EOR Implementation: Your 90-Day Roadmap for Hiring in the US

What would break first if you had to put a US employee on legal payroll 30 days from now, the contract, the state tax registrations, or the benefits enrollment?

For most companies the honest answer is all three at once, which is why EOR implementation fails on sequencing rather than on strategy. An employer of record can carry the legal employment, payroll and filings for you, but only if the first 90 days are run as a project with named owners and hard dates. This guide is that project plan, written specifically for hiring in the United States.

What exactly is EOR implementation, and why do the first 90 days decide the outcome?

EOR implementation is the structured process of moving your US hiring onto a third party that becomes the legal employer while you keep day-to-day direction of the work. The first 90 days decide the outcome because almost every downstream cost, from payroll accuracy to tax registrations to termination exposure, is locked in by decisions you make before a single employee starts.

The mechanics are simple enough on paper. The provider signs the employment agreement, runs payroll, withholds and deposits federal and state taxes, sponsors the benefit plans and holds the formal employment process, while you set priorities and manage performance.

Where it gets complicated is the American detail underneath: 50 state payroll regimes, a classification test that has been in flux since 2025, and a compliance calendar that starts on day one rather than day 30 (read: how an EOR actually works).

Atul Gawande's argument about high-stakes rollouts in The Checklist Manifesto applies almost word for word: "Good checklists, on the other hand are precise. They are efficient, to the point, and easy to use even in the most difficult situations."

That is exactly what the next three sections are: a precise checklist, split into three 30-day blocks. The first block is where you decide who you are hiring, and who you are hiring them through.

What should you lock down in days 1 to 30 of your EOR implementation?

Days 1 to 30 are for decisions, not configuration: who owns the project internally, exactly which US roles and states you are hiring into, and which provider you are signing. Nothing technical should be built before those three are settled, because every one of them changes the build.

Who needs to be at the table before you sign anything?

Four functions, and no more: HR owns job design and onboarding, finance owns the cost model and the funding schedule, legal owns the contract and IP terms, and one executive sponsor owns the go-live date. Anything wider than that turns a 30-day decision into a 60-day committee.

Give each function a named decision rather than general involvement. The sponsor's real job is to settle the argument that shows up in every kickoff: whether this is genuinely an employee role, or a scoped project that should have stayed a contractor engagement (see: contractors versus employees). Getting that wrong is the single most expensive mistake available to you in the US, and it is cheapest to fix in week one.

Once ownership is clear, the next job is assembling the US-specific facts your provider will ask for on day one.

What US-specific information does your EOR need upfront?

Your provider needs the work state, the annualized pay, the exempt or non-exempt classification, the start date and the benefits tier for every planned hire before it can quote or register anything. Withhold any one of those and the registrations stall.

In practice, the intake sheet that unblocks the whole implementation looks like this:

• Work state and city for each hire. This drives state income tax withholding, unemployment insurance registration and local ordinances, and it decides whether a second state payroll account is needed at all (read: state tax reciprocity).
• Exempt or non-exempt status under the FLSA, because that one field determines overtime liability for the life of the role, and reclassifying later is retroactive.
• Annualized base, bonus and equity plan, so finance models true cost rather than salary. Our employee cost calculator does that in one pass.
• Benefits tier and effective date. Medical, dental, vision, retirement and PTO accrual all key off the start date, so a start date that slips by a week can move a coverage month (see: employee benefits packages).
• Pay frequency, because moving a US employee from semi-monthly to biweekly after go-live is a change you have to communicate, re-document and reconcile (read: pay cycles and pay periods).
• Work authorization status and any sponsorship expectation, flagged before an offer goes out rather than during onboarding (see: work authorization).

Hand that sheet over in week one and your provider can run registrations in parallel with your own vendor decision, which is the next thing to get right.

How do you pick a provider that will not break in month six?

Pick on entity ownership, US payroll depth and named human support, not on the demo. The providers that fail in month six are usually the ones routing your employment through a partner they do not control.

John Lee put the structural risk plainly in his LinkedIn analysis Employers of Record: The Case For And Against: "Some Employers of Record operate through third-party partners (i.e. they don't actually own the legal entity in the country they're operating in) and this can provide challenges from a service consistency perspective, and potentially some risks."

