The most common commercial model in Indian BPO work is still the FTE (full-time equivalent) rate: the client pays a fixed monthly amount for each agent seat dedicated to its process. It looks simple from the outside. Inside, the rate is a stack of costs, assumptions and a margin, and each layer is a place where a pricing team can get it wrong. Let us build one from the bottom up.
Layer one: the fully loaded agent cost
Salary is only the starting point. A fully loaded cost per agent per month usually includes:
- Gross salary plus statutory contributions such as employer PF and ESI where applicable, and a provision for gratuity and bonus.
- Facility cost per seat: rent, power, housekeeping, security, allocated across seats and shifts.
- Technology per seat: desktop or thin client, headset, telephony licences, CRM licence if the BPO provides it, network bandwidth.
- Transport for night shifts, which is significant for international processes in Indian cities and often a legal or company-policy requirement for women working late shifts.
- Recruitment and initial training, amortised over the expected tenure of an agent.
Suppose all of this comes to ₹32,000 per agent per month for a domestic voice process in a tier-2 city. That number already embeds an assumption about attrition: if agents stay only eight months instead of fourteen, the amortised hiring and training cost per month goes up.
Layer two: the people who do not take calls
A process of 80 agents also needs team leaders, quality analysts, trainers, a workforce planner, an MIS analyst and some share of an operations manager. These roles are rarely billed separately in an FTE model, so their cost is spread over the agent seats.
The overhead depends on spans of control. A team leader for every 15 to 20 agents and a quality analyst for every 25 to 40 agents are common ranges, but clients with heavy compliance requirements may insist on tighter ratios. Say the support layer adds 22% to agent cost. The loaded cost per billable seat becomes ₹32,000 × 1.22 = ₹39,040.
Layer three: margin, and the markup mistake
The company wants a 25% gross margin on the account. This is where a surprising number of pricing sheets go wrong.
Markup is profit as a percentage of cost. Margin is profit as a percentage of revenue. Adding 25% to cost gives ₹39,040 × 1.25 = ₹48,800. The profit is ₹9,760, which is 25% of cost but only 20% of the price. The target was missed by five points before the contract was even signed.
To achieve a 25% margin, divide cost by (1 − margin): ₹39,040 ÷ 0.75 = ₹52,053. Now profit is ₹13,013, which is exactly 25% of the price. For 80 seats, monthly revenue would be about ₹41.6 lakh against costs of about ₹31.2 lakh.
| Approach | Rate per FTE | Actual margin |
|---|---|---|
| Cost + 25% markup | ₹48,800 | 20.0% |
| Cost ÷ 0.75 (25% margin) | ₹52,053 | 25.0% |
The gap widens as the target rises. A 40% markup gives only 28.6% margin. Anyone reviewing a quote should ask which of the two the sheet means.
The assumptions hiding in the rate
The arithmetic is the easy part. The rate is only as good as the assumptions behind it, and these are the ones that most often turn a good-looking account into a loss-maker.
What counts as a billable FTE. Some contracts bill per logged-in seat; others bill only for agents who are production-ready, excluding trainees in nesting. If new joiners take six weeks to reach production and attrition is high, a meaningful share of heads is always non-billable. If, say, an extra 8% of heads is needed to cover trainees and buffer, the cost to recover per billable seat rises to about ₹42,163, and the 25%-margin price becomes about ₹56,218. A quote built on 100% billability will undershoot.
Productivity expectations. An FTE rate implicitly promises a certain output. If the client expects each FTE to handle a volume of calls based on a lower handle time than the process actually needs, the BPO will have to add unbilled agents to hit the service level.
Salary inflation and contract length. A three-year contract priced on today’s salaries will be squeezed by annual increments unless there is a cost-of-living adjustment clause. Many master agreements include an annual rate revision linked to an index or a fixed percentage.
Currency. International contracts are often priced in US dollars or pounds while costs are in rupees. A stronger rupee shrinks the margin. Larger providers hedge; smaller ones sometimes discover this risk the hard way.
Ramp-up and transition costs. The first two or three months involve knowledge transfer, travel, parallel runs and lower productivity. These are either billed as a one-time transition fee or spread into the rate. Leaving them out entirely is a common reason why year-one profitability disappoints.
When FTE pricing stops being the right model
FTE pricing puts productivity risk on the client: if handle time rises, the client needs more seats and pays for them. It also gives the provider little reason to reduce effort, since fewer seats mean less revenue. That tension becomes sharp when automation or AI tools reduce handle time.
Alternatives include per-transaction pricing (a rate per call, chat or processed document), per-hour pricing on actual logged hours, and outcome-based pricing tied to results such as resolved cases or collections recovered. Each shifts risk differently. Transaction pricing makes the provider carry productivity risk, so the provider then needs reliable volume forecasts and minimum volume commitments from the client. Outcome-based pricing can align incentives well but needs clean, agreed measurement, which is harder than it sounds.
Hybrid models are common: a fixed base for a core team plus a variable component for volume above a band, sometimes with gain-sharing when efficiency improves. Whatever the model, the same discipline applies. Know the true loaded cost, separate margin from markup, and write down every assumption the rate depends on so it can be revisited at renewal.
Written by the RP Strativa Team. Figures in the examples are illustrative.
Leave a Reply