Industries
Choosing the Right CX Pricing Model
Choosing the Right CX Pricing Model
Key Takeaways
By Andy Schachtel, CEO of Sourcefit | Global Talent and Elevated Outsourcing
- There is no universally best pricing model for outsourced CX; the right model depends on the client’s volume predictability, quality requirements, tolerance for cost variability, and the degree to which they want visibility into what they are paying for, which means that choosing a pricing model is a strategic decision that should be made with the same rigor as choosing the provider itself.
- Cost-plus pricing, where the client pays the actual cost of their team plus a transparent management fee, offers the highest visibility and alignment of interests because the provider has no incentive to cut corners on quality or staffing to protect hidden margins, but it requires the client to trust the provider’s cost reporting and to accept that costs will fluctuate with headcount changes.
- Per-agent pricing offers budget predictability and simplicity, making it the most common model in traditional BPO, but it can create misaligned incentives where the provider profits from understaffing or overloading agents because the revenue per agent is fixed regardless of how many interactions each agent handles.
- Output-based pricing ties cost directly to results, which sounds ideal but requires extremely precise definition of what constitutes a completed output, robust measurement systems, and shared agreement on quality thresholds, because without these controls, the provider is incentivized to maximize volume at the expense of quality.
In 2021, a rapidly growing Asia-Pacific fintech platform signed a per-agent contract with a CX provider at a competitive monthly rate for a team of 15. The rate looked reasonable. The provider assured them that 15 agents would handle their projected volume across compliance review, customer service, and executive assistance comfortably. Six months in, the company discovered that actual volume had grown significantly as they expanded into new markets. The provider had not added agents because the contract was per-agent, not per-interaction. Handle times had been compressed to compensate. Quality scores had dropped. Customer satisfaction had declined. The provider was earning the same revenue while delivering worse service.
When the company complained, the provider offered to add more agents at the same per-agent rate, increasing the monthly cost substantially. The company realized they were paying more to fix a problem the pricing model had created. The per-agent rate had appeared simple and predictable. In practice, it had misaligned the provider’s incentives with the client’s interests. The company subsequently transitioned to a fully managed service model with cost-plus pricing, where they paid actual team costs plus a transparent management fee. Under the new model, quality scores improved to 96%, CSAT reached 97%, and productivity increased by 40%, because the provider’s incentive was aligned with running the operation well rather than with maintaining a fixed headcount regardless of quality.
Pricing model conversations in outsourced CX are often treated as procurement details, negotiated by finance teams after the operational decision has been made. This is a mistake. The pricing model shapes the provider’s behavior, the client’s risk exposure, and the quality of service delivered. It is not an administrative detail. It is the structural incentive system that governs the entire relationship.
Cost-Plus: Maximum Transparency, Maximum Trust
In a cost-plus model, the client pays the actual, fully loaded cost of their CX team, including salaries, benefits, office space, technology, management, and overhead, plus a fixed management fee that represents the provider’s margin. The management fee is typically a flat monthly amount per employee or a percentage of total costs, and it is disclosed upfront.
The advantage of cost-plus is transparency. The client sees exactly what they are paying for. Every line item is visible: what each agent earns, what the benefits cost, what the office space and technology add, and what the provider charges for management and profit. There are no hidden margins embedded in a blended rate. If the client wants to increase compensation to reduce attrition, they see the cost of that decision directly. If the provider recommends adding a team leader, the cost is clear.
Cost-plus aligns interests in a way that other models do not. The provider’s margin is fixed regardless of how many agents are on the team or how hard they work. The provider has no incentive to understaff because adding or removing agents does not change the management fee structure. The provider has no incentive to compress handle times or cut corners on quality because those savings do not flow to the provider’s bottom line. The provider’s incentive is to run the operation well enough that the client renews the contract, which means the provider is motivated to optimize for quality and retention rather than for margin extraction.
The requirement is trust. The client must trust that the provider’s reported costs are accurate and that the overhead allocation is fair. Reputable providers address this through open-book accounting, regular cost reviews, and the client’s ability to audit cost reports. For clients who value transparency and want a partnership-oriented relationship, cost-plus is often the strongest structural foundation.
Per-Agent: Simplicity with Hidden Risks
Per-agent pricing is the traditional BPO model: the client pays a fixed monthly rate for each agent on their team. The rate includes everything, salary, benefits, overhead, management, technology, and the provider’s margin, bundled into a single number. The client knows their monthly cost with precision: 25 agents at $2,800 equals $70,000.
The simplicity is genuine and valuable. Budget forecasting is straightforward. Procurement can compare providers on a single number. Invoicing is clean. For companies that value predictability above all else, per-agent pricing delivers it.
