HR Data Extraction and HRIS Portability When Leaving a PEO

The hardest part of leaving a PEO is getting the employer's own employee data back in a form anyone can actually use. Most companies planning an exit spend their energy lining up a new payroll provider and a new benefits broker, and assume the data will simply follow, exported in full the moment someone asks for it. That assumption usually breaks somewhere in the middle of the transition, when a request for "all our employee data" comes back as a partial file, a PDF archive, or a flat no.

The PEO's portal is the PEO's system. Data sits inside a multi-tenant platform built to serve many client companies at once, and the employer using it does not own or control that platform the way it would its own HRIS. Access to it exists only as long as the co-employment relationship does, and the clock on retrieving anything starts the moment notice is given. Companies that wait until the final weeks of a contract to think about data extraction are negotiating from a position that has already weakened.

The underlying cause is the co-employment model itself. The same structure that makes a PEO efficient during the engagement becomes a liability the moment the engagement ends, because the employer's data was never truly separable from the PEO's own operating system.

Contracts often do little to soften that liability. Employers who entered co-employment expecting a clean handoff at departure frequently discover that the language governing data portability is vague, or written entirely to the PEO's advantage. As more companies grow past the headcount and cost threshold where PEO pooling advantages fade, more of them are hitting this problem directly, not as an edge case but as a routine part of scaling past the size where a PEO makes financial sense. Costs tend to rise with every hire once a company passes that threshold, making data planning a problem for a much larger population of operators than it used to be.

The HRIS data categories and the formats they take

Before an employer can negotiate an exit or execute one, it needs an accurate map of what is actually sitting inside the PEO's HRIS. Treating all of it as a single "export" is the first mistake most companies make, and it is the one that creates the gaps discovered later, often too late to fix cleanly. The data falls into distinct categories, and each one has its own export path, its own format quirks, and its own level of employer entitlement.

Employee records make up the most obvious category: personal information, employment history, compensation history, and performance documentation. HR documents form a fourth category: offer letters, signed handbooks, disciplinary records, I-9s, and onboarding paperwork. Time and leave records, including accrued PTO balances and leave history, make up a fifth. Tax documents, including W-2s issued under the PEO's EIN and quarterly filings, round out the list.

The assumption that a single export button will produce all of this in usable form is what creates the gaps employers discover mid-transition. Each category tends to require its own request, often its own format, and its own confirmation that what came back is actually complete.

The W-2 split deserves particular attention because it produces a concrete, employee-facing problem. That split creates confusion at tax time for employees who have no reason to understand the mechanics of co-employment, and it requires the employer to communicate proactively rather than let employees discover two tax forms in their mailboxes without warning. Timing an exit at the end of a calendar year rather than in the middle of one is one of the strongest practical arguments for this approach, a point the transition timeline later in this piece returns to directly.

What employers are contractually entitled to

What an employer can actually extract, in what format, and on what timeline is governed by the PEO contract itself, not by any general principle about who owns employee data. Reading that contract before giving notice is the step that determines how much leverage the employer has for everything that follows.

The termination clause carries the most weight. It defines the notice period, which most PEO agreements set within a specific window. It defines the triggers for an early termination fee. And in the best cases, it defines, in specific terms, what data portability and post-exit access actually look like. Data portability language varies enormously across PEO contracts. Some agreements commit the PEO to delivering a full export in a specified format within a defined window. Others leave format, timeline, and completeness entirely to the PEO's discretion, which in practice means the employer finds out what it's getting only after notice has already been given and much of its leverage has already been spent.

State-level mechanics add another layer that the contract alone won't fully resolve. In states where the PEO files unemployment taxes under its own EIN, often called PEO-level states, the employer has to establish its own State Unemployment Insurance account from scratch at exit. In states where the employer already holds its own SUI account and has simply granted the PEO third-party access to file on its behalf, often called client-level states, the transition is more straightforward, but the employer still has to revoke the PEO's access formally and confirm it can file directly before the handoff is complete.

The employer's negotiating position is strongest in the window before formal notice goes out. Once notice is submitted, the PEO has little incentive to renegotiate format, timeline, or completeness, because the relationship is already ending on terms the contract already sets. Written confirmation that notice was received and accepted matters just as much. Verbal assurances from an account manager do not hold up when a dispute surfaces three weeks before a final payroll run.

The 90-day extraction and transition sequence that protects data continuity

A clean handoff depends on running three tracks at the same time: contract and legal work, data extraction, and setup of the replacement system. These tracks have interdependent deadlines, and running them one after another instead of in parallel is what causes the whole timeline to collapse under its own weight. A 90-day sequence, broken into three 30-day phases, gives each track room to move without starving the others.

The first 30 days belong to contract and planning work. Sending notice before this groundwork is complete is the single most common mistake in the entire process: companies give notice before they've confirmed a replacement stack even exists, and the data and coverage gaps that follow are a direct result of that sequencing error.

The next 30 days cover sourcing and setup. Employers should verify that all data has actually been received before the final exit date arrives, not after it has already passed and there's no PEO account left to call.

