Supply Strategy: Kraljic Matrix and JIT

← All maturity guides

SEF Reference Guide G14  |  Manufacturing OpEx Maturity Assessment

What it is

A supply strategy decides, for each group of purchased parts, how you will buy it, from whom, from where, and how it will reach the line. Without one, every part gets the same treatment: the same purchasing process, the same order quantities, the same safety stock rules. That wastes effort on low-risk parts and leaves critical ones exposed.

The Kraljic Matrix is the standard starting point. Peter Kraljic introduced it in “Purchasing Must Become Supply Management” (Harvard Business Review, September 1983). It sorts purchased items by two questions: how much does this item affect our profit, and how risky is its supply? The answers put each item into one of four quadrants, and each quadrant gets its own strategy.

Once you know the strategy, you decide how material arrives. Just-in-Time (JIT) supply delivers key parts in small, frequent quantities matched to consumption. Parts that cannot be supplied JIT (long lead time, distant supplier, erratic demand) get a planned, sized buffer instead of a guess.

Why it matters

• Cost: leverage items are where competitive bidding and volume consolidation save real money. Non-critical items are where you save by cutting purchasing effort.
• Delivery and continuity: strategic and bottleneck items are where a single supplier failure stops the line. Naming them lets you protect them before it happens.
• Cash: JIT supply for high-value, high-volume parts cuts raw material inventory. Right-sized buffers for the rest stop over-buying “just in case.”
• Resilience: comparing local and global sources on total landed cost and risk, not just unit price, avoids cheap parts that cost more once freight, duty, inventory, and disruption are counted.
• Focus: buyers spend their time where the money and the risk are.

Key terms

Term	Meaning
Kraljic Matrix	A purchasing strategy tool (Peter Kraljic, 1983) that sorts purchased items by profit impact and supply risk into four quadrants: strategic, leverage, bottleneck, and non-critical, each with its own sourcing strategy.
JIT	Just in Time. Producing and delivering only what is needed, when it is needed, in the amount needed. Anything made before it is needed is waste.
Buffer	A planned quantity of raw material or WIP placed before a workstation so it can keep working when supply varies (defined in G12).
Profit impact	How much the item affects the bottom line: annual spend, share of product cost, and effect on product quality or growth.
Supply risk	How likely supply is to be disrupted: number of qualified suppliers, availability, substitutes, lead time, geographic and political risk, make-or-buy options.
Strategic items	High profit impact, high supply risk. Few suppliers, critical to the product.
Leverage items	High profit impact, low supply risk. Many capable suppliers, meaningful spend.
Bottleneck items	Low profit impact, high supply risk. Low spend, but only one or two sources.
Non-critical items	Low profit impact, low supply risk. Commodity parts, many sources.
Total landed cost	Unit price plus freight, duty, packaging, handling, inventory carrying cost, quality cost, and an allowance for supply risk. The basis for comparing local and global sources.
Milk run	A scheduled pickup route that collects small quantities from several suppliers on a fixed timetable.
Supplier kanban	A pull signal (card or electronic) that authorizes a supplier to ship a set quantity when the customer consumes it.

How to do it

1. Build the spend cube. Export 12 months of purchases: part or category, supplier, quantity, spend, lead time, number of qualified sources, and supplier location.
2. Group into categories. Classify parts into sensible buying groups (castings, machined parts, electronics, fasteners, packaging, and so on). Kraljic works best on categories, then drills down to key parts.
3. Score profit impact and supply risk. Use a simple 1 to 5 scale for each. Profit impact: share of spend, share of product cost, quality impact. Supply risk: number of sources, lead time, substitutes, supplier financial health, geographic risk. Agree on the scoring rules first so the result is not just opinion.
4. Plot the matrix and set strategy per quadrant.

