The salary increases that were frozen across the Indian services industry last year have been reinstated, attrition is climbing back toward pre-pandemic levels and past them in digital skills, and recruiters are quoting premiums for exactly the profiles that back-office delivery centres depend on. Meanwhile a large number of outsourcing contracts are sitting in their third or fourth year, priced per seat, with an annual productivity discount of two or three per cent written into the schedule. Those two facts cannot both hold. Something in the offshore cost model is about to give, and the organisations that look at it this quarter will negotiate from a much better position than the ones that wait for the renewal letter.
The arbitrage was never only about wages
The original business case compared a salary here with a salary there and produced a number large enough that nobody interrogated the rest of the model. The rest of the model is where the erosion is happening. Four variables sit between the wage gap and the saving you actually bank: Replacement cost. Recruiting, onboarding and the eight to fourteen weeks before a new processor is genuinely productive on your work. At fifteen per cent annual attrition this is a rounding error. At thirty per cent it is a standing tax on the account, and it is paid in quality as well as cost. Span of control. High-churn teams need more supervision, more checking and more quality assurance. A team with a fifth of its members inside their ramp period does not run at a one-to-twelve supervisory ratio. Retained management. The onshore people who specify, govern, escalate and explain. This cost never appears in the comparison and never goes away. Rework and escape rate. Errors found downstream cost several times what they cost at the point of capture, and error rates during ramp are materially higher than steady state. Put plausible numbers through that and a seat bought at a quarter of the onshore salary does not deliver a seventy-five per cent saving. It delivers something closer to half of that — and in a year when wages move by double digits and attrition moves with them, the half shrinks while the contract assumes it is growing.
| Cost input | Account-level evidence to request |
|---|---|
| Replacement | Recruiting, onboarding and time to productive work. |
| Supervision | Checking, quality assurance and management effort during ramp-up. |
| Retained management | Client-side specification, governance and escalation work. |
| Rework | Errors, downstream correction and the costs not in the supplier invoice. |
Qualitative summary of this article's source text, not a measured outcome or performance estimate.
The pricing unit is the actual problem
If you buy full-time equivalents, you have bought inputs. That means three things follow automatically: you pay for wage inflation, you pay for attrition, and you receive none of the benefit when the provider automates part of the process. That last point is worth sitting with. A provider who automates thirty per cent of the keystrokes on an input-priced contract has just improved its own margin and has no commercial reason to tell you. The same provider on a transaction-priced contract with a declining unit rate has to share the gain to keep the account. The negotiation is therefore not really about rates. It is about who owns the productivity curve. Moving the unit from a seat to a transaction, a case or an outcome — with a volume band, a service level and an agreed treatment for exceptions — changes the incentive rather than the price, and it is the only structural answer to wage inflation available to a buyer.
Three levers, in order of value
Reduce the work. The cheapest transaction is the one nobody performs. Before relocating or repricing anything, look at what generates volume: duplicate approvals, supplier invoices that arrive by email into a mailbox, customer queries caused by an unclear statement, exceptions created by a master data defect. Elimination beats relocation every time and is rarely on the agenda because it is nobody's contract. Re-segment the work. The low-cost seat is genuinely well suited to high-volume repetitive processing. It is poorly suited to judgement, negotiation, and anything requiring institutional memory. Most delivery arrangements have drifted: judgement work migrated offshore because it was adjacent to the transactions, and it is now being performed by the population with the highest turnover. Separate them explicitly, and price the judgement roles for retention rather than for arbitrage. Reprice the unit. Then, and only then, renegotiate. A buyer who has removed volume and re-segmented roles has a credible story; a buyer who simply asks for a discount gets a discount and a quieter form of degradation.
What to ask your provider this quarter
Three questions, and the answers are more informative than any benchmark: What was the actual salary movement, over the last three years, for the grades working on our account in the specific city they sit in? Company-wide averages conceal everything. What is the attrition rate on our account, not across the business, and what is the average tenure of the people doing our work today? What proportion of our volume is now touched by automation, and how is that reflected in the price? A provider who cannot or will not answer those has told you something about the relationship.
Practical Guidance for Delivery Cost Strategy Review
- Rebuild the business case with replacement cost, supervisory ratio, retained management and rework included. The honest number is lower than the original and still probably positive.
- Move at least one process from seat pricing to transaction pricing at the next renewal, with a volume band and a unit rate that declines annually.
- Ask for account-level attrition and city-level wage movement in writing. Make both reportable monthly, not annually.
- Put an attrition threshold in the contract with a remedy attached, and name continuity requirements for the handful of roles that carry institutional knowledge.
- Define the wage indexation clause precisely if you accept one: which index, which basket, which cap, and what the buyer gets in return.
- Separate judgement work from transaction work and price them differently. Paying a retention premium for twelve people is cheaper than re-learning your process twice a year.