The risks are structural. A per-agent rate incentivizes the provider to maximize the margin within each agent by controlling costs the client cannot see. If the provider can reduce the actual cost of an agent by 10% through lower compensation, fewer benefits, or reduced training investment, that saving flows directly to the provider’s profit. The client sees the same rate and the same headcount. They may not see the quality erosion that results from paying agents below market or cutting training from four weeks to two.
The volume mismatch problem is the more common failure mode. Per-agent pricing does not account for interaction volume. If volume increases, the same agents handle more interactions, which increases handle pressure, reduces quality, and degrades the customer experience. The provider has no financial incentive to flag the problem because adding agents increases their costs without increasing their per-agent margin. The client discovers the problem through declining satisfaction scores, not through the provider’s proactive communication.
Per-Hour: Flexibility with Monitoring Requirements
Per-hour pricing charges the client for actual agent hours worked rather than for a fixed headcount. If an agent works 160 hours in a month at $18 per hour, the charge is $2,880. If volume drops and the agent works 120 hours, the charge drops to $2,160. The cost flexes with actual utilization.
The flexibility advantage is significant for operations with variable demand. Seasonal businesses, companies with unpredictable volume patterns, and organizations that are scaling up or down benefit from a model that does not lock them into fixed headcount during low-volume periods. Per-hour pricing eliminates the waste of paying for idle agents during slow periods and provides a natural mechanism for scaling during peaks.
The monitoring requirement is the tradeoff. Per-hour pricing requires reliable time tracking that both parties trust. Disputes over hours worked, productive versus non-productive time, and the definition of billable activities are common in per-hour arrangements. Clear definitions of what constitutes a billable hour, agreed time-tracking systems, and regular reconciliation processes are essential. Without them, the model produces invoicing disputes that erode the relationship.
Per-hour pricing can also create a subtle incentive misalignment: the provider earns more revenue when agents take longer to resolve interactions. A per-hour provider has no financial incentive to improve efficiency because faster resolution means fewer billable hours. This does not mean providers deliberately slow their agents down. It means the structural incentive does not reward the efficiency improvements that benefit the client. Pairing per-hour pricing with quality and efficiency KPIs that carry financial consequences addresses this misalignment.
Output-Based: Results-Oriented with Definitional Challenges
Output-based pricing charges the client for completed work units rather than time or headcount. A completed chat session, a resolved ticket, a processed application, a qualified lead. The client pays per unit of output, which directly ties cost to productivity. If the team resolves 10,000 tickets in a month at $3.50 per ticket, the cost is $35,000 regardless of how many agents worked or how many hours they logged.
The appeal is obvious. The client pays for results, not effort. If the provider can resolve 10,000 tickets with 20 agents instead of 25, the client’s cost is the same and the provider’s margin improves. The provider is incentivized to optimize efficiency because efficiency gains flow to their bottom line. In theory, this creates the tightest alignment between cost and value.
In practice, output-based pricing is the most complex model to implement correctly. The definition of a completed output must be precise enough to prevent gaming. Does a “resolved ticket” include tickets that the customer reopens within 48 hours? Does a “completed chat” include chats where the customer abandoned before the agent resolved the issue? Does a “qualified lead” include leads that sales subsequently disqualifies? Every ambiguity in the output definition becomes a potential dispute and a potential incentive for the provider to maximize counted outputs at the expense of actual quality.
Quality thresholds are essential in output-based models. Without them, the provider is incentivized to resolve tickets as quickly as possible, counting volume regardless of whether the resolution was complete or the customer was satisfied. A well-designed output-based contract includes a quality gate: only outputs that meet a minimum quality standard count toward the billable volume. This requires a QA framework that can evaluate output quality at scale and a shared agreement on what constitutes acceptable quality.
