[BPO Insights] By 2028, 50% of BPO Revenue Will Come From Technology Margins, Not Labor

Last reviewed: February 2026

The Revenue Mix That Defines an Industry

The BPO industry generates approximately $280 billion in annual revenue globally. Roughly 95% of that revenue comes from one source: charging clients for human labor.

The model is simple. Hire agents. Train agents. Put agents in seats. Bill clients for the hours those agents work. The revenue is labor revenue. The margin is the spread between what the client pays per hour and what the agent costs per hour. The entire industry — its operations, its pricing, its real estate, its org charts — is structured around this single revenue mechanism.

Labor margins in BPO run 15-25% gross, depending on geography, vertical, and scale. Offshore delivery pushes margins toward the higher end. Onshore delivery compresses them. Specialized verticals (healthcare, financial services) command slight premiums. Commodity verticals (general customer service, back-office processing) compress margins further.

This margin structure has been stable for two decades. And it's about to break.

By 2028, I believe 50% of revenue at the surviving BPOs will come from technology margins — not labor. The sources: AI platform fees, outcome-based pricing spreads, technology licensing to clients, and managed AI services. These revenue streams carry 60-80% gross margins. Four to five times higher than labor margins.

This isn't a prediction about what would be nice. It's a prediction about what's structurally necessary. Here's why.

Why the Shift Is Structural, Not Optional

Three forces are making the revenue mix shift inevitable.

Force 1: Labor margin compression is accelerating. The spread between client rates and agent costs is narrowing from both directions. Client rates are under pressure from AI alternatives — enterprise buyers know that AI can handle routine interactions at a fraction of the cost of human agents, so they're renegotiating rates downward. Agent costs are rising — wage inflation in major delivery markets (Philippines, India, Colombia, South Africa) runs 4-8% annually. Benefits costs are increasing. Regulatory requirements around worker protections are expanding.

A BPO that earned 25% gross margin on labor five years ago is earning 18-20% today. In five more years, that margin hits 12-15% for commodity work. At 12% gross margin, the business model doesn't support corporate overhead, technology investment, and growth. It barely supports survival.

Force 2: AI creates a technology revenue opportunity that didn't exist before. Before AI, a BPO's technology was operational infrastructure — telephony, CRM, workforce management. These were costs, not revenue sources. The BPO didn't charge clients for its technology stack. It charged clients for the humans who used the technology stack.

AI changes this equation. When a BPO deploys an AI agent that handles 40% of a client's call volume, that AI agent is a product, not infrastructure. The BPO built it (or licensed and configured it). The BPO maintains it. The BPO improves it. The client receives a service — AI-handled customer interactions — that has a clearly attributable cost and value.

The BPO can charge for that service at technology margins, not labor margins. The AI agent costs the BPO $0.06-$0.15 per minute to operate. The BPO charges the client $0.30-$0.60 per minute (or an equivalent outcome-based rate). The gross margin on AI-delivered interactions is 60-80%. Four to five times the gross margin on human-delivered interactions.

Force 3: The surviving BPOs will be the ones who make the shift. This is the evolutionary pressure argument. BPOs that remain 95% labor-revenue businesses will face continuous margin compression until the business is unviable. BPOs that shift 30-50% of revenue to technology margins will have a blended margin structure that supports investment, growth, and competitive pricing.

The market will select for the shifted model. Clients will prefer BPOs that offer AI-augmented services because the economics are better and the capabilities are broader. Investors will prefer BPOs with technology revenue because the margins are higher and the growth is more scalable. Talent will prefer BPOs that invest in technology because the work is more interesting and the career path is more durable.

Natural selection applies to business models. The labor-only BPO is heading toward the same fate as the labor-only manufacturing company or the labor-only logistics company. The survivors integrate technology into the revenue model.

The Revenue Mix Evolution: 2024, 2026, 2028

Here's how the revenue mix shifts across three time horizons.

2024 Revenue Mix (Current State)

Labor revenue: 95%+ of total revenue. Human agents handling customer interactions, billed per seat or per hour.

