After the EOR

Executive Summary

The Employer of Record market is projected to reach $10 billion by 2028. The model works: companies can hire in new countries within weeks, without setting up a legal entity, and stay compliant with local employment law. That problem is solved.

But a different problem has taken its place. We keep seeing the same pattern across our deployment work: a company engages an EOR, hires a team in India or Eastern Europe, gets the payroll running and the contracts signed, and then realizes that nobody designed the team. Nobody validated whether the people hired can actually do the work to the standard required. Nobody established an operating model, quality protocols, or governance framework. The employment wrapper is in place. The capability layer is missing.

This paper examines that gap. It is written for companies that have used or are considering EOR for global team deployment, and who need to understand what the EOR does, what it does not do, and what needs to sit on top of it for the deployment to actually deliver results.

We use EOR ourselves. This is not an argument against the model. It is an argument that the model is incomplete without a structured capability and governance layer above it.

1. What the EOR solved

Before EOR became mainstream, a company wanting to hire five engineers in India had two options: set up a legal entity (six to twelve months, $200K–$500K, and a stack of regulatory filings) or classify workers as independent contractors and hope nobody noticed. The first was slow and expensive. The second was illegal in most cases.

EOR fixed this. The model is straightforward. A third-party organization becomes the legal employer of your workers in a country where you have no entity. The EOR handles payroll, tax withholding, statutory benefits, employment contracts, and local labour law compliance. You manage the workers’ day-to-day tasks. The EOR manages the legal relationship with the state.

It is a good model. For companies testing a new market, hiring a small team before committing to an entity, or deploying workers in multiple countries simultaneously, EOR removes the single largest barrier to global hiring: the entity requirement. The market has responded accordingly. Deel, Oyster, Papaya Global, Remote, and dozens of other providers now offer EOR services in 100+ countries.

1.1 The problem that remains

Here is what EOR providers do not say in their marketing: employment compliance is a necessary condition for global deployment, but it is not a sufficient one.

An EOR gets your people employed. It does not get them organized into a functioning team. It does not validate whether their skills match your delivery requirements. It does not establish how they will work together, what quality standards they will meet, or how their performance will be measured against delivery outcomes rather than attendance.

Think of it this way. If you hire a property management company to handle your rental agreements, that company ensures the lease is legal, the rent is collected, and the maintenance schedule is followed. What it does not do is design the building.

EOR is property management. The building still needs an architect.

2. Five gaps the EOR does not close

Over the past three years, we have deployed teams for CROs, system integrators, enterprise data programs, and companies scaling globally through EOR-backed arrangements. The gaps we see are consistent across industries, geographies, and team sizes.

2.1 Team design

EOR providers hire individuals. They do not design teams. The distinction matters.

When a company tells its EOR to hire six data engineers in Hyderabad, the EOR sources and employs six people with “data engineer” on their CV. What it does not do is assess whether those six people, together, cover the full spectrum of capability the program requires. Are all six pipeline builders, or does the team also need a data architect, a DataOps engineer, and someone who understands governance? Nobody asked, because team composition is not the EOR’s job.

The result is a team that looks staffed on paper but has structural holes in practice. We wrote about this at length in our earlier paper on the enterprise data talent gap: most organizations hire for three of the six roles their data platform actually needs. EOR makes this problem worse, because the hiring happens faster and with less scrutiny than internal recruiting would apply.

2.2 Competency validation

EOR providers verify that a candidate is legally eligible to work, that their documents are in order, and that they accept the employment terms. They do not typically assess whether the candidate can do the specific work the client needs done.
In regulated environments, this gap can be dangerous. A medical writer hired through EOR to produce clinical narratives for a pharmaceutical sponsor needs validated competency in ICH-GCP guidelines, therapeutic area knowledge, and the sponsor’s documentation standards. The EOR confirms the employment contract is compliant with Indian labour law. Whether the writer can produce a compliant clinical study report is a separate question that nobody in the EOR arrangement is responsible for answering.
In technology roles, the gap is less dramatic but equally real. A Kafka engineer hired for a banking platform needs to understand the client’s architecture, security requirements, and deployment standards. The EOR has no visibility into any of this.

2.3 Operating model

A team without an operating model is a group of individuals sharing a Slack channel. They need defined ways of working: sprint cadences, stand-up rhythms, code review protocols, documentation standards, escalation paths, and handoff procedures. Someone has to design this, train the team on it, and enforce it.

EOR providers do not do this. Their scope ends at employment administration. The operating model is the client’s responsibility. But many clients, especially those deploying their first offshore team, assume the EOR covers it. They discover otherwise around week four, when the team is employed and paid but not producing anything coherent.

2.4 Quality governance

In our world, quality governance means defined acceptance criteria for deliverables, review protocols, quality checkpoints, and audit-ready documentation. In pharmaceutical environments, this is governed by ICH-GCP, FDA 21 CFR Part 11, and company-specific SOPs. In enterprise technology, the standards are less formalized but no less real: code quality gates, testing coverage requirements, deployment approval workflows.

EOR does not touch any of this. The gap is most visible when something goes wrong. A deliverable fails quality review. A data pipeline introduces errors into production. A clinical narrative does not meet the sponsor’s documentation standard. In each case, the EOR’s obligation was met. The employment contract is valid. The payroll was processed. The failure occurred in a layer that the EOR does not operate in.

2.5 Retention and development

People hired through EOR can feel disconnected. Research from multiple EOR providers themselves acknowledges this: EOR-employed workers sometimes feel like “second-class citizens” compared to directly employed staff. They sit outside the client’s internal career framework, outside its learning and development programs, and outside its performance management system.