Pricing Models Compared
| Dimension | Cost-Plus | Per-Agent | Per-Hour | Output-Based |
|---|---|---|---|---|
| Cost Predictability | Moderate; varies with headcount and actual costs | High; fixed rate per agent | Moderate; varies with hours worked | Variable; depends on output volume |
| Transparency | Highest; open-book cost visibility | Low; costs bundled into blended rate | Moderate; hourly rate visible, margin hidden | Low; provider margin embedded in per-unit rate |
| Provider Incentive | Quality and retention (margin is fixed) | Margin protection (reduce hidden costs) | More hours billed (lower efficiency incentive) | Volume maximization (quality risk without gates) |
| Quality Risk | Low; no incentive to cut corners | Moderate; hidden cost reduction can erode quality | Low to moderate; time-based, not quality-based | High without quality gates; low with them |
| Volume Flexibility | High; team scales with need | Low; fixed headcount regardless of volume | High; hours flex with demand | Highest; cost scales directly with output |
| Best For | Partnership-oriented clients valuing transparency | Clients wanting budget simplicity and predictability | Variable-volume operations; seasonal businesses | High-volume transactional operations with clear outputs |
| Implementation Complexity | Moderate; requires cost reporting infrastructure | Low; simple rate card | Moderate; requires time tracking and reconciliation | High; requires output definition, QA gates, measurement |
Choosing the Right Model for Your Operation
The pricing model should be chosen based on four factors: volume predictability, quality sensitivity, budget flexibility, and relationship orientation. Companies with stable, predictable volume and a strong preference for budget certainty may find per-agent pricing acceptable if they negotiate volume-based adjustment clauses. Companies with variable volume benefit from per-hour or output-based models that flex with demand. Companies that prioritize quality and want a collaborative partnership tend to gravitate toward cost-plus because the transparency and incentive alignment support a long-term, quality-focused relationship.
Hybrid models are increasingly common and often the best answer. A cost-plus base for the core team, ensuring transparency and quality alignment, combined with a per-hour or output-based component for overflow and peak capacity, provides stability for the baseline operation and flexibility for demand variation. The hybrid approach captures the advantages of multiple models while mitigating the risks of relying on any single one.
The most important factor is not which model you choose but whether you understand the incentives it creates. Every pricing model has structural incentives that influence provider behavior. The client who understands those incentives can design contractual protections, monitoring mechanisms, and performance standards that keep the provider’s behavior aligned with the client’s interests regardless of the pricing structure. The client who chooses a model based solely on the quoted rate without understanding the incentives is optimizing for the wrong variable.
Frequently Asked Questions
Can we switch pricing models during a contract if the initial choice is not working?
Yes, but the transition should be planned carefully. Switching from per-agent to cost-plus requires the provider to open their books and restructure their billing, which some providers resist because it exposes margins they prefer to keep private. Switching from per-hour to output-based requires building the output definition, measurement, and quality gate infrastructure before the switch. Most contracts include annual reviews where pricing model changes can be negotiated. The smoothest transitions happen when both parties agree the current model is not serving the relationship and collaborate on the alternative rather than one party imposing a change unilaterally.
How do we prevent providers from gaming output-based pricing?
Three controls are essential. First, define outputs precisely, including what does and does not count, how rework and reopened cases are handled, and how partial completions are treated. Second, implement a quality gate that requires outputs to meet a minimum quality score before they count toward billable volume. Third, include a customer satisfaction threshold that the provider must maintain; if CSAT drops below the threshold, the per-unit rate is reduced or a penalty applies. These controls create a system where the provider can only increase revenue by increasing the volume of quality outcomes, which aligns their incentive with the client’s interest.
Is cost-plus always more expensive than per-agent pricing?
Not necessarily. Cost-plus reveals the actual cost, which may be lower than the per-agent rate a provider quotes because the per-agent rate includes a margin buffer that accounts for the provider’s risk of cost overruns. In a cost-plus model, the client pays actual costs plus a known fee. In a per-agent model, the client pays a rate that includes a hidden margin sized to protect the provider against worst-case cost scenarios. For stable operations where costs are predictable, cost-plus often produces a lower total cost because the provider does not need to embed risk premiums in the rate.
Which pricing model works best for a new outsourcing engagement where volume is uncertain?
Per-hour pricing is typically the best starting point for new engagements with uncertain volume because it provides flexibility without requiring the precise output definitions that an output-based model demands. The client pays for actual hours utilized, which accommodates the volume fluctuation that is normal during the first three to six months of a new engagement. After the engagement stabilizes and volume patterns become predictable, transitioning to cost-plus or a hybrid model that provides more transparency and better incentive alignment is a common and effective evolution.
How should we compare pricing across providers who offer different models?
Normalize the comparison to a common denominator: total cost per resolved interaction at a defined quality standard. Ask each provider to estimate the total monthly cost for your projected volume and quality requirements, regardless of their pricing model. Divide by the projected number of resolved interactions to get a comparable cost-per-resolution figure. Then evaluate the quality risk, transparency, and incentive alignment of each model separately. The provider with the lowest cost-per-resolution may have the highest quality risk if their model incentivizes volume over quality. The provider with the highest transparency may cost slightly more per resolution but produce better outcomes over 24 months.
To learn more about how SourceCX offers flexible pricing models designed to align provider incentives with client outcomes, visit sourcecx.com or contact our team for a consultation.