Mary Joseph's screening advice is the practical version of the same point: "Make sure the company follows local labor and tax legislations before entrusting your business to a vendor." Ask for evidence, not assurances, and ask during diligence rather than after signature (vendor screening tips).

The risk is not theoretical. A long-running Hacker News thread titled "Ask HN: Negative Experiences with Remote and Other Employer of Records (EOR)?" collects first-hand accounts from engineering and operations leaders whose service quality dropped sharply after the contract was signed. Read it before you shortlist, and use it to build your reference questions.

Score every provider against a short written list: do they own the entity, do they run US payroll in-house, who is your named contact, what is the payroll error remedy, and what happens to your employees if you leave (read: EOR vendor selection).

If you are still building the shortlist, a side-by-side of the best EOR companies is a faster starting point than a round of discovery calls.

With a signed agreement in hand, days 31 to 60 shift from choosing to building.

What happens in days 31 to 60 when you wire the EOR into your stack?

Days 31 to 60 are integration and paperwork: connecting the provider's platform to your HR and finance systems, finalising the employment agreement and IP terms, and training managers on the boundary between your instructions and the provider's employer duties. Nothing here is glamorous, and all of it is load-bearing.

How do you connect the EOR platform to your HR and finance systems?

Map the four data flows that actually matter, then agree which system is the source of truth for each: new-hire records in, payroll cost out to the general ledger, time off in both directions, and invoices into accounts payable.

Decide the direction of each flow in writing before anyone builds a connector, because the failure you want to avoid is two systems both believing they own headcount (read: EOR technology integration). If a manual CSV handoff is honest about being manual, it beats a half-built integration that silently drops records.

If you do not yet have a system of record on your side, sort that out now rather than after go-live, because retrofitting employee data later is far more painful than importing it once (see: HRIS versus HRMS).

Data plumbing is the easy half of this phase. The paperwork is where the US-specific risk actually lives.

What has to be in the employment paperwork before a US hire starts?

At minimum: an offer letter with correct at-will language, the provider's employment agreement, a confidentiality and invention-assignment agreement that routes IP to you, any state-required notices, and Form I-9 with Section 1 completed on or before day one.

Employment in every US state is at-will with one exception. Montana's Wrongful Discharge from Employment Act requires good cause for dismissal once an employee is past the probationary period, which defaults to six months. If you are hiring into Montana, that changes your termination language and your probation design, and it is far cheaper to catch in the template than in a dispute.

IP needs particular attention under an EOR model. Because work made for hire vests in the employer, and the provider is the employer, ownership has to travel two hops: employee to provider, then provider to you. Both hops need to be explicit in the master services agreement and in the employee's assignment clause. One caveat worth knowing: California Labor Code section 2870 voids any assignment of inventions an employee developed entirely on their own time without company resources.

Review the whole package against your own templates rather than accepting the provider's standard forms unread, paying particular attention to notice periods, indemnities and the payroll-error remedy (read: EOR contract management).

If your legal team has not worked through this structure before, a primer on how employment contracts are put together will make that review a lot faster.

Paperwork signed, the last piece of this phase is making sure your own managers do not accidentally undo it.

How do you train managers on what they can and cannot do?

Train them on a single rule: they direct the work, and the provider decides employment terms. Pay changes, formal discipline, leave of record and termination all route through the EOR, in writing, before anything is said to the employee.

The reason for the rule is co-employment exposure: the more your managers behave like the legal employer, the weaker the separation you paid for becomes. A one-hour session with real examples of what to say and what to escalate is genuinely enough.

Reviews, goals and feedback stay entirely yours, and should. Just agree upfront how a performance conversation becomes a documented process if it ever needs to (see: performance management under an EOR).

With systems connected, contracts signed and managers briefed, you are ready to run the thing for real.

What does go-live look like in days 61 to 90?

Go-live is three things in sequence: a pilot payroll on real data, the first onboarding cohort, then a 30-day measurement window against the KPIs you set in phase one. Doing them out of order is how implementations end up being repeated.

What must happen in a new US hire's first three business days?

Form I-9 Section 2 has to be completed by the employer within three business days of the employee's first day of work, with Section 1 done by the employee on or before day one. Records are retained for three years after the hire date or one year after termination, whichever is later (USCIS Form I-9 guidance).

