State of Employment Verification 2026

The first look inside how employment verification actually works in 2026, combining more than 170,000 real verifications run for banks and background screeners with published industry data on the market's structure.

Why this report exists

Employment verification is the quiet load-bearing step in every mortgage, apartment lease, and background check, and almost no one measures it. The public numbers that exist come almost entirely from the incumbent database that dominates the category, so the picture is partial by design.

Two things make it worth correcting now.

Employment verification is the largest slice of the screening market. It generated the biggest share of background-screening revenue in 2025, roughly 62% of the category (Mordor Intelligence, 2026). The broader background-screening market sat around $7-8B in 2026 and is growing at a high-single-digit to low-double-digit CAGR depending on the analyst (Mordor Intelligence; Fortune Business Insights, 2026).

The cost of getting it wrong is climbing fast, and AI just changed the math. That story runs through the next section.

This report pairs what we see in our own production data with what the public record says about the market around it.

The cost of getting verification wrong is rising

Verification used to be a box-check. In 2026 it's a control under real pressure, because the thing it checks against, the self-reported resume, is getting cheaper to fake and more expensive to trust.

Fraud is closer to the base rate than the exception. Surveys consistently find that more than half of job applicants admit to having lied on a resume, with one widely cited figure putting it at 70% of U.S. workers at least once (CrossChq, 2024). Roughly 85% of hiring managers report having caught a lie, and about 1 in 4 background checks surfaces a discrepancy between what a candidate claimed and what's actually true (AMS Inform, 2024). It concentrates in employment history: overstating length of employment and inflating job titles are perennially among the most common fabrications (ResumeLab; SHRM), which are exactly the fields an employment verification confirms.

AI changed the economics of lying. Gartner projects that by 2028 as many as 1 in 4 candidate profiles worldwide could be fake, some using AI-generated audio and video to get through screening (Gartner, 2025). That future is already visible in the pipeline: about 17% of hiring managers report having encountered candidates using deepfake technology in interviews (Huntress, 2026), and in one real-time screening window between September and November 2025, 23.2% of applicants were flagged as a fraud risk (Huntress / Endorsed, 2026). Detection confidence is moving the wrong way, with only about 19% of hiring managers extremely confident their current process would catch a fraudulent applicant (Checkr, 2025). Discrepancies themselves are closer to the norm than the exception: more than three-quarters of employers globally found at least one discrepancy in a candidate's background over the past year, and employment verification is consistently where the most inconsistencies surface (HireRight, 2025).

The downstream cost is real money. Resume fraud is estimated to cost U.S. businesses on the order of $600B a year (CrossChq, 2025). A single bad hire traced to a fraudulent resume can run $17,000 or more (AMS Inform, 2024), and turnover from bad hiring decisions costs 16-30% of an entry-level employee's annual salary (Equifax, 2026). In regulated and safety-sensitive roles, an unverified hire isn't just a productivity loss, it's negligent-hiring liability.

Why this lands on verification. As fabrication gets cheaper and more convincing, a document the candidate wrote (and increasingly, one an AI wrote for them) is worth less, and an independent check against the employer's own records is worth more. Verification is the one step in the stack that doesn't take the applicant's word for it. The rising cost of fraud is, functionally, a rising premium on getting verification right, and getting it right fast.

Methodology

The Superunit figures below are based on 170,000+ employment verifications run through our platform between February 2025 and June 2026, on behalf of 50+ customers (banks, mortgage lenders, and background screening companies). All figures are aggregated and anonymized. Timing statistics are drawn from 100,000+ completed verifications.

One scope note: this reflects the verification activity of screeners and banks, not a census of the whole labor market. That specificity is the point. It's the slice no incumbent database can measure. Industry figures are attributed inline and listed in full under Sources.

Finding 1: One company runs most of the instant-verification layer, and that layer covers less than half the workforce

Superunit data: When an employment verification routes through a named third-party platform, about 73% of the time that platform is Equifax's The Work Number. In our data, the instant-database tier isn't a competitive market. It's effectively one incumbent.

