The economics of financial services operations are undergoing a fundamental shift. Providers are being measured increasingly not by the headcount they deploy, but by the business outcomes they deliver — reduced credit losses, improved collections, and lower fraud exposure. This article examines the move from effort-based to outcome-based delivery, and the transformation and commercial models that must evolve alongside it.
From labor arbitrage to measurable outcomes
For twenty years or so, BPM service providers sold the same thing: “Give us your processes, we'll run them cheaper, and you'll need fewer people”. That deal built an entire industry. And it's quietly changing. A report from ISG last year found that leading service providers are now tying somewhere between 70 and 100 percent of contract value not to headcount, but to business outcomes, resulting in fewer bad loans, more cash collected, lower fraud losses, etc. You don't get those outcomes by automating a broken process faster. You get them by transforming how the whole operation works. The question is about what change did actually happen.
Business outcomes create greater value
Removing five underwriters is a cost story. Reducing the default rate on the loan that team approves is a business outcome story. One saves money once; the other compounds every single quarter. And crucially, that outcome doesn't come from technology alone. It comes from redesigning the credit decision, fixing the data feeding it, reskilling the people making the judgment calls, and rewiring the governance around it. Technology is probably a third of the answer. The market has moved with this too — the BPO industry is now worth over 400 billion dollars, and analysts are explicit that buyers are reallocating spend away from pure labor arbitrage toward value creation.
Let’s see where the distinction bites the hardest across financial services. Pick almost any function and the reframe is obvious once you see it. In mortgage underwriting, the old pitch was “process applications faster with fewer people”, but the outcome that matters is loan quality. You get there by redesigning the whole origination journey. This is done with better data at the point of decision, sharper credit policy, and automation wherever it helps. In collections, the outcome is cash — more dollars recovered, sooner — and that comes as much from rethinking customer segmentation, contact strategy, and agent coaching as from using any tool. In financial crime and fraud, it's fewer false positives while catching more genuine risk, which is a process and operating-model change, not just a system. And on the front end — wealth management, customer service — it's redirecting people toward growing assets under management. Transformation could free up quite a bit of an adviser's time. That's not cost saving, but the capacity you redeploy toward revenue and relationships.
The shift vs. traditional provider economics
For service providers, this might sound detrimental for their business model, since for years they have sold FTE reduction as the primary USP. Their entire commercial DNA was built on labor arbitrage — they priced effort, seats, transactions. Moving to outcomes means paying the providers not for how hard they work but for the improvements they deliver to clients. That exposes them and demands that they bring more than delivery muscle. Real transformation needs consulting depth, domain expertise, data and technology, and change management, working as one. If service providers can't move the needle, they don't get paid. It also breaks internal incentives. Teams have been rewarded on utilization and headcount for so many years. So, the hardest part isn't technology. It's that outcome-based, holistic transformation attacks the very economics that made providers successful in the first place.
A few practical things hold providers back, with a clear conflict of interest baked in. First, attribution — outcomes like default rates or AUM are influenced by the economy and a hundred factors you don't control, so proving that you moved the needle is hard. This measurement challenge is the biggest reason why only a minority of vendors have implemented true outcome-based pricing. Second, data access: To stand behind an outcome, a provider needs visibility into the client's actual results, and many won't share collections or loss data with a vendor. Third, risk appetite: Outcome models mean taking on downside, and most provider balance sheets weren't built for that. And fourth, capability breadth: You can't promise a better default rate if all you've ever done is process transactions. You need to bring process redesign, data, technology, and talent transformation together. That's exactly why providers are buying domain and agentic AI capabilities precisely so they can stand behind outcomes across the whole stack, not just running the back office.
The approach and outcomes
Given all that, you don't start with a grand outcome-based contract, and you don't start by ripping everything out. Both are mistakes. You start narrow and instrumented. Pick one process where the outcome is measurable and reasonably attributable — collections or fraud where losses prevented are trackable. Then diagnose it end to end before you touch anything. Find out where the value is leaking; is it process, data, skills, or technology? Baseline it properly, because without a clean pre-transformation picture you can never prove impact. Then transform that slice holistically — redesign the process, fix the data, reskill the people, apply the right tech — and prove you moved one specific lever. Get a track record on that narrow claim first. Trust in outcomes is earned on small, provable wins, not promised in a slide.
Measurability of outcomes is a fair challenge, and this is exactly where the rigor lives, because a metric you can't define you can't get paid on.
- In mortgage and lending, the outcome metrics are early-payment-default rate, approval-to-funding cycle time, and cost per funded loan.
- In commercial lending, it's covenant-breach detection lead time, exposure-at-default avoided, and time-to-decision on credit renewals.
- In collections, it's dollars recovered per account, recovery rate, right-party-contact rate, promise-to-pay conversion, and roll-rate into later delinquency buckets.
- In financial crime, it's false-positive reduction, alert-clearance time, cost per alert, and SAR conversion quality.
- In fraud, it's fraud losses prevented in dollars, false-decline rate, and detection-to-block latency.
- In payments, it's straight-through-processing rate, settlement-fail rate, break-resolution time, and operational-loss provisions.
- And in wealth and capital markets, it's adviser capacity redeployed, net new assets, trade-fail rate, and confirmation-matching accuracy.
The discipline is picking two or three per contract, baselining each, and agreeing the measurement window up front — not twenty vanity metrics which can be difficult to attribute.
Let’s see what the way forward is once you have the early proof points. You climb a maturity curve on two axes at once — commercial and transformational.
On the commercial side, at the first stage, you're still on input pricing but reporting outcomes alongside it. At stage two, there are outcome-linked bonuses and penalties on a base fee. At stage three, for processes where you've proven attribution, obtain a genuine gain-share where a slice of your revenue is the value created. And in parallel, the transformation deepens — from fixing individual processes, through redesigning the end-to-end operating model, to reimagining the business capability itself. Different processes can be at different stages at the same time. Nobody flips the whole book overnight, and nobody transforms everything at once.
Pricing the outcomes in the right way
Let’s examine how to price an outcome, because if the price is wrong, you either give value away or you price yourself out. This is the discipline that separates the serious players. Pricing starts with quantifying the value pool. Find out what is a one-percent lift in collections recovery, or a ten-basis-point cut in default rate, actually worth to this client in hard currency. After estimating the price agree the baseline and the attribution method jointly — a shared, transparent number, not something that you assert. Then, negotiate the share of the value created — and there's a real prize for getting this right: Firms using value-based pricing achieve higher profit margins than peers on traditional models. And critically, you build between a floor and a cap — a floor so you cover your cost-to-serve and the investment you’ve put into the transformation if outcomes lag for reasons outside your control, and a cap so the client isn't horrified when you massively overdeliver. Because in this shift, you are not just running a process anymore, you’re investing capital and capability up front to transform it. The pricing has to reflect that you’ve got skin in the game. Get that structure right and pricing stops being a fight over rate cards and becomes a shared bet on results.