E-Verify is voluntary for most employers, but mandatory for federal contractors carrying the FAR E-Verify clause and in a number of states. Contractors have to enroll within 30 days of award and verify new hires within 90 days (E-Verify), so confirm which category you fall into before your first start date rather than after.

New-hire reporting to the state directory is a federal requirement due within 20 days of hire, and it is the one people forget because it feels administrative (federal new-hire reporting rules). Your provider should own the filing; you should be able to see proof of it.

Everything else in week one is experience rather than compliance, and it is still worth designing: equipment before day one, accounts provisioned the night before, a first-week plan and a named buddy (read: EOR onboarding best practices).

Documents done, the next test is the money.

How do you run the first payroll without a surprise?

Run a pilot cycle on real data before the live one: fund the account early, check gross-to-net on every single hire, and reconcile employer taxes line by line against the model finance built in phase one. A pilot costs you one week and catches almost everything.

The errors that show up in a pilot are almost always deduction-side: a benefits election that did not carry over, a retirement deferral applied to the wrong earnings code, a state withholding set to the residence rather than the work location (see: how payroll deductions work).

Ask your provider to walk you through funding timing in concrete terms too: when the invoice lands, how long the transfer takes to clear, and how many days before payday the money has to be with them (read: the payroll process step by step). A late transfer is the most common cause of a late paycheque, and it is entirely on your side of the line.

Once payroll clears cleanly twice, you can start judging the implementation on numbers rather than on feel.

Which numbers tell you the implementation actually worked?

Six, and they should all be visible on one page: time to onboard, payroll accuracy, compliance deadline hit rate, cost variance against model, manager escalations, and 90-day retention. Anything else is reporting for its own sake.

These are the targets we hold ourselves to on a first US cohort, and they are a fair benchmark to hold any provider to:

If all six read green at day 90, your implementation is finished and the thing worth auditing next is your cost model.

What does a US employee actually cost through an EOR in 2026?

Salary is roughly 70% of the story. On top of gross pay you carry the employer share of FICA, federal and state unemployment tax, benefits, workers' compensation and the provider's own fee, and most of the statutory pieces have hard 2026 numbers you can plan against.

Two of those lines deserve a second look. The Social Security wage base rose to $184,500 for 2026, up from $176,100 the year before, which materially changes the employer cost of any hire paid near or above that line.

The benefits share is not a rule of thumb either. The Bureau of Labor Statistics Employer Costs for Employee Compensation release for March 2026 puts benefits at 30.1% of total employer compensation cost in private industry: $14.01 per hour of benefits against $32.60 per hour of wages, for $46.60 all in.

FUTA looks alarming at 6.0% until you apply the state credit, which normally brings it down to 0.6% on the first $7,000 of each employee's wages (IRS Topic 759). States in credit reduction are the exception, so ask which of your work states are affected.

If you want the full mechanics of what gets withheld, deposited and reported on each cycle, our guide to employer payroll taxes covers the deposit schedules and the forms behind each line above.

On the fee itself, the number to interrogate is not the headline rate but what sits inside it: platform access, benefits administration, filings, support, equipment logistics and offboarding are all sometimes billed separately (read: EOR pricing and cost breakdown).

Once you have an all-in figure, compare it against running your own entity instead of guessing at the crossover point. Our EOR versus entity calculator does that side by side.

Cost, at least, is predictable once modelled. What catches teams out is the compliance calendar sitting behind it.

Which US compliance rules most often derail an EOR implementation?

Five: worker classification, the ACA coverage threshold, WARN notice on layoffs, work-authorization timing, and state rules on final pay. Each has a hard trigger, none of them care that you are new to the country, and your provider should own every filing while you stay close enough to ask for evidence.

Here is what each one actually requires.

