Warehouse Strategy, Location and Network Decisions

Introduction

A warehouse is not merely a building in which inventory is stored. It is a strategic node that connects suppliers, manufacturing facilities, markets, customers and reverse flows. Decisions about the number, location, ownership and operating model of warehouses influence logistics cost, customer service, working capital, resilience, sustainability and competitive positioning. Because these decisions usually involve substantial capital, long-term contracts and organisational commitments, they must be aligned with the overall supply-chain strategy rather than made as isolated operational choices.

Warehouse strategy concerns the design of the physical and organisational network through which products flow. It addresses questions such as:

• Should inventory be held in one central warehouse or several regional facilities?
• Where should facilities be located?
• What capacity and technology should each facility possess?
• Should warehouses be owned, leased or outsourced to a third-party logistics provider (3PL)?
• What service levels are required for different customer segments?
• How should the network respond to demand uncertainty, disruptions and regulatory constraints?

The appropriate answer depends on product characteristics, demand geography, customer expectations, supplier structure, transportation economics, risk exposure and the organisation’s strategic priorities.

Learning Objectives

After studying this chapter, the learner should be able to:

1. Explain the strategic role of warehousing in supply-chain design.
2. Compare centralized, decentralized and hybrid warehouse networks.
3. Apply warehouse-location models and interpret their assumptions and limitations.
4. Evaluate capacity, ownership and outsourcing alternatives.
5. Design service requirements and align them with customer and supply-chain objectives.
6. Assess 3PL arrangements, governance mechanisms, technology and risk.
7. Formulate a defensible warehouse-network recommendation using cost, service, resilience and sustainability criteria.

1. Warehouse Strategy as a Network Decision

A warehouse network comprises facilities, transportation links, inventory positions, information systems, operating rules and governance arrangements. It should be viewed as an integrated system. A low-cost warehouse may be strategically poor if it creates excessive outbound transport, long order cycles or vulnerability to disruption.

A useful strategic alignment chain is:

Business strategy → customer value proposition → supply-chain priorities → warehouse network design → operating policies → performance outcomes

For example, an organisation competing on rapid delivery may require geographically dispersed stock, while a producer of slow-moving industrial equipment may benefit from centralised inventory and direct shipment. A low-price retailer may prioritise scale economies and efficient replenishment; a healthcare distributor may prioritise traceability, temperature control and regulatory compliance.

The principal decision dimensions are:

Strategic decisions should distinguish total cost from individual cost categories. Minimising warehouse rent alone can increase line-haul and last-mile costs. Similarly, maximising warehouse utilisation can reduce responsiveness if congestion causes delays.

2. Centralized, Decentralized and Hybrid Networks

2.1 Centralized networks

In a centralized network, inventory for a broad market is held in one major facility or a small number of national facilities. Customer orders are served from these locations, often through parcel, less-than-truckload or direct transport services.

Advantages include:

• pooling of demand, which can reduce safety stock;
• economies of scale in labour, equipment, systems and supervision;
• simpler inventory control and standardisation;
• lower risk of duplicate slow-moving stock;
• easier implementation of specialised processes such as postponement or quality inspection.

Disadvantages include longer average delivery distances, dependence on fewer facilities, possible exposure to regional disruption and higher delivery cost for remote customers. Centralisation can also produce a single point of failure unless contingency arrangements exist.

2.2 Decentralized networks

A decentralized network uses several regional, market-facing or local warehouses. Stock is positioned closer to customers, reducing delivery time and often lowering outbound transportation distance.

Its costs may include additional buildings, management, systems, handling equipment and duplicated inventory. Forecast errors can be amplified because each site holds its own safety stock. Inventory balancing between sites may also be difficult.

2.3 Hybrid networks

A hybrid network combines central and regional facilities. Fast-moving or service-critical items may be stocked regionally, while slow-moving, high-value or specialised products remain centralised. A central fulfilment centre may replenish forward stocking locations, while selected customers receive direct shipment from suppliers.

Hybrid design is often more realistic than a simple centralised/decentralized choice. It permits differentiated inventory policies, such as:

• regional stock for high-volume products;
• central stock for unpredictable demand;
• vendor-managed inventory for selected customers;
• cross-docking for stable, fast-moving flows;
• postponement, whereby final configuration occurs near demand.

