Workday: The Buyer and the User

Workday is rated 4.5 stars by the people who buy it and 1.1 stars by the people who use it. The 83-point gap between those two audiences is the entire trade.

On April 28, 2026, Andreessen Horowitz published an analysis titled "Workday's Last Workday?" that opened with a line so clean it became the consensus framing of the stock within 48 hours:

More than 10,000 organizations run it. Tens of millions of employees live inside it. It is approaching $10 billion in annual revenue. Its gross retention rate is 97%. And on Trustpilot, across 462 reviews, it has a 1.1 out of 5.0 stars.

I want you to hold both of those numbers in your head at once. Ninety-seven percent retention. One-point-one stars.

That gap is not a contradiction. It is the architecture of the business. The people who buy Workday are enterprise HR and finance leaders at Fortune 500 companies. The people who review it on Trustpilot are the job applicants who watch their resumes get mangled by the parsing engine and the frontline employees who navigate the system to check their pay stubs. The buyers are satisfied. The users are furious. And neither group has anything to say to the other, because in enterprise software, the buyer and the user are almost never the same person.

Today, June 4, 2026, Workday trades at approximately $157, down 42% from its 52-week high of $257. The question is whether the stock is pricing the 97% or the 1.1.

See the Investment Council's verdict on $WDAY → Investment Council: $WDAY →

All current verdicts: The Verdict Board →

The decline has three layers, and only one of them is about Workday specifically.

The first layer is the SaaSpocalypse. In February 2026, over a 48-hour window, approximately $285 billion of market value was erased from SaaS stocks. The trigger was a cascading realization among institutional investors that AI agents—software that performs work autonomously—would structurally erode per-seat pricing models. If a company deploys an AI agent that does the work of three human employees, it needs three fewer Workday seats. The entire sector was repriced. Workday, as one of the purest per-seat enterprise plays, was hit harder than most.

The second layer is the CEO departure. On February 9, 2026, Carl Eschenbach stepped down as CEO and left the board. Co-founder Aneel Bhusri returned permanently—not as interim, but as the actual CEO. Eschenbach had been sole CEO since February 2024 and co-CEO with Bhusri before that. The market read the departure as instability. On the same day, Workday announced a 2% workforce reduction, its second round of layoffs in twelve months. The first round, in February 2025, had cut 1,750 positions—8.5% of the workforce. Combined: 2,150 jobs eliminated in a year, with $303 million in restructuring charges.

The third layer is the growth deceleration. In its Q4 FY2026 earnings report on February 24, 2026, Workday guided FY2027 subscription revenue growth at 12-13%. That was below some bullish expectations for reacceleration. Analysts at Jefferies downgraded the stock, citing AI disruption risk. Goldman Sachs cut its price target from $206 to $151. The stock hit its 52-week low of $110.36 on April 9.

Then the Q1 FY2027 earnings report on May 21, 2026, cracked the narrative. Revenue of $2.542 billion beat consensus by $25 million. Non-GAAP EPS of $2.66 beat by $0.14. AI product ACV surged 200% year-over-year, with over 4,000 customers now using at least one Workday AI agent—doubled from the prior quarter. The stock jumped 10.8% that day and has climbed further since, landing near $157.

That is the financial surface. Underneath it is the customer story, and the customer story has a statistical anomaly at its center.

Full-year FY2026: revenue $9.55 billion (+13.1%), subscription revenue $8.83 billion (+14.5%), free cash flow $2.78 billion (+27%).

These are not crisis numbers. This is a company growing at low-teens with 32% operating margins, $3.5 billion of net cash, a $27 billion backlog, and 97% gross retention. The financial profile says premium SaaS compounder. The stock is trading at roughly 11x forward revenue, which is in line with other large-cap SaaS names that have decelerated from 20%+ growth. There is no balance sheet problem here.

The question is whether the revenue growth can reaccelerate, hold, or will erode as AI agents shrink enterprise headcounts—and with them, per-seat subscriptions.

We collected customer, employee, and end-user sentiment data from the following channels:

Total sample: approximately 9,700+ reviews and data points across 10 channels.

The central finding of this report is the divergence between how enterprise buyers rate Workday and how end-users rate it. We tested this with a two-proportion Z-test.

Enterprise platforms (G2, Capterra, Gartner Peer Insights, TrustRadius): combined N = 4,626 reviews. Proportion rating Workday 4+ stars (or equivalent): 88.1%.

Consumer/end-user platform (Trustpilot): N = 462 reviews. Proportion rating 4+ stars: 5.0%.

Gap: 83.2 percentage points.

Two-proportion Z = 43.08, p < 0.001. 95% confidence interval on the gap: [81.0%, 85.4%].

