15 = 30.0 days. Do not add site cover values or treat a total dollar balance as a cover calculation.

Approve thresholds before assigning status

These are illustrative company rules, not industry standards. The operations executive approves the service thresholds; the inventory owner approves the cover range. Record their effective date with the scorecard.

Measure Direction Green Amber Red
Customer on-time delivery Higher is better At least 95.0% At least 90.0%, below 95.0% Below 90.0%
Supplier on-time receipt Higher is better At least 92.0% At least 88.0%, below 92.0% Below 88.0%
Available stock cover Within approved range 15.0 through 25.0 days, inclusive No amber band in this rule Below 15.0 or above 25.0 days

Missing or invalid inputs receive an unavailable status, not green and not a numeric zero. A site without eligible orders needs an explanation rather than a division error disguised as performance.

ASCM's SCOR performance framework distinguishes reliability from asset performance and other attributes. That helps explain why a service rate and an inventory measure belong in different units. This small operating scorecard is not a complete assessment of every framework attribute.

The completed review layout

Rates are rounded to one decimal place. Calculate changes from unrounded rates before rounding their display. The cover review has no prior-period input, so its trend is explicitly unavailable.

Measure Site Current actual Prior actual Change Approved target Status Owner
Customer on time Dayton 90.0% 92.0% −2.0 points At least 95.0% Amber Distribution manager
Customer on time Reno 95.0% 90.0% +5.0 points At least 95.0% Green Distribution manager
Supplier on time Dayton 90.0% Unavailable Unavailable At least 92.0% Amber Procurement manager
Supplier on time Reno 90.0% Unavailable Unavailable At least 92.0% Amber Procurement manager
Available stock cover Dayton 20.0 days Unavailable Unavailable 15.0–25.0 days Green Inventory manager
Available stock cover Reno 30.0 days Unavailable Unavailable 15.0–25.0 days Red Inventory manager

The delivery formulas are 108 / 120 = 90.0% for Dayton and 57 / 60 = 95.0% for Reno. Dayton is 5.0 percentage points below its target; Reno meets it. Supplier rates are 72 / 80 = 90.0% and 18 / 20 = 90.0%, each 2.0 points below target. Reno cover exceeds the approved upper bound by 30.0 − 25.0 = 5.0 days.

Those gaps guide the discussion. They do not prescribe purchasing decisions or supplier sanctions. A red stock-cover result asks the planner to validate demand and open commitments before deciding whether stock is avoidable.

Add actions without losing the measurement evidence

Exception Action Owner Due date Evidence needed to close
Dayton customer service below target Review late due orders and document causes Distribution manager October 8, 2026 Order-level cause review
Supplier service below target at both sites Confirm missed commitments and follow-up owners Procurement manager October 9, 2026 Reviewed commitment schedule
Reno stock cover above range Validate demand assumption and open purchase commitments Inventory manager October 9, 2026 Approved item-site review

The general KPI scorecard guide covers the same review discipline across teams. For alternative presentation choices, see supply chain dashboard examples and the supply chain KPI dashboard spreadsheet guide.

Prove the rollup before the meeting

Company September delivery is (108 + 57) / (120 + 60) = 165 / 180 = 91.7%. August is (92 + 45) / (100 + 50) = 137 / 150 = 91.3%. The unrounded change rounds to +0.3 percentage points, even though subtracting the displayed rates would suggest +0.4.

Company supplier performance is (72 + 18) / (80 + 20) = 90 / 100 = 90.0%. Inventory values reconcile to 24,000 + 18,000 = $42,000. Tie these totals back to the accepted export populations and retain the report cutoff with the results.

Where it goes wrong

  • Revising targets after seeing actuals. Preserve the approved threshold version so status means the same thing across periods.
  • Using one direction for every measure. Higher stock cover can breach an upper bound even when service uses a minimum target.
  • Hiding unavailable inputs. Missing prior data cannot support a flat trend, and a blank denominator cannot support a rate.
  • Averaging site rates. Aggregate the eligible counts, then divide.
  • Closing an action because the color changed. Require the agreed completion evidence; a different order mix can change a rate without resolving its cause.

Review your own scorecard inputs in Covirage

Bring your due-order export, supplier commitment file and inventory snapshot, with the approved targets and reporting definitions. Covirage maps the columns, and its tools compute and check the comparisons; the external AI model explains the results with every figure traced to its rows. Create a free Covirage account and bring a sample of your own data; creating the account takes about a minute.

See Covirage for supply chain teams for the supporting industry context.

Questions people ask

What columns belong in a supply chain KPI scorecard?

Include the measure, site, period, current actual, prior actual, approved target, direction, status, owner, action and due date. Retain numerator and denominator fields in a supporting schedule. This lets reviewers distinguish a changed result from a changed definition or population.

How do you choose scorecard thresholds?

Have the accountable business owner approve thresholds using customer commitments, replenishment requirements and the company's operating priorities. Define the boundaries precisely, including equality. Record the effective date, and do not present illustrative thresholds from another business as universal industry standards.

Can every KPI be shown as higher is better?

No. Service rates often use a minimum, delays may use a maximum, and stock cover may use an approved range. Preserve each measure's direction and unit. A stock balance that rises can improve availability while also increasing the amount of cash tied up.

Should site percentages be averaged for the company score?

Usually not. For a service rate, sum the successful events and divide by the summed eligible events. That preserves the underlying volume weights. A simple average gives a small site the same influence as a large one and can hide where the work occurred.