Most offshore contracts don’t fail because of the talent. They fail because of the commercial structure wrapped around that talent. The choice between time and material, fixed price and a dedicated team decides who absorbs risk, who keeps the knowledge, and how fast you can change direction. That is why offshore engagement models deserve more scrutiny than the vendor shortlist.
Gartner’s July 2026 forecast puts worldwide IT spending at $6.37 trillion this year, 14.2% higher than in 2025. Its April 2026 update had already named IT services the largest spending category, at more than $1.87 trillion.
A large share of that services spend runs through offshore software development contracts. The way those contracts are structured affects delivery more than most buyers admit. The gap shows up fastest in legacy work.
Mainframe estates still run core banking, insurance and logistics. The 2026 BMC Mainframe Survey covered more than 1,300 practitioners, and leaders said they plan to point AI investment at AIOps, application development and knowledge transfer over the next two years. When you combine scarce COBOL skills, undocumented business logic and AI-assisted refactoring, a poor choice among offshore engagement models gets expensive quickly.
This guide breaks down each model the way a delivery lead would: what it costs, where it breaks, and whether it fits modernization, maintenance or greenfield work. It draws on patterns the Supersourcing delivery team has seen across 527+ IT projects in fintech, healthtech and enterprise SaaS. You will find cost ranges, red flags and a six-step selection process.
TL;DR
This guide compares the three standard offshore engagement models: time and material, fixed price and dedicated team. It is written for CTOs, engineering heads and procurement leads, with a focus on legacy and mainframe programs.
The most important number is the risk buffer. On legacy code, fixed-price bids typically include a 15–30% contingency, and you pay it whether or not the risk happens. A dedicated development team, by contrast, can be interview-ready in 7–10 working days.
By the end, you will know which of the offshore engagement models to sign, which clauses to push back on, and when to switch models without stalling the program.
What Are Offshore Engagement Models?
Offshore engagement models are the commercial and operating structures that define how a company works with a remote technology partner, including how work is scoped, priced, staffed and governed. The three standard options are time and material, fixed price and dedicated team, and each one places cost risk and control in different hands.
In short, time and material buys hours, fixed price buys an outcome, and a dedicated team buys capacity. Every other clause in offshore engagement models follows from that one choice.
Legacy Modernization Staffing: Why the Model Choice Gets Expensive
Legacy programs break estimates in ways that greenfield work rarely does. Finding people who can write code is the easy part. The hard part is finding people who can read 30-year-old code, explain why it exists, and change it without breaking a batch job nobody has touched in a decade.
Discovery on a mainframe modernization program routinely turns up dead programs, duplicated copybooks, business rules hard-coded in JCL, and integrations that appear in no diagram. In most legacy engagements we’ve run, scope after discovery lands at 1.5–3x what the original RFP described. That gap is where offshore engagement models either protect you or punish you.
Three realities make this worse:
- Knowledge concentration. Critical batch and CICS logic often sits with two or three people, and some of them are a few years from retirement.
- Ramp-up time. Even strong engineers typically need 4–8 weeks to become productive on an unfamiliar mainframe codebase, and longer if documentation is thin.
- Moving targets. A maintenance contract often turns into a COBOL to Java migration, and then into API enablement, within 12–18 months.
Each of these realities favors a different commercial structure. Choosing between offshore engagement models is therefore a delivery decision first and a procurement decision second.
How the Three Offshore Engagement Models Actually Work
Every model answers one question: who carries the risk when reality drifts from the plan? Here is how each one behaves on live programs.
Time and Material: Flexibility You Have to Govern
Time and material is the default for discovery and evolving scope. Under a time and material contract, you pay for actual hours at agreed rates, and priorities can change every sprint. It is also the commercial model behind most IT staff augmentation.
The catch is governance. Without burn tracking at the sprint level, hours expand to fill the budget. Insist on named engineers, a rate card by seniority, and visibility into the bill rate vs pay rate gap. A markup well above 40–50% usually signals a layered subcontracting chain.
T&M also suits short specialist work. If you need two performance engineers for a six-week batch-tuning sprint, it is faster to hire contract developers on hourly terms than to re-scope a fixed bid.
Best for: discovery phases, unclear scope, R&D and short specialist needs.
Fixed Price: Certainty That Holds Only on Known Scope
Fixed price works when the deliverable can be written down in full. Under a fixed price contract, the vendor commits to a defined scope, timeline and cost, and absorbs any overrun.
You pay for that certainty up front. Vendors add contingency, typically 15–30% on legacy work, and every change request gets its own quote. On a mainframe program where discovery keeps finding new dependencies, the change log can quietly grow larger than the original contract.