For a two-person team hired for a six-month project, this may not matter. For a fifteen-person team running a multi-year platform build, it creates attrition risk that compounds over time. People leave. Knowledge walks out with them. Replacement cycles restart. The cost of attrition in a structured delivery program is rarely just the recruitment fee; it is the four to eight weeks of ramp-up time that each replacement requires.

3. The capability layer

The solution is not to abandon EOR. The employment infrastructure it provides is genuine and difficult to replicate. The solution is to build a capability layer on top of it.

We call this the EOR + Capability Stack. The EOR handles what it does well: legal employment, payroll, tax, statutory benefits, and labour law compliance. The capability partner handles everything above that: team design, competency validation, operating model, quality governance, and ongoing people development.

The table makes the split visible. EOR covers the left column. Without a capability partner, the right column is empty. Someone has to fill it. In most failed global deployments we have seen, nobody did.

4. What this looks like in practice

Three patterns from our deployment work illustrate how the EOR + Capability Stack operates.

A CRO building a biometrics team in India

A mid-size CRO needed seven people: two biostatisticians, three SAS programmers, a clinical data manager, and a medical writer. They had no Indian entity. Their previous attempt, through a standalone EOR, had produced a team that was employed but could not meet the sponsor’s quality standards. The SAS programmers had programming skill but no experience with CDISC SDTM standards. The medical writer had written marketing content, not clinical narratives.

We redesigned the team against the sponsor’s actual requirements, assessed candidates for domain-specific competency (not just employment eligibility), deployed a governance framework including SOPs, quality review protocols, and training validation, and had the team delivery-ready in 18 days. The EOR handled the employment contracts and payroll. We handled everything else.

An SI augmenting a cloud migration program

A system integrator needed twelve cloud engineers for a client’s Azure migration. Their staffing partner, operating through EOR, filled the roles individually over six weeks. By week eight, the SI had twelve employed cloud engineers who had never worked together, had different naming conventions, different deployment approaches, and no shared operating rhythm. The client escalated.

We replaced the individual-hire approach with a team deployment. Same EOR infrastructure for employment. But the twelve people were assessed as a team, deployed with a defined operating model, given shared tooling standards and code review protocols, and started with a governance framework that matched the SI’s delivery methodology. Delivery stabilized within two sprints.

A fintech company scaling a data platform from London

A UK-based fintech wanted to build a ten-person data team in India to extend its Databricks platform. The founder had used EOR before for individual hires. This time, the scale required a different approach.

We designed the team: two data engineers, a data architect, an analytics engineer, a DataOps specialist, and a data governance lead, plus four additional pipeline developers. The EOR employed them. We validated their competencies against the fintech’s platform architecture, established an operating model that synced with the London team’s sprint cadence, and deployed a data quality monitoring framework from day one. The team shipped its first production pipeline in week three.

5. When EOR alone is enough

6. How to evaluate your deployment model

7. The EOR market is growing. The capability gap is growing faster.

The EOR market is expanding at roughly 7% annually. More companies are hiring through EOR. More countries are being covered. The technology platforms are getting better. The compliance infrastructure is maturing.

But the capability gap is growing faster than the EOR market can close it, because the EOR market was never designed to close it. EOR providers are employment infrastructure companies. They are specialists in payroll, tax, and labour law. Asking them to also design teams, validate competencies, and govern delivery quality is like asking your accountant to run your factory. They are excellent at what they do. What you need is a different kind of partner for the rest.

The companies that will succeed with global deployment in the next three to five years are those that treat employment and capability as two separate problems, each requiring a specialist. EOR for the employment wrapper. A structured capability partner for everything above it.

That is the stack that works. Employment compliance on the bottom. Capability and governance on top. Neither layer is optional, and neither layer can do the other’s job.

8. About SIRO Functional Services

SIRO Functional Services deploys structured capability teams across technology, data, platforms, and regulated environments for complex enterprises. We have operated in highly regulated industries for over two decades, where governance is the operating system, not a slide in the onboarding deck.

We work with EOR providers. We are not one. Our job starts where theirs ends: designing teams, validating competencies, establishing operating models, deploying governance frameworks, and managing delivery quality. Our clients are system integrators, CROs, pharmaceutical sponsors, enterprise data leaders, and companies scaling globally.

If you are deploying a team through EOR and wondering who fills the capability layer, that is the conversation we are built for.

For more information, visit www.sirofsp.com or contact bd@siroclinpharm.com.

References

1. Business Research Insights. “Global Employer of Record Platform Market.” BRI Market Report, 2025.
2. Exactitude Consultancy. “Global Employer of Record Market Size, Share, and Forecast to 2034.” Exactitude Consultancy, 2024.
3. Atlas. “The Global Atlas: for HR Leaders in 2025 — Challenges, Strategies, and Solutions.” Atlas HXM, 2025.
4. Everest Group. “Employer of Record Solutions PEAK Matrix Assessment 2025.” Everest Group, 2025.
5. Zinnov. “Employer of Record: Accelerating Global Expansion.” Zinnov Management Consulting, 2024.
6. NASSCOM. “Technology Sector in India: Strategic Review 2025.” National Association of Software and Service Companies, 2025.
7. OECD. “2025 Update to the Model Tax Convention: Permanent Establishment Commentary.” OECD, 2025.
8. ICH. “ICH E6(R2) Guideline for Good Clinical Practice.” International Council for Harmonisation, 2016.
9. FDA. “21 CFR Part 11: Electronic Records; Electronic Signatures.” U.S. Food and Drug Administration.
10. India. “Digital Personal Data Protection Act, 2023.” Ministry of Electronics and Information Technology, Government of India.