This is not close. It is a chasm. And it is not random. On Trustpilot, the dominant review profile is a job applicant who submitted a resume through Workday's talent acquisition portal and watched it get shotgunned into the wrong fields, who received an automated rejection minutes later, who could not reach a human, and who came to Trustpilot to say what a 26-year recruiter wrote: "The absolute worst system."

On G2, the dominant review profile is an HRIS administrator or VP of HR who deployed Workday to unify payroll, benefits, and talent management across a 10,000-employee organization and rates the organizational charting at 9.7/10 and the pay calculation engine at 9.5/10.

These are reviews of the same product. The 83-point gap is not a measurement error. It is the architecture of enterprise software: the buyer optimizes for administrative power, the user experiences the interface, and the feedback loops between them are broken.

The a16z analysis captured this perfectly: Workday's 97% retention is "driven by switching costs, not satisfaction." The implementation costs $300,000 to $1 million. It takes 6 to 18 months. There are 10,500 certified consultants who form what a16z called a "services cartel" that lobbies for the status quo. Once you're in, you don't leave—not because you love it, but because leaving costs more than staying.

The second test examines Workday's internal health. We compared Workday's Glassdoor recommend-to-friend rate against its closest HCM competitor, SAP.

Workday: 60% recommend (N = 4,525).
SAP: 85% recommend (N = 24,962).

Two-proportion Z = -39.57, p < 0.001. The 25-percentage-point gap is statistically significant with massive sample sizes on both sides.

Within Workday, the departmental breakdown contains a red flag: the Sales department shows a 25% recommend rate against the company-wide 60%. In a company whose growth depends on net-new enterprise logos, the people responsible for selling have the lowest morale.

Glassdoor reviewers in the last six months describe a company whose culture has "fallen off a cliff." The specific complaints are consistent: rolling layoffs, RSU elimination that lowered effective compensation, outsourcing of roles to India and Costa Rica, and a leadership narrative around AI that employees describe as "unconvincing and disconnected from reality." One reviewer wrote: "Complete decimation of morale and culture due to widespread layoffs, cutbacks and outsourcing." Compensation ratings have declined 5% year-over-year.

For context: Workday was on Glassdoor's "Best Places to Work" list as recently as 2025. The deterioration is recent, concentrated in the last twelve months, and coincides with the restructuring.

The financials show a 97% retention rate and a $27 billion backlog. What they do not show is how fragile the moat's foundation is.

Workday's stickiness is real but brittle. It is built on three pillars: implementation cost (switching requires a multi-million-dollar project), data gravity (years of HR and financial records are locked inside Workday's proprietary format), and ecosystem lock-in (10,500 certified consultants who have built their careers around the platform). None of these pillars requires the product to be good. They only require it to be painful to replace.

The AI disruption thesis attacks all three simultaneously. If an AI-native HCM system can ingest Workday data, replicate its administrative functions, and deploy in weeks instead of months, the switching cost collapses. This is exactly what a16z's analysis argued—and it is what the market priced into the stock between February and April 2026.

But here is what a16z's analysis missed: Workday is not standing still. In Q1 FY2027, AI product ACV grew 200% year-over-year. Over 4,000 customers are using at least one Workday AI agent, doubled from the prior quarter. The company acquired Sana for $1.1 billion (AI knowledge management), FlowiseAI (low-code AI agent builder), and Pipedream (3,000+ pre-built connectors). It launched "Flex Credits," a consumption-based pricing model that explicitly decouples revenue from headcount. On June 2, 2026, it announced an Agent Passport security framework and a major AWS Data Cloud integration. The stock jumped 20% that day.

The Flex Credits model is the single most important strategic move. If AI reduces enterprise headcounts, per-seat pricing shrinks. Flex Credits let Workday charge for AI agent usage instead—preserving or growing revenue even as the number of human users declines. Whether this transition is fast enough and large enough to offset the structural headwind is the entire question.

Recovering:

Financial performance. Revenue growth 13.5%, margins expanding, FCF $616M in Q1 alone. Not deteriorating. AI product traction. 200% ACV growth, 4,000+ customers on AI agents. Not vaporware. Stock price. Up from $110 low to $157. The relief trade is real. Partnership ecosystem. AWS, Google Cloud, Microsoft, Cisco, Databricks—major cloud partners are integrating, not abandoning.

NOT recovering:

Employee morale. Glassdoor 3.5, declining. Sales recommend at 25%. RSU elimination. Outsourcing. Two rounds of layoffs in twelve months. The people building the product are not happy. End-user experience. Trustpilot 1.1. BBB rating B-. Job applicant experience described as "the worst system" by multiple reviewers. The 1.1 billion rejected applications cited in the AI discrimination lawsuit. Growth rate. 14.3% subscription growth in Q1 is solid but decelerating from 15.7% two quarters ago. The FY2027 guide of 12-13% confirms the trend.