The final 30 days cover transition and handoff. The new provider goes live, the first independent payroll run gets verified end to end, and every exported dataset gets checked to confirm it has loaded correctly into the new HRIS.

One piece of this sequence gets skipped more often than any other, and it causes more disruption in the first week of independence than any data export ever does: custom workflow logic. Approval chains, routing rules, and automated processes built inside the PEO's system fail to export. They have to be documented by hand during the planning phase and rebuilt inside the new system before go-live, or the new HRIS launches without the operational logic the HR team has relied on for years.

Timing matters as much as sequencing. Exiting at the end of the calendar year, or at the end of the benefits plan year, eliminates the W-2 duplication problem described earlier and aligns payroll tax wage base resets with the transition, cutting down both employee confusion and administrative cleanup afterward. Benefits that reset on a plan-year basis, FSAs especially, can create real coverage gaps if the exit date and the plan year don't line up.

Where data loss happens and how to prevent it

Data loss in a PEO exit almost never happens as one dramatic failure. It accumulates across a handful of predictable gaps, each of which looks manageable on its own, until they stack into a compliance problem or an operational one.

The first gap comes from incomplete export requests. Employers ask for "employee records" and receive only the core employee profile, missing compensation history, document attachments, and time-off balances entirely, because each of those lives in its own module and requires its own separate request. The prevention is specific: request exports by module name, in writing, against a checklist, and do not rely on the PEO to interpret a vague request comprehensively on the employer's behalf.

The second gap is a format mismatch. The PEO delivers a PDF archive or a proprietary file format that the new HRIS has no way to ingest, forcing manual re-entry that introduces errors and burns hours of HR staff time that could have gone elsewhere. The fix is to confirm the target HRIS's import format requirements before ever requesting exports from the PEO, and to specify that required format, typically CSV with a defined field mapping, directly in the written export request. Enrollment records deserve the same caution: collect them for reference, but plan re-enrollment with the new carrier as a fresh process rather than assuming PEO enrollment data will transfer directly into the new system.

A third gap sits in payroll tax records. Taxes filed under the PEO's EIN during the co-employment period belong to the PEO's own tax record, not the employer's, and there is no mechanism to transfer that filing history onto the employer's own EIN. The employer still needs copies of that payroll history for its own reference, covering both year-to-date and prior-year data, and those copies should be stored as permanent records regardless of whether any of it loads into the new system.

A fourth gap involves 401(k) and other retirement plan records. If the retirement plan is held under the PEO's own plan umbrella rather than the employer's, the employer has to establish its own plan and execute a transfer or rollover, a process with its own timeline that cannot be compressed into the final weeks of a transition. Identifying early whether the 401(k) sits under the PEO's plan, and engaging a retirement plan administrator during the planning phase rather than the transition phase, is what keeps this gap from turning into a last-minute scramble.

The compliance obligations that activate the moment co-employment ends

Retrieving data is the precondition for meeting a set of legal obligations that the employer has owned all along, even while the PEO was the one filing the paperwork. The moment co-employment ends, every compliance function the PEO was administering reverts to the employer immediately and in full, because the underlying legal exposure always belonged to the employer. It was only administered on the employer's behalf.

Under the federal health-coverage law governing large employers, the employer mandate applies to any employer with 50 or more full-time-equivalent employees, regardless of whether it uses a PEO. The PEO handled the administration of coverage offers, but the employer was always the party legally responsible for them, and failure to provide Forms 1095-C after the exit carries penalties assessed per form.

Under ERISA, Form 5500 must be filed by the last day of the seventh month following the end of the plan year, absent an extension, and failure to file triggers substantial daily penalties. The payroll and benefits records extracted from the PEO during the transition are the source data this filing depends on, which is part of why incomplete exports create compliance risk and not just administrative inconvenience.

COBRA brings its own set of deadlines. Employers with 20 or more employees must offer continuation coverage, and notices have to go out promptly following a qualifying event such as a termination. Penalties for failing to provide proper notice include significant per-day, per-employee charges, on top of potential lawsuits. The obligation starts at the qualifying event itself: the employer has 30 days to notify the plan administrator, and the plan administrator then has to send election notices to qualified beneficiaries promptly after receiving that notification, for any employee who isn't moving onto the new plan.

Multi-state registration adds a geographic dimension to all of this. Employers typically have only a short window from the first wages paid in a new state to complete registration there, and a single remote employee is enough to create nexus, triggering withholding and filing obligations in that state. The PEO's filing history under its own EIN does not carry over, so the employer has to register independently in every state where it has employees. State unemployment insurance is the sharpest version of this problem: in PEO-level states, the SUI account was held under the PEO's EIN, and the employer has to open its own account with no prior claims history behind it for rate-setting purposes.

Fiduciary responsibility under ERISA, ACA, and COBRA stays with the employer no matter which vendors it chooses afterward. Brokers, third-party administrators, and payroll providers all act on the employer's behalf, but the legal obligation itself sits with the employer, exactly as it did before the PEO relationship began and exactly as it will after the exit is complete.