2.4 Risk pooling and inventory implications

Centralisation may reduce aggregate safety stock through risk pooling. If demand variability across locations is not perfectly correlated, combining demand streams can reduce the relative amount of safety stock required. However, the benefit depends on demand correlation, replenishment lead time, service targets and product substitutability. Centralisation does not automatically reduce total inventory if it creates much longer replenishment lead times or if customer service requires local availability.

For an item with independent demand across n locations, equal standard deviation σ at each location, and a common safety factor z, a simplified comparison is:

versus decentralized safety stock:

Here, SS is safety stock in units, z is the standard-normal service factor, σ is the standard deviation of demand during the relevant protection period at one location, and n is the number of locations. The expressions assume independent, normally distributed demand, identical locations and the same protection period. They illustrate the pooling effect, not a complete inventory policy. If demand is correlated, the general variance of aggregate demand is:

where D_i is demand at location i and the second summation covers each distinct pair of locations. Positive correlation reduces the advantage of centralisation; negative correlation increases it.

3. Customer Service and Strategic Segmentation

Warehouse design must begin with service requirements, not with a preferred building or technology. Service should be translated into measurable requirements, including:

• order cycle time;
• promised delivery date;
• order completeness or fill rate;
• inventory availability;
• delivery-window compliance;
• return-processing time;
• product condition and accuracy;
• traceability and documentation.

A single service target for all products and customers is usually inefficient. Strategic segmentation can use profitability, urgency, product criticality, order frequency, regulatory sensitivity and substitutability. For example, emergency medical components may justify local safety stock, whereas infrequently ordered replacement parts may be centrally stocked with expedited transport.

The inventory cycle-service level is the probability of meeting demand during a replenishment cycle without a stockout. It differs from the fill rate, which is the proportion of demand quantity supplied immediately from available stock. Confusing these measures can lead to inappropriate network decisions.

A practical service-cost analysis examines the marginal cost of improving service. Moving from a 90% to a 95% service level may be relatively inexpensive; moving from 98% to 99.5% may require disproportionately more inventory, capacity and premium transport. Service commitments should therefore be economically and operationally credible.

4. Warehouse Location Models

Location analysis is a structured process rather than a single mathematical calculation. It normally includes:

1. defining demand points, supply points and candidate sites;
2. estimating volumes, transport rates and service constraints;
3. screening sites for legal, labour, infrastructure and environmental suitability;
4. modelling costs and service outcomes;
5. testing uncertainty and disruption scenarios;
6. evaluating qualitative and strategic factors;
7. selecting and validating the network.

4.1 Centre-of-gravity model

The centre-of-gravity method provides an initial geographic approximation for a facility serving several demand points. The weighted coordinates are:

where x_i,y_i are the coordinates of demand point i, and w_i is its weight, commonly annual shipment volume, demand, freight spend or a volume-distance proxy. The result is a continuous point, not necessarily a feasible site. The method assumes that coordinates are meaningful, transport cost behaves approximately proportionally with distance and other constraints are temporarily ignored.

Illustrative calculation. Suppose three markets have coordinates and annual demand as follows:

Then:

x* = (100×10 + 200×40 + 100×30) ÷ 400 = 30

y* = (100×20 + 200×30 + 100×60) ÷ 400 = 35

The initial point is (30, 35). A road network, land availability, flood risk, labour supply or planning restriction may require selection of a nearby feasible site rather than the exact point.

4.2 Factor-rating method

The factor-rating method evaluates qualitative and quantitative criteria. Each candidate receives a score for factors such as transport access, labour availability, land cost, utilities, tax environment, security, expansion potential and sustainability. With weights summing to one, the weighted score for site j is:

where S_j is the total score of site j, w_k is the importance weight of factor k, and r_jk is the rating of site j on that factor. Scores must use a consistent scale. The method supports judgement but does not eliminate subjectivity; sensitivity testing is essential.

4.3 Total-cost and network models

A simplified annual total-cost model for candidate network design is:

where F is facility fixed cost, H is handling and operating cost, T_in is inbound transport cost, T_out is outbound transport cost, I is inventory carrying cost, P is penalty or service-failure cost, and R is risk, resilience or disruption-related cost where it can be estimated. All terms must refer to the same time period and currency.