Quadrant	Profit impact / Supply risk	Management focus (Kraljic’s term)	Typical actions
Strategic	High / High	Supply management	Long-term partnership, joint planning and improvement, dual sourcing where practical, make-or-buy review, risk monitoring.
Leverage	High / Low	Sourcing management	Competitive bidding, volume consolidation, target pricing, regular supplier switching where it pays.
Bottleneck	Low / High	Materials management	Secure supply: longer contracts, safety stock, qualify alternates, redesign to use standard parts.
Non-critical	Low / Low	Purchasing management	Cut transaction cost: catalogs, blanket orders, vendor-managed inventory, two-bin or supplier-managed replenishment.

1. Compare local and global options. For strategic and leverage categories, compare sources on total landed cost, lead time, flexibility, and risk. Include inventory: a supplier with a 12-week ocean lead time forces you to hold far more stock than one 2 hours away.
2. Decide the supply mode for each key part.
   • JIT candidates: high value or high volume, stable demand, reliable supplier, short and predictable transit. Use supplier kanban, frequent deliveries, milk runs, or sequenced delivery to the line.
   • Planned buffer parts: long lead time, distant supplier, erratic demand, or unreliable supply. Size a buffer from demand, lead time, and variation, and mark its location and max and min.
3. Size the buffers. A common starting point for safety stock, covering demand variation only, is z x (standard deviation of daily demand) x square root of (lead time in days), where z sets the service level (for example, about 1.65 for 95%). Add cover for supply variation where the supplier is unreliable. Review quarterly. Buffers of raw material or common components also let the frozen production plan be shorter than the supply lead time (see G27). Every buffer is inventory, and inventory ties up cash: it raises net working capital, adds carrying cost (storage, handling, obsolescence, and the cost of capital), and can hide problems. Size each buffer to the risk it covers, put it where it protects the most products (raw material or common components rather than finished goods), and take it to S&OP (G20) for approval, so the working capital is a deliberate business choice, not a side effect of purchasing decisions.
4. Write it down and review it. Record the quadrant, strategy, supply mode, and buffer size for each category and key part. Review it once a year, and after any major supply disruption, new product, or supplier change. Buffer quantities and their inventory value are reviewed every month in the S&OP supply review and approved in the executive meeting alongside the rest of the inventory plan.

What each maturity level looks like

FL-4 Part and Supplier Strategy

Level	Assessment wording	What it looks like on the floor	How to move up
1	No part or supplier strategy.	Every part is bought the same way. Buyers chase price and shortages. Stock levels are set by habit. Line stops for missing parts are common.	Build a 12-month spend cube and list the parts that caused line stops or expediting last year.
2	Part and supplier strategy is considered ad hoc.	Some key suppliers get special attention, usually after a problem. Decisions depend on who is buying. Nothing is written.	Score categories on profit impact and supply risk and plot the Kraljic Matrix with purchasing, engineering, and operations together.
3	As Level 2, plus Kraljic analysis is used to develop the supply strategy.	A current Kraljic Matrix exists. Each category has a documented strategy. Buyers can say which quadrant their parts are in and what that means.	For strategic and leverage categories, compare local and global sources on total landed cost and risk, not unit price.
4	As Level 3, plus sourcing is optimized across local and global options.	Sourcing decisions show total landed cost comparisons. Some parts moved closer to shorten lead time, others moved to lower cost regions where risk is acceptable. Dual sources exist for critical items.	Define which key parts will be supplied JIT and set planned, sized buffers for the rest.
5	As Level 4, plus Just-in-Time (JIT) supply for key parts and planned buffers for non-JIT parts.	Key parts arrive in small, frequent deliveries triggered by consumption. Non-JIT parts sit in marked buffers with max and min levels. Raw material inventory is falling while line stops for parts are rare.	Sustain: review the matrix annually, extend JIT to more parts as suppliers prove reliable, and share results with other sites.

Evidence to look for

Document

• Kraljic Matrix with date, scoring rules, and the team who built it.
• Category strategy sheets.
• Total landed cost comparisons for recent sourcing decisions.
• List of JIT parts with delivery frequency and signal type.
• Buffer sizing sheet for non-JIT parts.