On legacy work, fixed price breaks down less because of scope creep and more because of scope discovery. On known code it performs well. On unknown code, every surprise becomes a negotiation.
Best for: a defined report migration, a single interface rewrite, or test automation against a frozen spec.
Dedicated Team: Owned Capacity, Shared Context
With this model, you reserve full-time engineers who work only on your program, under your direction, for a monthly fee. When you hire dedicated development teams, knowledge stays inside one stable group instead of resetting with every new statement of work.
For legacy programs that run longer than six months, this is the model whose value builds over time. The team learns your copybooks, batch windows and release politics. By month four, it is fixing problems that a rotating T&M bench would still be discovering.
The risk shifts to you. You own prioritization and utilization, and if you can’t keep five engineers busy, a dedicated team becomes expensive idle capacity. Of the three offshore engagement models, this one depends most on strong internal product ownership.
Best for: long-running modernization, ongoing maintenance, and programs where retained context is the main asset.
Time and Material vs Fixed Price for Software Projects: The Real Cost Math
Hourly rates mislead buyers more than any other number. Typical bill rates for mainframe developers in India run $25–45 per hour, depending on seniority and CICS/DB2 depth. Comparable US onshore contractors typically bill $90–150. Treat these as working ranges and check them against live quotes.
The honest comparison is total program cost. Take a five-person team at a $35 blended rate, working about 160 hours a month each. That costs roughly $28,000 a month, or $336,000 a year on T&M or a dedicated model. If the same 12 months is priced as a fixed bid with 20% contingency, you start near $403,000, before any change request.
Rates also move with scarcity, so the cost to hire offshore developers in India depends more on the skill than on the city. Across offshore engagement models, the cheapest hourly rate rarely produces the cheapest program.
How to Choose an Offshore Engagement Model in 6 Steps
Run this sequence before you sign. It applies to all offshore engagement models and works for maintenance, migration and greenfield programs.
- Score scope stability. If more than 30% of requirements are still open, rule out fixed prices for that phase.
- Estimate duration. Programs under three months favor T&M or fixed price. Beyond six months, a dedicated team usually wins on total cost.
- Map knowledge risk. If critical logic lives with a few SMEs, choose the model that keeps the same engineers longest. Build a structured staff augmentation onboarding process around the SME handover.
- Decide who owns prioritization. Strong internal product owners suit T&M or dedicated teams. If internal capacity is thin, use fixed prices on small, bounded work.
- Price the change path. Compare change-request rates, not just day rates. Ask for three worked examples of how a scope change in the middle of a sprint gets billed.
- Write the switch clause. Agree up front how a fixed phase converts to T&M or a dedicated team, and what happens to named engineers and IP when it does.
How to Hire COBOL Developers in India Without Stalling the Program
The shortage is real but uneven. When enterprises set out to hire cobol developers, the gap is rarely COBOL syntax. It is engineers who combine COBOL with CICS, DB2 and JCL, and who understand how batch interacts with modern APIs.
Screen for that combination. Ask candidates to walk through a real copybook, trace a failed batch job, and explain how they would expose a CICS transaction as a REST service. Messaging matters too. Many programs need to hire IBM MQ engineers alongside COBOL talent, because MQ is where mainframe and cloud meet.
If the roadmap includes migration, put legacy specialists and target-stack engineers in one squad. Teams that hire Java developers with real conversion experience avoid the classic failure where translated code compiles but nobody can maintain it. The same rule applies when you hire offshore developers for both stacks: one squad, one backlog, one owner.
Speed matters when a retiring SME has a fixed exit date. Supersourcing’s AI-powered sourcing surfaces the top 2% of vetted talent and typically delivers an interview-ready shortlist in 7–10 working days.
What This Looks Like in Real Engagements
The two patterns below are anonymized composites that show how offshore engagement models play out on legacy programs. Client details and confidential figures have been removed.
Pattern 1: From fixed bid to dedicated squad. A mid-size insurer started a policy-administration modernization as a fixed bid. Discovery found undocumented batch dependencies, and the change log grew every month. Converting the remaining phases to a dedicated squad of COBOL and Java engineers moved scope changes inside a predictable monthly fee and made SME handover a planned workstream.
Pattern 2: T&M discovery feeding a dedicated pod. A payments company ran a six-week T&M discovery on its settlement batch, then kept the three strongest engineers as a dedicated pod. Because the people stayed the same, there was no second onboarding cycle. The pod started delivery with the context that discovery had already built.