Technology revenue: Less than 5% of total revenue. Primarily limited to reporting dashboards, analytics tools, or minor automation features that some BPOs charge incremental fees for.

Blended gross margin: 18-22%.

The typical 2024 BPO P&L: $50M revenue. $40M cost of delivery (agent wages, facilities, management). $10M gross profit. $7M SG&A. $3M operating income. 6% operating margin.

2026 Revenue Mix (Transition State)

Labor revenue: 75-80% of total revenue. Still the majority, but declining as AI handles an increasing share of interactions.

Technology revenue: 20-25% of total revenue. Sources: AI platform fees charged to clients for AI-handled interactions, outcome-based pricing spreads (the difference between what the BPO charges per resolution and what the AI costs per resolution), and managed AI service fees for deployment, monitoring, and optimization.

Blended gross margin: 28-35%.

The 2026 transition BPO P&L: $50M revenue ($38M labor, $12M technology). $28M cost of delivery. $22M gross profit. $9M SG&A (higher because of technology investment). $13M operating income. 26% operating margin.

Same top-line revenue. More than 4x the operating income. The technology revenue doesn't just add margin — it transforms the operating model.

2028 Revenue Mix (Target State)

Labor revenue: 50% of total revenue. Human agents handle complex, high-value, relationship-driven interactions. The work is harder, the agents are more skilled, and the rates are higher because the commodity work has been automated.

Technology revenue: 50% of total revenue. Sources: AI platform fees (largest component), outcome-based pricing (growing component), technology licensing to clients who want to bring AI in-house but need the BPO's platform and expertise (emerging component), and managed AI services including conversation design, optimization, and compliance management.

Blended gross margin: 40-50%.

The 2028 target BPO P&L: $65M revenue ($32M labor at higher rates, $33M technology). $32M cost of delivery. $33M gross profit. $12M SG&A. $21M operating income. 32% operating margin.

Top-line growth from $50M to $65M. Operating income growth from $3M to $21M. Seven times the operating income on 30% more revenue. The margin expansion is the story, not the top-line growth.

The Four Technology Revenue Streams

The 50% technology revenue target comes from four distinct streams, each with different economics and maturity levels.

Stream 1: AI Platform Fees (35-40% of technology revenue)

The most straightforward stream. The BPO deploys AI agents for clients and charges a per-minute or per-interaction fee for AI-handled volume. The BPO pays the AI platform vendor $0.06-$0.15 per minute. The BPO charges the client $0.30-$0.60 per minute. Gross margin: 60-75%.

This stream scales directly with AI adoption across the client base. As more interactions shift from human to AI handling, this revenue stream grows. The BPO's role is deployment, configuration, quality monitoring, and ongoing optimization — all of which are technology services, not labor services.

Stream 2: Outcome-Based Pricing Spreads (25-30% of technology revenue)

When a BPO charges per resolution rather than per minute, the spread between the outcome price and the delivery cost becomes technology revenue. The client pays $1.50 per appointment scheduled. The AI costs $0.12 per appointment scheduled (2 minutes at $0.06/minute). The spread is $1.38 per resolution at 92% gross margin.

This stream rewards the BPO for quality and efficiency, not volume. Better AI means lower cost per resolution, which means higher margin on the same client rate. The BPO's incentive shifts from "handle more calls" to "resolve more efficiently" — which also aligns with what the client wants.

Stream 3: Technology Licensing (15-20% of technology revenue)

This is the emerging stream that most BPOs haven't considered yet. Some enterprise clients will eventually want to bring AI capability in-house rather than running it through a BPO. Instead of losing these clients entirely, the BPO licenses its AI platform — the conversation designs, the compliance frameworks, the integration layer, the monitoring tools — to the client on a SaaS basis.

The client gets the technology without the BPO's operational layer. The BPO gets recurring technology revenue at 70-85% gross margins without the cost of delivery operations. It's a different relationship — vendor rather than operator — but it preserves revenue from clients who would otherwise churn entirely.

Stream 4: Managed AI Services (15-20% of technology revenue)

Consulting-adjacent services that support AI deployment: conversation design, prompt engineering, compliance configuration, quality monitoring, performance optimization, and analytics. These are professional services, but they're technology professional services with margins of 50-65% — significantly higher than labor delivery margins.