Unknown:

Whether Flex Credits can offset per-seat erosion at scale. Too early to tell—the program just launched. Whether Bhusri's return stabilizes or further disrupts the organization. Employees are skeptical. The market is neutral. The AI discrimination lawsuit outcome. A federal judge allowed ADEA age discrimination claims to proceed in March 2026. The potential class includes "hundreds of millions" of rejected applicants. The liability is unquantifiable. Whether an AI-native competitor can actually replicate Workday's HCM functionality. No one has done it yet, but the window is open.

The Trustpilot sample (N = 462) skews heavily toward job applicants and frontline end-users—people who interact with Workday's talent acquisition portal, not its core HCM or financial management modules. The 1.1 rating reflects the applicant experience specifically and should not be interpreted as a comprehensive assessment of Workday's full product suite. Enterprise review platforms that capture buyer and administrator experiences show 4.1-4.5/5.0 ratings.

The two-proportion Z-test compares different populations (enterprise buyers vs. end-users/applicants). The gap is real and statistically significant, but it measures a known structural feature of enterprise software, not necessarily a Workday-specific deficiency. SAP SuccessFactors and Oracle HCM likely show similar (though probably narrower) buyer-user divergence.

Glassdoor employee ratings have inherent selection bias: employees who review tend to have stronger sentiments (positive or negative) than those who don't. The 5% decline in compensation ratings should be interpreted directionally, not as a precise measurement.

Reddit data was gathered via search indices due to access limitations. Exact post counts and date distributions should be considered approximate.

Bear case (35% probability): The SaaSpocalypse thesis is correct. AI agents reduce enterprise headcounts, shrinking per-seat demand. Flex Credits cannot scale fast enough. Growth decelerates to single digits by FY2028. Employee attrition in engineering and sales (25% recommend) hollows out the product and pipeline. The AI discrimination lawsuit results in material liability. An AI-native competitor launches a credible alternative with faster implementation. Stock returns to $100-$110 range.

Base case (45% probability): Growth holds at 12-14%. Flex Credits gradually shift the revenue mix, stabilizing the per-seat exposure. The AI product line reaches $1 billion ARR by FY2028. Bhusri's return steadies the organization, but employee morale remains below 2024 levels. The lawsuit settles for a manageable amount. No AI-native competitor achieves meaningful enterprise share in the next 18 months. Stock trades in the $155-$200 range.

Bull case (20% probability): AI product traction accelerates beyond 200% ACV growth. Flex Credits adoption is faster than expected, proving the model can grow revenue regardless of headcount trends. Workday Rising in October 2026 is a catalyst event (major product launches, customer case studies). Elliott Management's $2 billion stake catalyzes further operational improvement. Subscription growth reaccelerates to 16%+. Stock recovers to $220-$250.

Now: The stock is fairly valued in the base case. The weighted expected value of $159 implies essentially no upside from current levels—the base and bull scenarios offset the bear. This is not a screaming buy.

The catalyst: Workday Rising US, October 12-15, 2026, at the Venetian in Las Vegas. This is Workday's annual conference with 10,000+ attendees. Expect GA releases for the Sana, Pipedream, and FlowiseAI integrations into the Illuminate platform. If management can demonstrate that Flex Credits are converting pipeline at scale and that the AI agent count has grown past 8,000-10,000 customers, the growth narrative shifts from "holding" to "reaccelerating." That would move the stock.

The decider: Q2 FY2027 earnings, late August or early September 2026. The market needs to see subscription growth hold at 14%+ (above the 12-13% guide), and it needs to see AI product ACV sustain 200%+ growth for a second consecutive quarter. If both happen, the bull case probability rises from 20% to 35% and the weighted expected value moves to $185+. If subscription growth slips to 11-12% and AI ACV growth decelerates, the bear case probability rises and the stock tests $130.

When Workday Rising opens on October 12, subscribers will get our updated analysis covering:

Whether Flex Credits have converted pilot customers to paid at scale—the specific metric is new ACV attributable to consumption-based pricing. Whether the AI agent customer count has grown from 4,000+ to 8,000+, validating the doubling cadence. Whether the Sana acquisition ($1.1B) has produced a GA product or remains in integration limbo. How the AI discrimination lawsuit has progressed—whether Workday has disclosed the customer list as ordered by the court. Whether the Sales department's 25% Glassdoor recommend rate has improved or deteriorated further—because if the people selling the product don't believe in it, the pipeline will eventually show.

The 83-point gap between buyer satisfaction and user satisfaction is not a bug in Workday's model. It is the model. The bet is whether that gap is a moat or a vulnerability—whether the switching costs hold long enough for Flex Credits to build a new revenue engine, or whether an AI-native competitor collapses the implementation barrier and turns Workday's most-important-and-least-loved status into just least-loved.

We'll have the answer—or at least the first real data point—in October.

See the Investment Council's verdict on $WDAY → Investment Council: $WDAY →

All current verdicts: The Verdict Board →