A more detailed mixed-integer network model may use binary variables indicating whether a facility is opened and allocation variables indicating flows between facilities and customers. Its objective commonly minimises fixed facility cost, transportation, handling and inventory cost subject to demand satisfaction, capacity, service and policy constraints. Such models can be powerful, but their results are only as reliable as the data and assumptions. They may also omit political feasibility, managerial capability, community impact and extreme-event uncertainty.

4.4 Break-even analysis for network alternatives

When comparing two alternatives, the break-even volume can be found if one has lower fixed cost and the other has lower variable cost. If alternative A has fixed cost F_A and variable cost v_A per unit, and alternative B has corresponding values F_B and v_B, the break-even volume is:

This applies when v_A ≠ v_B and both alternatives are compared over the same period and relevant volume range. It excludes nonlinear congestion, quantity discounts and capacity constraints unless they are incorporated into the cost functions.

5. Capacity and Facility Configuration

Capacity is more than floor area. It includes receiving doors, storage positions, picking locations, dock throughput, labour, equipment, systems and time available for operations. Capacity should be expressed in the constraint that limits flow. A building may have adequate cubic space but insufficient dock doors or picking labour.

Important capacity concepts include:

• Design capacity: theoretical maximum under specified conditions.
• Effective capacity: realistic maximum after maintenance, breaks, changeovers and operating restrictions.
• Utilisation: actual output divided by design or effective capacity, provided the denominator is clearly defined.
• Throughput: quantity processed per unit of time.
• Peak capacity: capability during seasonal or promotional demand.
• Scalability: ability to add capacity without disproportionate cost or disruption.

High utilisation is not always desirable. Near saturation, congestion may increase travel, queueing, errors and overtime. A facility with 95% average utilisation may be unable to absorb a demand peak or equipment failure. Capacity planning should therefore use scenarios rather than only average demand.

Design choices include expandable buildings, modular automation, flexible labour, shared-user facilities, overflow contracts and postponement. Automation is justified when volume, product stability, labour cost, accuracy requirements and operating horizon support the investment. Highly automated systems can be efficient but may be difficult to modify when product dimensions, order profiles or volumes change.

6. Ownership, Leasing and Outsourcing

6.1 Own versus lease

Owning a warehouse provides control over layout, technology, security, operating standards and long-term asset development. It may create residual property value and protect against rent escalation. However, it requires capital, creates location commitment and exposes the organisation to property and utilisation risk.

Leasing preserves capital and can provide flexibility in network redesign. It may permit faster implementation and easier access to modern facilities. Its limitations include contractual restrictions, rent escalation, renewal uncertainty and possible difficulty in installing specialised equipment.

The choice should consider total economic cost, not only accounting treatment. Relevant factors include:

• present value of lease payments and ownership cash flows;
• tax and financing effects;
• expected residual value;
• flexibility and exit cost;
• required fit-out investment;
• strategic control;
• market availability and expansion options.

6.2 Outsourcing to a 3PL

A third-party logistics provider performs logistics activities for a client under contract. Services may include storage, receiving, picking, packing, transport coordination, customs support, returns and value-added operations. A 3PL may operate a dedicated facility, a shared multi-client facility or a network of facilities.

Potential benefits include:

• access to specialist knowledge and systems;
• variable rather than fully fixed cost;
• faster entry into a new market;
• scalable labour and space;
• access to transport purchasing power;
• reduced managerial burden.

Risks include loss of direct control, data and process dependence, switching costs, service inconsistency, unclear accountability and provider financial or operational failure. Outsourcing does not transfer ultimate responsibility for customer outcomes, legal compliance or ethical conduct.

A robust 3PL contract specifies scope, operating standards, pricing, data ownership, audit rights, liability, business continuity, cybersecurity, subcontracting, continuous improvement and termination assistance. Service-level agreements should define measurement rules, exclusions and remedies. Incentives should avoid encouraging behaviour that improves one measure while damaging the system—for example, shipping incomplete orders merely to improve dispatch speed.

7. Technology, Visibility and Contemporary Developments

Warehouse-management systems (WMS) coordinate receiving, put-away, inventory records, replenishment, picking, packing, shipping and reporting. Transport-management systems (TMS) support carrier selection, routing, freight audit and delivery visibility. Integration with enterprise systems creates a common data foundation for network decisions.