This stream is valuable because it's sticky. A client that uses the BPO for managed AI services develops dependency on the BPO's AI expertise. Switching costs are high because the AI configurations, conversation flows, and optimization history are specific to the BPO's platform and approach.

The Math That Changes Everything

The critical calculation is how the blended margin shifts as the revenue mix changes.

Consider a BPO with $50M in annual revenue. Here's what happens to their margin at different technology revenue percentages:

0% technology revenue (today): $50M labor revenue at 20% gross margin = $10M gross profit. Blended margin: 20%.

10% technology revenue: $45M labor at 20% + $5M technology at 65% = $9M + $3.25M = $12.25M gross profit. Blended margin: 24.5%.

20% technology revenue: $40M labor at 20% + $10M technology at 65% = $8M + $6.5M = $14.5M gross profit. Blended margin: 29%.

30% technology revenue: $35M labor at 20% + $15M technology at 65% = $7M + $9.75M = $16.75M gross profit. Blended margin: 33.5%.

50% technology revenue: $25M labor at 20% + $25M technology at 65% = $5M + $16.25M = $21.25M gross profit. Blended margin: 42.5%.

The inflection point is around 20-25% technology revenue. At that level, the blended margin crosses 30% — a threshold that transforms the BPO from a low-margin staffing business into a mid-margin technology-enabled services business. Different valuation multiples. Different investor interest. Different strategic options.

A BPO doesn't need to reach 50% technology revenue to transform their economics. Getting to 30% changes the business fundamentally. The path from 0% to 30% is the strategic priority. The path from 30% to 50% happens naturally as AI adoption accelerates across the client base.

What Has to Change to Get There

The revenue mix shift isn't just a pricing strategy change. It requires structural evolution in how BPOs operate.

Commercial model. BPOs need to learn how to price, sell, and contract technology revenue. Today, most BPO sales teams sell seats. They know how to calculate seat-based pricing, negotiate seat-based contracts, and manage seat-based delivery. Selling AI platform fees and outcome-based pricing requires different skills: value-based pricing, technology ROI modeling, and SaaS contract structures.

Talent model. Technology revenue requires technology talent. Conversation designers, AI engineers, data analysts, compliance automation specialists. These aren't roles that exist in most BPO org charts today. Building the talent base takes 12-24 months and requires competing for talent against technology companies that pay technology salaries.

Investment model. Technology revenue requires technology investment — in AI platforms, in integration infrastructure, in quality monitoring systems, in simulation and testing tools. This investment has to come from current operating cash flow or external capital. BPOs that can't fund the investment will be locked out of the technology revenue opportunity.

Client relationship model. The relationship shifts from "we provide you with agents" to "we provide you with AI-augmented customer experience." This is a different conversation with a different stakeholder. The agent conversation happens with the VP of Operations. The AI conversation happens with the CTO, the CIO, and the VP of Customer Experience. BPOs need to sell up and across the client org chart.

The Valuation Implication

There's a reason this matters beyond operating economics.

BPOs with labor-only revenue trade at 5-8x EBITDA. Low margins, labor-intensive operations, limited scalability, client concentration risk. The multiples reflect the structural limitations of the model.

Technology companies trade at 15-25x EBITDA (or higher for SaaS models). High margins, scalable delivery, recurring revenue, defensible intellectual property.

A BPO that shifts 50% of revenue to technology margins doesn't just double their operating income. It changes their valuation framework. Acquirers and investors start applying technology multiples to the technology revenue stream and labor multiples to the labor revenue stream. The blended valuation multiple expands.

A $50M BPO at 6% operating margin and 6x EBITDA is worth approximately $18M. A $65M BPO at 32% operating margin with 50% technology revenue might command a blended 10-12x multiple, valuing the business at $200M+.

Same founder. Same team. Same clients. Different revenue mix. Ten times the enterprise value.

That's the prize. And it's available to any BPO operator who makes the shift in the next 24 months.

Richard Lin is the CEO and founder of Anyreach, an agentic AI platform for enterprise CX.