- Spend the first two weeks of any review on volume elimination, not on rates. Expect to find between five and fifteen per cent of transactions that should not exist.
- Compare offshore against your own fully loaded in-house seat, not against a Western benchmark. In this region the comparator is much closer than the rate cards suggest.
The Regional Angle
That last bullet is the regional point, and it is the one most likely to change a decision here. When a European or American buyer compares an offshore seat with an in-house seat, the gap is enormous. When a Gulf-based group does the same arithmetic, the comparison is far tighter, because the in-house seat in this region is filled from substantially the same labour markets as the offshore centre. The difference is relocation, not skill. Against that, the fully loaded cost of employing somebody here is not only salary: it is visa and processing fees, medical insurance, accommodation or housing allowance, transport, an annual air ticket, end-of-service gratuity accruing every month, and the administrative overhead of maintaining all of it. Count those honestly and the in-house seat becomes more expensive than managers assume — but the offshore seat plus governance, connectivity, transition cost and provider margin also lands closer than the sales deck suggests. Groups that have run this calculation properly often find the two within twenty per cent of each other, at which point the decision is about control, continuity and scalability rather than about cost. The second regional factor is overlap, and it is an advantage buyers here consistently fail to price. A Gulf head office sits within about ninety minutes of Indian delivery hours and shares most of the working day with South Asia generally. Western buyers can offshore batch work — things that can be processed overnight and reviewed in the morning. Buyers in this region can offshore interactive work: live query handling, same-day exception resolution, a shared morning call that actually happens. That is worth real money and should be reflected in what you expect from the arrangement, not merely enjoyed. The corollary is the working week. Where the local week begins on Sunday and the delivery centre works Monday to Friday, Sunday coverage has to be bought deliberately rather than assumed, and Friday cover for the local team frequently gets provided by the offshore team without anybody pricing it. The third factor is wage transparency, and it is uncomfortable. Because the in-house team and the offshore team are often drawn from the same communities and stay in touch, a double-digit increase announced in Bengaluru or Manila is known in your Dubai or Riyadh finance department within days. Offshore wage inflation therefore creates retention pressure on your own staff, who can see the delta narrowing and can now be recruited remotely by companies in other countries without leaving home. Any delivery cost strategy that does not include a retention position for the local team is solving half the problem. One procurement note worth adding: where you bid for government or semi-government work, tender scoring increasingly rewards local employment and in-country spend. The cheapest delivery model can lose the bid, which makes offshoring a commercial decision with a revenue consequence rather than purely a cost decision.
The objection worth taking seriously
The strongest objection is arithmetical. A wage base at a fifth or a quarter of the alternative, growing at eight or ten per cent while the alternative grows at three, takes a great many years to converge. The offshore advantage has been declared over roughly every three years since the late nineties, and each time the industry has responded by moving to a smaller city, adding a new geography, or raising productivity, and the gap has persisted. Anyone forecasting the end of labour arbitrage on the basis of one year of post-pandemic hiring competition should be treated with scepticism. That is fair, and this is not an argument for bringing work home. The gap is not closing to zero in this decade. The narrower claim is the one that should reach the board. Your existing contract was priced on an assumption of improving unit economics, and the underlying inputs are moving the other way. The saving you report is drifting away from the saving you receive, the difference is concealed by the pricing unit, and the marginal decision — where to place the next fifty seats, which process to move next — is genuinely closer than it was three years ago. Those are the decisions in front of you this year, and they turn on a much smaller margin than the founding business case did.
Common Questions
Should we move to a lower-cost city or country?
Sometimes, but choose on wage trajectory and skill depth rather than today's rate. A cheaper city with a thin talent pool delivers a lower rate and higher attrition, which is the wrong trade.
Is building our own centre better than using a provider?
At scale and with a multi-year horizon, a captive captures the margin and the institutional memory; below roughly a couple of hundred seats it usually just relocates your management problem. Either way you inherit the attrition rather than outsourcing it.
How do we stop the provider taking all of the automation benefit?
Price per transaction with a declining unit rate, require reporting on the proportion of volume automated, and make the productivity commitment a unit cost decline rather than a headcount reduction.
What should we expect over the next twelve months?
Expect double-digit increases and elevated attrition to persist through the year, because demand for the same skills is being bid up simultaneously by global companies now hiring remotely into those cities. Expect providers to arrive at renewals asking for wage indexation clauses and to resist fixed unit pricing; treat that as the negotiation rather than as a formality. Expect more delivery capacity to be announced in tier-two cities and in newer geographies, including this region's nearer neighbours, and expect the first year of any such centre to run hot on quality. And expect the most valuable move available to you — removing transactions that should never have existed — to remain the one nobody's contract rewards.
Delivery Cost Strategy Review — we rebuild the offshore business case with the costs the original one omitted, test it against your fully loaded in-house seat, and identify what to eliminate before you renegotiate anything.