1. Worker classification: The IRS applies a common-law test across three categories of evidence, behavioral control, financial control and the type of relationship (IRS classification guidance). A formal Form SS-8 determination can take six months or more, so it is not a tool you use mid-implementation.
2. The cost of getting it wrong: Under IRC section 3509, an unintentional misclassification where you did file a 1099-NEC settles at 1.5% of wages plus 20% of the employee FICA share. Without that filing it doubles to 3% and 40%, and the employer share is always 100%. Intentional disregard voids the relief entirely.
3. The federal test is in flux, so states are doing the work: The Department of Labor stopped applying its 2024 independent-contractor rule in May 2025 and issued a proposal to rescind it on 26 February 2026 (DOL rulemaking status). In the meantime the binding test in many cases is the state one, and California's ABC test, codified at Labor Code sections 2775 to 2785, is the strictest of them.
4. ACA coverage: Once you averaged 50 or more full-time-equivalent employees in the preceding calendar year you are an applicable large employer and must offer affordable, minimum-value coverage to full-time staff or risk a shared-responsibility payment. Full-time means an average of 30 hours a week, or 130 hours a month.
5. WARN notice: Employers with 100 or more employees must give at least 60 calendar days' written notice of a plant closing or a mass layoff affecting 50 or more people at a single site (WARN Act compliance). Violations carry back pay and benefits for up to 60 days per employee, and several states have their own stricter mini-WARN statutes.
6. Work-authorization timing: DHS ended the automatic up-to-540-day EAD extension for renewal applications filed on or after 30 October 2025. Plenty of older guidance still says otherwise, and believing it will cost you a start date.
7. Non-competes: The FTC's blanket non-compete rule was formally removed from the Code of Federal Regulations effective 12 February 2026, and the agency now challenges individual agreements case by case under Section 5 of the FTC Act. State law is what governs your restrictive covenants in practice.
8. Final pay: There is no single federal deadline for a final paycheque. States set it, and some require payment on the day of termination, which means an exit decided on a Friday afternoon can be a compliance event by Friday evening (read: how to terminate an employee properly).

Build a quarterly evidence review into the contract rather than trusting a dashboard, and ask for filing confirmations rather than status labels (see: how to run an EOR compliance audit).

One more US-specific trap is worth flagging in phase one rather than at offer stage: under IRC section 422, an incentive stock option has to be granted to an employee of the granting corporation or its parent or subsidiary. An EOR employee is neither, so ISOs are generally off the table. Use non-qualified options or RSUs from the parent instead, and route the withholding through EOR payroll.

Whichever instrument you land on, align the vesting schedule with the employment start date recorded by your provider, not the date the offer was accepted, or your cap table and your payroll records will disagree at the first exercise.

Knowing the rules is one thing. The mistakes that actually sink implementations are more mundane than any of them.

What are the most common EOR implementation mistakes, and how do you avoid each one?

The recurring six are process failures rather than vendor failures: unclear ownership, no pilot payroll, treating the contract as boilerplate, skipping manager training, no data-security review and no exit plan. Every one of them is cheap to fix inside the first 30 days and expensive after go-live.

Each mistake below is paired with the fix that actually works in practice:

• Unclear ownership between your team and the provider: The fix is one named owner per workflow, written into the agreement before signature rather than discovered during the first escalation.
• No pilot payroll: The fix is one full cycle on real data before go-live. This is the single highest-return week in the whole project, and it is the first thing teams cut when the timeline slips.
• Treating the employment agreement as boilerplate: The fix is a clause-by-clause read of IP, indemnity, notice and termination terms against your own template (see: EOR risk management).
• Skipping manager training: The fix is a one-hour session on the employer boundary. A manager who promises a raise directly to an employee has created a problem that takes weeks to unwind.
• No data-security review: The fix is confirming where employee data is stored, who can access it and how long it is retained before you upload a single record (read: EOR data security).
• No exit plan: The fix is agreeing data portability and employee-transfer terms on day one, so leaving later is a scheduled decision rather than a crisis (see: how to switch EOR providers).

Avoid those six and the only strategic question left is how long you should stay on an EOR at all.

When should you graduate from an EOR to your own US entity?

When headcount concentrated in one or two states stays consistently in the mid-teens, or when you need equity and benefit structures a provider cannot sponsor for you. Below that, the fixed compliance cost of an entity usually loses to the per-employee fee.

The comparison is not only about money. An entity brings you direct control, your own benefit plans and full ISO eligibility, and it also brings state registrations, corporate filings, an audit trail and a payroll function you now staff yourself (read: EOR versus your own entity).

If you are a non-US company hiring into the States, there is a second trigger to watch: activity that creates a taxable presence. Sales-closing authority, a fixed office or a dependent agent can all create permanent establishment risk regardless of who signs the payslip, so involve tax counsel before the first quota-carrying hire.