Useful technologies include barcode and radio-frequency identification, mobile computing, slotting analytics, warehouse robotics, automated storage and retrieval, computer vision and digital twins. Artificial intelligence can support demand forecasting, labour planning, anomaly detection, slotting and predictive maintenance. Its value depends on data quality, process discipline and human oversight. Models should be monitored for bias, drift and unsafe recommendations; automated decisions involving employees, customers or regulated products require governance.

Contemporary network design increasingly considers:

• e-commerce and small-order profiles;
• omnichannel fulfilment and store-as-fulfilment nodes;
• returns and circular supply chains;
• geopolitical and climate disruption;
• carbon emissions and energy use;
• urban micro-fulfilment;
• collaborative and shared-user networks.

Sustainability should be evaluated across the network rather than by building efficiency alone. A highly energy-efficient warehouse that causes substantially more delivery kilometres may not reduce total environmental impact. Measures may include energy per order, emissions per unit delivered, transport distance, packaging intensity, waste diversion and renewable-energy share.

8. Risk, Resilience and Governance

Network design must account for uncertainty in demand, supply, transport, labour, utilities, cyber systems, weather, regulation and geopolitical conditions. A mathematically low-cost design can be strategically fragile if it depends on one port, one supplier, one carrier or one facility.

Resilience options include:

• multi-sourcing and alternate transport routes;
• backup warehouses or overflow agreements;
• strategic inventory of critical items;
• reciprocal capacity arrangements;
• emergency operating procedures;
• cyber and data-recovery controls;
• scenario exercises and stress testing.

Resilience has a cost, but the relevant comparison is expected total cost over the planning horizon, including disruption consequences. Exact probability estimates are often unavailable; scenario analysis and sensitivity analysis are therefore important. Managers should distinguish efficiency reserves, which may be removed to lower cost, from resilience reserves, which provide protection against severe events.

Governance also includes regulatory and ethical responsibilities. Warehouses may be subject to planning, fire, occupational safety, environmental, customs, food, pharmaceutical or hazardous-material requirements. Ethical management requires safe work design, fair labour practices, responsible subcontracting, privacy protection and honest performance reporting. Compliance should be built into network and facility design rather than treated as a final inspection.

9. Illustrative Case: Northstar Medical Devices

Background

Northstar Medical Devices distributes diagnostic consumables to hospitals and clinics across a large country. Its single national warehouse provides strong inventory pooling but has experienced rising parcel costs, transport delays and service failures in remote regions. Demand is growing unevenly, and some products require strict traceability and controlled storage.

Problem

Senior management is considering three options: retain the central facility, open two regional warehouses, or adopt a hybrid design with regional stock for high-volume products and central stock for slow-moving products. The organisation must improve next-day availability without excessive inventory or loss of regulatory control.

Analysis

The team first segmented products by demand volume, criticality, storage requirement and forecast uncertainty. High-volume, predictable products accounted for most order lines in the two largest regions. Low-volume products were widely dispersed and had uncertain demand. A centre-of-gravity analysis identified broad candidate areas, but final screening excluded sites lacking qualified labour, compliant temperature control and reliable power.

A total-cost model included facility, labour, inbound and outbound transport, inventory, quality compliance and disruption costs. The team also modelled demand peaks and a temporary closure of one facility. The two-regional-warehouse option produced shorter delivery distances but duplicated safety stock and required more quality-management resources. The hybrid option retained central control of slow-moving and sensitive products while placing selected fast-moving items in regional facilities.

Approach and solution

Northstar selected a hybrid network. Regional facilities held an approved assortment under common WMS controls. The central warehouse retained products with low demand, high value or special handling requirements. Inventory policies were differentiated by product class, and emergency replenishment rules were established. A 3PL was used for one regional facility under a dedicated contract with audit rights, validated storage processes, temperature monitoring and business-continuity requirements.

Performance was governed through a balanced scorecard: order completeness, promised-date delivery, inventory accuracy, quality deviations, cost per order, stock days, labour safety and emissions per shipment. A quarterly network review was established to test demand changes and disruption scenarios.

Outcome

In the illustrative scenario, regional availability improved for the selected high-volume products while the central facility continued to pool uncertain demand. The solution did not minimise every individual cost category; it improved the combined service, cost and risk position. Its success depended on disciplined assortment decisions, system integration, contract governance and periodic reassessment.

