Hidden Labour Costs in Benefits Card Infrastructure Could Reach $33,750 a Year, Report Finds
A PYMNTS Intelligence report co-produced with SoFi Tech Solutions models how manual exception-handling in a 5,000-account health and wellness benefits programme could generate approximately $33,750 in hidden annual labour costs. The analysis argues these expenses are rarely itemised, leaving programme administrators unaware of the true cost of legacy card infrastructure.

Benefits card administrators running on legacy platforms may be absorbing significant untracked labour costs that only become visible when exception-handling is modelled as a discrete operational expense.
Hidden Labour Costs Lurk Inside Benefits Card Infrastructure, PYMNTS Report Finds
A PYMNTS Intelligence report co-produced with SoFi Tech Solutions finds that manual exception-handling inside health and wellness benefits programmes can generate tens of thousands of dollars in concealed annual labour costs — costs that programme administrators may not be tracking at all. The report, titled Making Health and Wellness Benefits Perform: What Modern Card Infrastructure Changes, was published in September 2026 as part of PYMNTS's tracker and embedded finance series, and is available via free registration on the PYMNTS site.
The Cost Model in Detail
The report's central illustration models a 5,000-account health and wellness benefits programme under a set of explicitly stated assumptions. Under that model:
- A programme of that scale generates approximately 250 manual exceptions per month
- Each exception is assumed to require 15 minutes of staff time to resolve
- That produces roughly 62.5 staff hours consumed per month — arithmetic that is internally consistent (250 × 0.25 hr = 62.5 hr)
- Applying an assumed $45 loaded hourly labour cost, the model produces an annualised hidden labour expense of approximately $33,750
These figures are illustrative model outputs, not independently validated operational benchmarks. The 15-minute resolution time and the $45 loaded hourly rate are model inputs; no independent HR operations study or Bureau of Labor Statistics benchmark is cited in the publicly available summary materials to corroborate them. The $33,750 annualised figure does, however, appear in headline coverage by Insurance Business Magazine of the same report, confirming the figure is drawn directly from the underlying PYMNTS analysis.
According to the report's modelling, a 5,000-account benefits programme could be absorbing the equivalent of more than 60 staff hours every month in exception resolution alone — a cost that may not appear in any single line of an administrator's budget.
Why Exceptions Happen
The PYMNTS summary article attributes the volume of manual exceptions to the interaction between card systems, merchant category codes, IRS eligibility rules and real-time account balance logic — a combination of requirements that, the summary notes, is materially more complex than those governing simpler corporate card programmes. The same article frames legacy card platforms as not having been built for real-time eligibility decisioning — though this characterisation reflects the framing of the PYMNTS summary article specifically, rather than an independently verified technical assessment.
Why It Matters
The report's scope is health and wellness benefits accounts — consistent with its published title. The PYMNTS summary article references the broader category of health and wellness benefits programmes rather than enumerating specific account structures; administrators considering how these findings apply to particular product types should consult the full report directly.
For the benefits administration and payments infrastructure industries, the significance of the model is less in any single figure than in the methodology it surfaces. Exception-handling costs are typically absorbed into general operations overhead and are rarely itemised, meaning the effective cost of running benefits card infrastructure on legacy platforms can be structurally obscured. Framing these costs in labour-hour terms — as the PYMNTS model does — provides administrators with a unit-economics lens that is easier to apply to procurement and technology investment decisions.
About the Co-Producer
SoFi Tech Solutions describes itself on its official website as a financial technology platform provider supporting banks, fintechs and brands with cloud-native, API-first infrastructure. The company states it supports 135 million accounts across 14 countries. The company co-produced the PYMNTS Intelligence report; the publication does not constitute an independent editorial assessment of SoFi Tech Solutions' products or capabilities.
The full report requires free registration to access. The Fin Desk was not able to independently review the complete document; coverage is based on PYMNTS's publicly available summary article and Insurance Business Magazine's headline coverage of the same report.
The Fin Desk Newsroom publishes verified reporting on the developments shaping fintech, payments and modern financial infrastructure.
Related Stories

Global Payments acquires Worldpay, sells payroll unit for $1.1bn in merchant-focused pivot
Global Payments has closed its acquisition of Worldpay, sold its payroll business to Acrisure for $1.1 billion and announced plans to divest its Issuer Solutions unit, executing a sweeping portfolio shift toward merchant payment technology across 2025 and into 2026. The Issuer Solutions buyer, price and closing timeline remain unconfirmed.

Agentic Payments Expose Liability Gap as Card Network Rules Lag Behind AI Automation
AI agents are completing purchases autonomously using stored credentials, but card network dispute and chargeback frameworks were written for human-authorised transactions — leaving liability for agent-initiated payments legally unresolved. Legal commentary from Goodwin Procter and Astraea Counsel confirms no clear rule yet determines who bears the loss when an AI tool acts beyond its mandate.

AI agents enter cross-border payments but cannot fix the structural flaws beneath
AI agents are being integrated into cross-border payment pre-processing workflows, including counterparty verification and sanctions screening, but primary research from the BIS, Federal Reserve and SWIFT consistently identifies structural, compliance and data-quality failures that technology acceleration alone cannot resolve.

Basware Acquires Fraud Prevention Platform Trustpair in Undisclosed Deal
Basware, an invoice lifecycle management firm, has completed its acquisition of Trustpair, a payment fraud prevention platform, in a deal whose financial terms were not disclosed. The combination targets AI-powered payment fraud and supplier impersonation risks within enterprise accounts-payable workflows.
The essential developments in modern finance
The essential developments across fintech, payments and modern finance — delivered to your inbox.
Free. No spam. Unsubscribe anytime.