The scale. The Work Number holds more than 823 million employee records contributed by nearly 5 million employers (Equifax, 2026). It fulfilled roughly 149 million verification requests in 2024 (Equifax figures via 3rd Degree Screening, 2026) and helped 58 million people get verifications outside regular business hours in 2025 alone (Equifax, 2026). Equifax built the position by acquiring TALX in 2007 for $1.4B (GoodHire, 2026), and the record count has grown from about 296 million in 2017 to 823 million-plus today. Concentration is accelerating, not easing (Fast Company, 2019; Equifax, 2026).

The gap that matters most. Despite that scale, only about 40% of U.S. employers participate in The Work Number. For the remaining ~60%, a verifier falls back to direct contact by phone, email, or fax (PreciseHire, 2026). Independent screening guides confirm the pattern: coverage gaps are widest among small and mid-size employers, and a "no record found" result is not a negative result, it's an absence of data that triggers manual work (GoodHire, 2026).

The pricing consequence. Concentration shows up in the invoice. Equifax raised the price of employment verifications run through The Work Number again effective January 1, 2026, with pricing set by Equifax rather than by the screening vendors who resell it (Mitratech, via JD Supra, 2026). Vendors have described successive annual increases (SwiftCheck, 2024), and the long-run trend is stark: a $20 verification fee that once looked steep now reads as a bargain (3rd Degree Screening, 2026).

The pushback has reached the courts. A proposed class action filed by the St. Louis Housing Authority alleges Equifax leaned on its "near-monopoly" position to raise Work Number prices by as much as 300% within a matter of months, using contract terms that let it change pricing on 30 days' notice (ClassAction.org, 2025). The complaint claims the agency couldn't route around the increases because employers it contacted directly redirected it back to The Work Number, a clean illustration of what happens when one database becomes the default (ClassAction.org, 2025). These are allegations, and the case is proposed rather than proven, but the dynamic it describes matches what our data shows.

Independent market estimates put Equifax's share of the vendor-to-vendor verification market above 85%, higher still by revenue (In Practise expert interview, 2025), broadly consistent with the concentration we measure in our own data.

The takeaway: These numbers aren't in tension, they're the whole story. The instant layer has the pricing power of a monopoly and still can't reach the majority of the workforce, so most verifications require someone to actually make contact. That's the market underneath the database, and it's the one no one has been measuring.

Finding 2: Verification is slow almost everywhere, and the manual majority is the reason

Superunit data: Across 100,000+ completed verifications, the median time to verify was 15.5 hours, with 64% completing within 24 hours and about 1 in 6 completing within the first hour. The 90th percentile stretched to roughly 91 hours. The long tail is where the pain lives.

For the ~60% of verifications that fall outside the instant database, published turnaround benchmarks are measured in days, not hours. Direct employer contact typically runs 3-5 business days, and international or complex cases 7 or more (InfoMart, 2026). A standard employment background check ranges from minutes to five business days depending on whether the employer is reachable (GoodHire, 2024). Even in the best case, a manual verification of employment can close in a single business day when everything goes right, and routinely runs longer when it doesn't (Truework). The incumbent itself markets Priority Next Day and Priority Two Day manual options, an implicit acknowledgment that its non-instant path is a multi-day process (Equifax, 2021).

The takeaway: The industry's own published benchmark for the manual majority is days. A 15.5-hour median and a majority inside 24 hours is the number a lender or screener feels directly, because slow verification is where deals and offers stall.

Finding 3: In 2026, verification still runs on phone, email, and fax

Superunit data: Among verifications where the successful channel is known, the mix was roughly 52% email, 46% phone, and about 1% fax. Fax is a rounding error in volume, but it isn't zero, in 2026.