Decision Framework: Comparing Offshore Engagement Models Side by Side
| Factor | Time & Material | Fixed Price | Dedicated Team |
| Scope stability needed | Low | Very high | Medium |
| Typical duration | Weeks to 6 months | 1–6 months, bounded | 6+ months |
| Who carries cost risk | Client | Vendor (priced in as contingency) | Client (utilization) |
| Knowledge retention | Low to medium | Low | High |
| Best legacy use case | Discovery, specialist work | Bounded migrations against a frozen spec | Long-running modernization and maintenance |
Most enterprises end up combining offshore engagement models: fixed price for bounded deliverables, a dedicated core team for the long term, and T&M for spikes in demand. If you are still deciding whether to partner at all, outsourcing vs dedicated team vs RPO covers that earlier decision.
What Most Teams Get Wrong About Offshore Engagement Models
The most expensive mistake is choosing a model for the first three months instead of the next eighteen. Teams sign a fixed price to win budget approval, then pay for every discovery surprise as a change request. In legacy code, scope certainty is usually an illusion. Price the program you will actually run, not the one written in the RFP.
A second pattern shows up on almost every mainframe program. Access provisioning, including RACF IDs, TSO access and test LPAR slots, typically takes 2–4 weeks. Teams that don’t raise access requests before signing spend the first month of a dedicated team’s fee on idle time.
Third, buyers treat knowledge transfer as a line item instead of a deliverable. Make runbooks, recorded walkthroughs and annotated copybooks dated acceptance criteria, or they won’t get done. Lock in named engineers with a replacement SLA as well, because a bench that rotates after month two quietly resets context.
Finally, many buyers worry about vendor lock-in with dedicated models but ignore it with fixed price, where the vendor often holds the only working knowledge of what was built. The familiar challenges in offshore hiring, such as attrition, timezone delays and uneven vetting, get worse when offshore engagement models conflict with how delivery actually works.
Pressure-Test Your Model Before You Sign
The model choice is easier to get right before signatures than after. If you are comparing offshore engagement models for a mainframe or legacy program and want a second opinion before committing, Supersourcing has made this decision across 527+ IT projects. Its record includes a 98% candidate joining rate and a 7–10 day replacement guarantee.
Share your scope, timeline and current stack, and we’ll map it to a model, a team shape and a realistic plan for the first 90 days. Email mayank@engineerbabu.com or book a consultation.
Frequently Asked Questions
What are the three main offshore engagement models?
The three main models are time and material, fixed price and dedicated team. Time and material bills for hours worked and suit evolving scope. Fixed price commits to a defined deliverable at an agreed cost, with contingency built in. A dedicated team reserves full-time engineers for a monthly fee and suits long programs where retained context drives value.
Which offshore engagement model is best for legacy modernization?
For programs longer than six months, a dedicated team is usually the strongest fit, because mainframe context builds over time. Time and material works well for discovery, when scope is still unknown. Fixed price fits only bounded components, such as a single interface rewrite with a frozen spec. Most teams combine offshore engagement models across phases rather than picking just one.
How much does a dedicated offshore team cost?
Cost depends on team size, seniority and how scarce the skills are. As a working example, five engineers at a $35 blended hourly rate cost roughly $28,000 a month. Scarce skills such as CICS, IMS or Assembler push rates higher. Compare total 12-month cost, including onboarding and access lead time, rather than headline hourly rates.
Can you switch offshore engagement models mid-project?
Yes, and on legacy programs you often should. A common path is T&M discovery, then a dedicated team for delivery, with fixed price reserved for bounded components. Switching goes smoothly only if the original contract defines conversion terms, named-engineer continuity and IP ownership. Without a switch clause, conversion becomes a renegotiation.
Are COBOL skills really disappearing?
Not uniformly. Core COBOL talent is still available, especially in India. Engineers who pair COBOL with CICS, DB2, MQ and modern API work are scarce. The tightest gap is people who can bridge legacy and target stacks during migration, which is why hiring for skill combinations works better than hiring for one language.
What should every offshore contract include?
Whatever the model, insist on named engineers, a replacement SLA, IP assignment, NDA coverage, a rate card by seniority, rules for pricing change requests, and a conversion clause. These terms decide whether a model can adapt when scope shifts, which on legacy work it almost always does.
How quickly can an offshore mainframe team start?
With a vetted sourcing pipeline, shortlisting can take 7–10 working days. The real start date usually depends on access provisioning and SME availability. If you face a modernization deadline or a retiring expert with a fixed exit date, a short scoping call can map the team shape and a first-90-day plan before contracts are signed.
