Scale Without Headcount: Rethinking the Economics of Customer Operations

For much of the history of business process outsourcing, scale has meant people.

When transaction volumes increased, providers recruited more agents. As a client grew, the operation expanded alongside it. Capacity and headcount grew together. The economics were therefore relatively predictable: more work meant more people, and more people meant more cost. 

Technology is beginning to break that relationship.

For Quantanite Chief Operating Officer Gavin Atkinson, that creates an opportunity to rethink not only how customer operations scale but also how their performance should be measured in the first place.

Stop measuring transactions. Start measuring outputs.

Atkinson sees measurement as one of the most important changes ahead for the industry.

“The way we measure has traditionally been transactional, rather than focused on outputs.”

Historically, transactional measurement made sense. Providers were selling capacity. Clients were purchasing people, seats or transactions. Performance metrics naturally evolved around the efficiency of that model.

But when technology begins performing part of the work, the transaction becomes a less useful unit of value.

Atkinson believes the industry needs to move “from a cost per seat or a cost per transaction to a cost per output or a cost per function that we need to fulfil.”

That sounds like a relatively small change in terminology. It isn’t.

• A seat-based model asks: How much capacity did we provide?
• A transaction-based model asks: How much work did we process?
• An output-based model asks: What did we accomplish?

Those questions create very different incentives.

Design capacity around the work, not the org chart

In a traditional contact centre, capacity planning has largely meant workforce planning.

• Forecast the expected volume
• Estimate handle time
• Calculate staffing requirements
• Recruit and schedule enough people to meet demand

That model assumes that human labour is the primary source of capacity. Once technology becomes capable of performing meaningful parts of the workflow, capacity becomes a different problem.

The question is no longer simply how many people are required. It becomes, “What is the most effective way to fulfil this function?” Some activities may remain entirely human. Some can be supported by technology. Some can be automated. And the optimal combination may change as demand, complexity, and technology change. 

This approach creates the possibility of an operation that is considerably more flexible than one built predominantly around fixed human capacity.

Resolution changes the unit economics

Changing the unit of measurement also changes what operational efficiency means. Take the service level: 

Traditionally, the customer services director would report back on service levels and assess whether we answered calls within X amount of time, indicating that we were doing well.

That tells an operations leader something useful about accessibility and capacity. But it does not tell them whether the work achieved its purpose. Atkinson argues for a different emphasis: “It’s more about, did I resolve that customer query? Number one, that customer doesn’t have to interact with me again.”

From an operating perspective, this matters because unresolved work creates more work. A transaction may look complete in the system while the underlying requirement remains open. The next interaction then creates additional volume, additional handling, and additional cost. 

Measuring output forces the operation to account for the full cost of producing the required result rather than simply the efficiency of completing each individual transaction. 

That is a more demanding measure. It is also a more useful one.

Customer experience and cost belong in the same operating equation

The traditional efficiency conversation often presents customer experience and cost as opposing forces. Atkinson believes the operation should be accountable for both.

“We need to be improving the customer experience but reducing our cost of service.”

That creates a more useful test of operational improvement. If an initiative lowers cost but materially damages the customer experience, it is not a sustainable improvement. However, if the experience improves but costs grow proportionally with every increase in demand, scaling may be difficult.

The operating challenge is to strengthen the relationship between the two. Technology creates new possibilities for doing that because it introduces capacity that does not behave economically like human capacity.

Technology breaks the link between growth and headcount

This area is where the biggest structural change occurs. Traditional BPO operations face a basic reality: recruiting, onboarding, training and managing people takes time.

When demand increases quickly, scaling human capacity is both expensive and slow. Atkinson puts it simply:  “Traditionally, if a company wanted to scale, that scale came at a high cost.”

A blended operating model changes what is possible: “By doing the blend of having our humans and our tech, scale now becomes A, much faster, and B, at a much reduced cost because you’re able to blend the two.”

This is a more useful way to think about AI in customer operations than simply asking how many agents it can replace. The operational question is not human versus technology. It is about how to construct capacity. If volume rises significantly, does the organisation need to add human resources at the same rate? Or can technology absorb a greater share of incremental demand? Can the operating model flex when volumes spike? Can more work move through the same underlying cost base?

Those questions determine whether technology is creating genuine operating leverage rather than simply automating isolated tasks.

The goal is a more elastic operation

This approach has major implications for BPO delivery. Historically, a large increase in client volume required a corresponding exercise in workforce expansion. Recruitment had to increase. Training capacity had to increase. Management structures expanded. Facilities and infrastructure potentially expanded with them. 

The operation could scale, but its cost base scaled too. A technology-enabled operation should behave differently.

Human capacity can be concentrated around the work where it is most valuable, while technology provides a more elastic layer of capacity around it. That allows the operation to respond to growth and volatility without rebuilding the workforce every time demand changes.

It also changes the economics of the provider-client relationship. The objective becomes less about providing a predetermined quantity of resources and more about delivering the required output at the best combination of quality, speed and cost.

The real test of scale

For years, the BPO industry has demonstrated scale through headcount.

• How many agents can we recruit?
• How quickly can we stand up a team? 
• How many seats can we operate?

Those capabilities still matter. But they are becoming an incomplete definition of scale. The more important test is how much additional business an operation can support without requiring an equivalent increase in resources.

• Can volume double without cost doubling?
• Can the operation absorb a sudden spike without a major recruitment cycle?
• Can technology take on incremental work while human expertise remains focused where it creates the highest value?
• Can the cost of delivering each output fall as the operation grows?

That is operating leverage. And it changes the economics of customer operations fundamentally. 

The traditional model made growth expensive because capacity and headcount were inseparable. More demand meant more people, and more people meant more cost.

Technology gives customer operations the opportunity to break that relationship. The real measure of progress will not be how many interactions an operation can handle or how many agents it can deploy. It will be how much additional demand it can absorb, how consistently it can deliver the required outcome, and how little incremental cost it incurs.

That is a fundamentally different model of scale. And ultimately, a fundamentally different model for BPO.