When an employer isn't in the instant database, the fallback is explicitly a phone call, an email, or a fax (PreciseHire, 2026). Industry write-ups still describe the default failure mode as a lender faxing a request on Saturday, HR seeing it Monday, and the applicant losing the car loan by Wednesday (Vetty, 2026). Roughly 23% of verifiers work evenings or weekends, when HR desks are dark, part of why asynchronous channels still dominate the mix (Vetty, 2026).

The takeaway: The category most people assume is digital and instant is, for most of its volume, still email tag and the occasional fax. That's the friction, quantified.

Finding 4: Where you're verifying changes everything

Superunit data: Cleaned and normalized by country, and filtered to exclude blocked-identifier requests, the cross-border picture:

UK employment verification takes roughly 3.3x longer than the US (57h versus 17h median), with only about a third of UK verifications completing inside 24 hours against two-thirds in the US and Canada. The US and Canada run neck-and-neck on the median; the UK is the clear outlier.

The public record agrees directionally: international or complex prior-employment verification is routinely quoted at 7 or more business days, and overseas credential checks at one to three weeks (InfoMart, 2026; PreciseHire, 2026). Cross-border hiring is a named growth driver for the screening market, which means this friction is expanding, not shrinking (Mordor Intelligence, 2026).

The takeaway: Cross-border verification-reachability numbers are almost impossible to find, because the incumbent database is overwhelmingly U.S.-centric. Verification difficulty isn't uniform. It's a function of where the employer sits.

How to cite this report

Superunit, State of Employment Verification 2026. [URL]

Each finding is written as a standalone, citable statistic. Superunit figures are attributable to Superunit; industry figures carry their original source.

Sources

Industry figures in this report are drawn from public sources, including:

• Equifax / The Work Number: database scale, record counts, verification volumes, manual-VOE turnaround options (theworknumber.com; Equifax press, 2021-2026)
• Mitratech (via JD Supra): January 1, 2026 Work Number VOE price increase (2026)
• SwiftCheck: successive annual Work Number price increases; top-of-waterfall pricing (2024)
• ClassAction.org: St. Louis Housing Authority proposed class action alleging up to 300% Work Number price increases (2025)
• In Practise: independent estimate of Equifax verification market share (expert interview, 2025)
• 3rd Degree Screening; Pre-Employ: verifier commentary on Work Number cost trends (2022-2026)
• GoodHire: The Work Number coverage gaps; background-check turnaround ranges (2024-2026)
• PreciseHire: ~40% U.S. employer participation in The Work Number; manual fallback channels (2026)
• InfoMart: direct-contact and international verification turnaround benchmarks (2026)
• Truework: manual verification-of-employment turnaround (verification knowledge base)
• Vetty: manual and fax workflow friction; off-hours verifier share (2026)
• Mordor Intelligence: background-screening market size, employment-verification revenue share, cross-border growth (2026)
• Fortune Business Insights: employment-screening market size and CAGR (2026)
• CrossChq: resume-fraud cost estimate; share of workers who have lied on a resume (2024-2025)
• Gartner: projection that 1 in 4 candidate profiles will be fake by 2028 (press release and report Mitigate Rising Candidate Fraud Through Identity Verification, 2025)
• Huntress / Endorsed: deepfake-encounter rate and internal Sept-Nov 2025 applicant fraud-flag rate (2026)
• Checkr: share of hiring managers confident of catching a fraudulent applicant (2025)
• HireRight: global candidate-discrepancy rates and where discrepancies concentrate (2025 Global Benchmark Report)
• AMS Inform: bad-hire cost; hiring-manager lie-detection rate; discrepancy rate (2024)
• Equifax (Total Verify blog): entry-level turnover cost, citing Harvard Business Review (2026)
• ResumeLab, SHRM: resume-lie prevalence and most-common-lie breakdowns
• Fast Company: historical Work Number record count (2019, referencing a 2017 investigation)

Figures reflect the most recent data available from each publisher at the time of writing. Where a source updates its figures periodically, the current number may differ.