Learning points

1. Network design should begin with customer and product segmentation.
2. A hybrid network can combine pooling benefits with regional responsiveness.
3. Regulatory and quality requirements can eliminate apparently attractive sites.
4. A 3PL relationship requires governance, not simply a purchase order.
5. Resilience and service outcomes must be included in economic evaluation.

10. A Managerial Decision Process

A defensible warehouse-network decision can follow this sequence:

1. Clarify strategic priorities. Identify whether the organisation competes through cost, speed, reliability, customisation, compliance or resilience.
2. Segment demand and products. Analyse volume, variability, criticality, value, storage needs and customer promise.
3. Map the current network. Include suppliers, plants, warehouses, customers, flows, lead times and constraints.
4. Generate alternatives. Consider centralised, regional, hybrid, cross-dock, direct-ship and outsourced options.
5. Estimate total cost and service. Use consistent time periods, realistic capacity and explicit assumptions.
6. Test uncertainty. Vary demand, fuel cost, labour cost, lead time, disruption and service requirements.
7. Evaluate qualitative factors. Include capability, compliance, community, sustainability, governance and strategic flexibility.
8. Select contracts and controls. Define ownership, SLAs, data rights, audit, escalation and exit provisions.
9. Implement in stages. Pilot processes, protect service during transition and validate master data.
10. Review continuously. Network design is not permanent; changing demand, technology and risk require periodic reassessment.

Challenges and Limitations

Warehouse-network models face several limitations. Data may be incomplete, demand forecasts may be biased and transport rates may change. Average costs can hide peak-period congestion. Qualitative factors are difficult to quantify, while model objectives can privilege short-term financial savings over resilience or social outcomes. A model may also assume that capacity, labour and transport are always available when real markets are constrained.

Accordingly, quantitative optimisation should support managerial judgement rather than replace it. Decision makers should document assumptions, conduct sensitivity analysis, validate results with operational experts and identify conditions under which the recommendation would change.

Chapter Summary

Warehouse strategy is a network-design discipline linking facilities, inventory, transportation, service, technology and governance to supply-chain objectives. Centralized networks benefit from scale and risk pooling but may create distance and concentration risk. Decentralized networks improve market responsiveness but duplicate facilities and inventory. Hybrid designs can segment products and customers to balance these effects.

Location analysis may use centre-of-gravity, factor-rating, break-even and total-cost or network-optimisation models. These tools require clear assumptions and cannot replace feasibility, compliance and risk assessment. Capacity decisions must address throughput, peak demand, bottlenecks and scalability rather than floor area alone.

Ownership, leasing and 3PL choices involve trade-offs among control, capital, expertise, flexibility and risk. Effective outsourcing requires precise contracts, service measures, audit rights and continuity provisions. Technology and AI can improve visibility and planning, but they require reliable data, cybersecurity and human governance. Ultimately, the best network is not necessarily the one with the lowest nominal cost; it is the one that delivers the required service and resilience at an acceptable total cost while supporting strategic, ethical and sustainability objectives.

Key Terms

Warehouse network: The connected set of facilities, flows, systems and policies used to store and move goods.

Centralized network: A network in which inventory is concentrated in one or a few major facilities.

Decentralized network: A network using multiple geographically dispersed stocking locations.

Hybrid network: A network combining central, regional or specialised facilities.

Risk pooling: Reduction in aggregate variability achieved by combining demand or inventory positions.

Service level: A measure of the probability or extent to which customer demand is fulfilled as promised.

Fill rate: The proportion of demand quantity supplied immediately from available inventory.

Centre of gravity: A weighted-coordinate method for approximating a facility location.

Factor rating: A weighted scoring method for comparing candidate sites.

Total cost: The combined relevant facility, handling, transport, inventory, service and risk costs.

Design capacity: The theoretical maximum output under specified conditions.

Effective capacity: Realistic available capacity after operating constraints are considered.

Utilisation: Actual output divided by a defined capacity measure.

Postponement: Delaying final product configuration until demand information is available.

3PL: A third-party logistics provider performing contracted logistics activities.

Service-level agreement: A contract section defining measurable service requirements and remedies.

Cross-docking: Moving inbound goods to outbound dispatch with little or no storage.

Resilience: The ability of a supply chain to absorb, adapt to and recover from disruption.

Warehouse-management system: Software coordinating warehouse inventory and operating processes.