Observation

• At receiving: how often do key parts arrive, and in what quantities?
• At the line: are there marked buffer locations with max and min for purchased parts? Are they within limits?
• Are there supplier kanban cards, bins, or electronic signals in use?
• How much purchased material is in the warehouse, and how old is it?

Statement (questions to ask and who to ask)

• Buyer: “Which of your categories are strategic, and what is the plan for them?”
• Purchasing manager: “When did you last move a part between a local and a global source? How did you compare them?”
• Materials planner: “How did you set the safety stock on this part?”
• Production supervisor: “When did the line last stop for a missing purchased part?”

Management report

• Raw material inventory value and days of supply, by category.
• Line stops or expedites caused by purchased parts.
• Supplier on-time delivery for JIT parts.
• Premium freight spend.
• Savings by category (see G23).

Common pitfalls

• Scoring on opinion. If the scoring rules are not agreed first, the matrix reflects whoever talked loudest. Write the criteria down.
• Treating the matrix as permanent. Supply risk changes with markets, geopolitics, and supplier health. A matrix older than a year is probably wrong in places.
• Choosing on unit price. A cheaper global source can cost more once freight, duty, inventory, quality, and disruption are included.
• JIT without reliable suppliers. JIT with an unreliable supplier just moves the shortage to your line. Prove delivery and quality first (see G09).
• Pushing inventory to the supplier. JIT should lower inventory in the chain, not just move it to the supplier’s warehouse with a price increase attached.
• Ignoring bottleneck items. A low-cost part with one source can stop the line as surely as a costly one. These often get the least attention.

Template

Kraljic scoring sheet (one row per category or key part)

Category / part	Annual spend	% of product cost	Quality impact (1 to 5)	Profit impact score (1 to 5)	No. of qualified sources	Lead time (weeks)	Substitutes? (Y/N)	Geographic / supplier risk (1 to 5)	Supply risk score (1 to 5)	Quadrant	Strategy	Supply mode (JIT / buffer)

Total landed cost comparison

Cost element	Source A (local)	Source B (global)
Unit price
Freight per unit
Duty and customs
Packaging and handling
Inventory carrying cost (stock held x carrying rate)
Quality cost (inspection, PPM, returns)
Risk allowance (disruption, currency)
Total landed cost per unit
Lead time (weeks)

Buffer sizing (non-JIT parts)

Part	Avg daily demand	Std dev daily demand	Lead time (days)	Service level z	Safety stock = z x std dev x square root of lead time	Max	Min (reorder point)	Location

First 30 days

• Week 1: Export 12 months of spend. Group into categories. List parts that caused line stops, expedites, or premium freight.
• Week 2: Agree on the scoring rules with purchasing, engineering, operations, and quality. Score the top categories (usually 80% of spend plus any known risk items).
• Week 3: Plot the matrix. Draft a one-line strategy per quadrant and a short plan for the top three strategic and bottleneck items.
• Week 4: Pick two or three JIT pilot parts with reliable suppliers. Size buffers for five long lead time parts. Put the plan on the monthly review agenda.

Further reading

• Peter Kraljic, “Purchasing Must Become Supply Management,” Harvard Business Review, September 1983. https://hbr.org/1983/09/purchasing-must-become-supply-management
• Cees J. Gelderman and Arjan J. van Weele, “Handling measurement issues and strategic directions in Kraljic’s purchasing portfolio model,” Journal of Purchasing and Supply Management (2003).
• Taiichi Ohno, Toyota Production System: Beyond Large-Scale Production (Productivity Press, 1988).
• Lean Enterprise Institute, Lexicon entry for just-in-time production: https://www.lean.org/lexicon-terms/just-in-time-production/
• Association for Supply Chain Management (ASCM), resources on sourcing, inventory, and supply planning: https://www.ascm.org