There is also a middle option people forget. If you already have a US entity and only want the administrative load lifted, a co-employment arrangement may fit better than full employer-of-record service (see: PEO versus EOR).

When you do decide to move, treat it as its own 90-day project with the same discipline as the original rollout, and transfer employment on a payroll-cycle boundary so nobody misses a paycheque (read: transitioning from an EOR to a legal entity).

The incorporation path itself is a separate checklist covering entity type, registered agent, EIN, state payroll accounts and insurance (how to set up a legal entity).

Whichever way that decision lands, the quality of your first 90 days is what makes it a choice rather than a rescue, and choosing the right partner is most of that quality.

Why do companies run their EOR implementation with Wisemonk?

Wisemonk is an India-native employer of record. We built our own entity, payroll engine and compliance stack in-country rather than renting someone else's, and that ownership is precisely why our implementations run to schedule instead of to a partner's queue.

Here is what that means in practice for a company running a 90-day rollout:

• An owned-entity model, not an aggregator, so the employment relationship sits with the company you actually signed with (read: owned entity versus aggregator EOR).
• Onboarding measured in days rather than weeks, because contracts, statutory registrations and benefits enrollment run in parallel from the moment we get your intake sheet (see: our employer of record service).
• Transparent per-employee pricing with no markup buried in payroll funding, quoted separately from pass-through costs so you can compare us honestly against anyone else.
• A dedicated named contact per account instead of a ticket queue, which is the difference between a payroll question answered the same day and one that ages for a week.
• Payroll, tax filings, health cover, equipment logistics and benefits administration under a single contract, so your finance team reconciles one invoice.
• Filing evidence and audit trails you can hand to your own auditors rather than a status dashboard (see: global compliance under an EOR).
• Support for converting existing contractors into employees without a gap in engagement or a fresh classification argument.
• Clean, documented offboarding when a role ends, handled by us and evidenced for your legal team (read: employee termination through an EOR).
• Free tooling to pressure-test decisions before you commit, including a misclassification quiz that takes about two minutes.
• Depth in tech and SaaS specifically, where headcount plans change quarterly (see: EOR for tech companies).

Our clients are global and our operating strength is Indian. Companies across the US, UK, and elsewhere work with us to build and pay teams in India, and we are planning to expand our own market coverage in future.

What do Wisemonk's clients say about their implementation?

They point at the same two things every time: how fast onboarding closed, and whether payments landed when they were supposed to. Three short examples, in their words.

How fast can a whole team go live? Senem RFP did it in days

Frank Menes, Founder and CEO of Senem RFP, needed an entire team on compliant payroll quickly, and cared about how the money moved: "Wisemonk onboarded all of my employees in one or two days. They paid my employees' salaries on the day after my payment cleared… We are an American company, so I was very happy to see that they have a US bank account where I can make ACH payments to minimize bank charges. All salary payments are timely." The implementation outcome was a full cohort live inside a week, ACH funding from a US bank account, and health-plan enrollment handled directly with each employee.

Can you build a specialist team on an EOR? Onereach built one in four months

Saurabh Sharma, Co-founder and CEO of Onereach, was hiring for narrow B2B SaaS skills rather than generic roles: "The Wisemonk team played a key role in helping us hire for specialized B2B SaaS marketing skills. We were able to build the team within four months, and hire experienced professionals from Tier 1/major B2B SaaS brands." The outcome was a full go-to-market bench covering SEO, demand generation, product marketing, content and business development, employed compliantly from day one instead of stitched together from contractors.

What if you need sourcing and onboarding in one track? Cobu ran both together

Dan Sampson, Head of Engineering at Cobu, wanted one accountable process rather than three vendors: "They helped us understand their pricing model, find top-qualified individuals, interview them, and then onboard them. I gave them criteria for the type of people we sought, and they delivered. The individuals they were able to find have been some of the best engineers I have ever worked with." The outcome was sourcing, interviewing and employment onboarding managed as a single implementation track with one owner.

More of these, unedited and in the clients' own words, sit on our reviews page.

If you would rather read the longer versions with numbers attached, our case studies walk through each engagement end to end.

If you want that same 90-day plan applied to your own first hire, the next step is a short conversation rather than another comparison spreadsheet.