Network segmentation: Designing different flows, facilities or policies for distinct product or customer groups.

Review Questions

Short Answer Questions

1. What are the principal advantages and disadvantages of a centralized warehouse network?
2. Distinguish cycle service level from fill rate.
3. What is the purpose of the centre-of-gravity method?
4. Define effective capacity and explain why it differs from design capacity.
5. Identify four provisions that should be included in a 3PL service-level agreement.

Long Answer Questions

1. Critically compare centralized, decentralized and hybrid warehouse networks, including inventory, transport, service and resilience implications.
2. Explain how a total-cost model can support warehouse-location decisions and discuss its limitations.
3. Evaluate the strategic considerations involved in owning, leasing or outsourcing a warehouse.
4. Discuss how product and customer segmentation should influence warehouse service requirements and inventory positioning.
5. Explain how technology, sustainability and risk governance are changing warehouse-network design.

Analytical / Application Questions

1. A company serves four markets with different demand volumes and coordinates. Calculate the centre-of-gravity location, then identify at least four practical factors that must be evaluated before selecting a site.
2. An organisation is deciding between one national warehouse and two regional warehouses. Design a comparison framework covering total cost, service, inventory, capacity, disruption risk and sustainability. State which data would be required.
3. A 3PL offers a low storage rate but refuses audit rights and provides no guaranteed recovery time after a major disruption. Assess the proposal and recommend contractual and governance changes.

Multiple Choice Questions with Answers

Question 1

Which feature is most characteristic of a centralized warehouse network?

A. Inventory is concentrated in one or a few major facilities

B. Every customer receives stock from a local warehouse

C. Each region independently selects its own inventory policies

D. Facilities are used only for cross-docking

Correct Answer: A. Inventory is concentrated in one or a few major facilities

Question 2

What is the principal inventory benefit of risk pooling?

A. It eliminates all demand variability

B. It can reduce aggregate safety stock when demands are not perfectly correlated

C. It guarantees a higher fill rate without additional cost

D. It removes the need for replenishment lead-time analysis

Correct Answer: B. It can reduce aggregate safety stock when demands are not perfectly correlated

Question 3

In the centre-of-gravity method, the weighting factor commonly represents:

A. The legal age of the facility

B. The number of warehouse employees

C. Demand, shipment volume or freight-related importance

D. The building’s ceiling height

Correct Answer: C. Demand, shipment volume or freight-related importance

Question 4

Which item is normally part of total warehouse-network cost?

A. Only warehouse rent

B. Only outbound transportation

C. Facility, transport, handling, inventory and relevant service costs

D. Only the purchase price of inventory

Correct Answer: C. Facility, transport, handling, inventory and relevant service costs

Question 5

Why may very high warehouse utilisation be undesirable?

A. It always reduces inventory accuracy

B. It can create congestion and reduce the ability to absorb peaks

C. It prevents all forms of automation

D. It automatically increases facility rent

Correct Answer: B. It can create congestion and reduce the ability to absorb peaks

Question 6

Which is a potential advantage of using a 3PL?

A. The client has no need to define performance measures

B. The client automatically transfers all legal responsibility

C. The client may gain specialist capability and scalable capacity

D. The client is guaranteed permanent control of operations

Correct Answer: C. The client may gain specialist capability and scalable capacity

Question 7

A warehouse has design capacity of 10,000 units per day and processes 8,000 units per day. Its utilisation, using design capacity as the denominator, is:

A. 20%

B. 40%

C. 80%

D. 125%

Correct Answer: C. 80%

Question 8

Which statement about the centre-of-gravity result is correct?

A. It is always the final legally feasible site

B. It ignores all demand weights

C. It is an initial geographic approximation requiring practical screening

D. It can be used only when all markets have identical demand

Correct Answer: C. It is an initial geographic approximation requiring practical screening

Question 9

What is postponement?

A. Delaying final product configuration until demand information is available

B. Eliminating all regional inventory

C. Increasing safety stock at every facility

D. Replacing warehouse management with manual records

Correct Answer: A. Delaying final product configuration until demand information is available

Question 10

Which contract provision is especially important in a strategic 3PL relationship?

A. A prohibition on measuring delivery performance

B. Undefined liability for inventory loss

C. Business-continuity requirements and audit rights

D. An agreement that excludes data ownership

Correct Answer: C. Business-continuity requirements and audit rights