ZipRecruiter: A Case Study in Platform Economics

The ads are everywhere. Radio spots promise to connect employers with qualified candidates “fast.” Bill Maher reads copy about revolutionary hiring technology. Banner ads tout AI-powered matching. ZipRecruiter’s marketing budget is clearly substantial. Yet a curious pattern emerges when you look at what actual users report about the platform.

The Three-Sided Problem

ZipRecruiter operates as what economists call a two-sided marketplace—platforms designed to connect distinct groups who provide value to each other. In theory, more job seekers attract more employers, and more employers attract more job seekers, creating a virtuous cycle. Airbnb connects travelers and hosts. Uber connects riders and drivers. ZipRecruiter connects job seekers and employers.

Except ZipRecruiter’s marketplace has quietly become three-sided: job seekers, employers, and third-party recruiters who use the platform for lead generation rather than actual hiring. This third group fundamentally changes the dynamics.

The Ghost in the Machine

Recent research has documented a phenomenon that users have been experiencing for years: a substantial portion of online job postings don’t represent real hiring intent.

The Congressional Research Service issued a report in April 2025 defining “ghost jobs” as online job postings for positions that do not exist, or that employers are not planning to fill immediately. Multiple studies have attempted to quantify the scale:

  • ResumeUp.AI’s analysis found 27.4% of U.S. job listings on LinkedIn are likely ghost jobs
  • A Clarify Capital survey found nearly 1 in 3 employers admit to posting job listings with no intention of hiring
  • Resume Builder’s survey of hiring managers discovered that 81% of recruiters have posted ghost jobs
  • Greenhouse Software reported that 18-22% of online job ads are fake or unfilled

According to the CIO analysis of these studies, ZipRecruiter appeared on 58% of recruiters’ lists for where they post ghost jobs—third behind company websites (72%) and LinkedIn (70%), but ahead of Indeed (49%) and Glassdoor (48%).

The reasons employers give for posting non-existent positions are revealing: keeping a “warm talent pool” for future hiring, appearing to be growing, testing market salary expectations, making current employees feel replaceable, or satisfying corporate directives to post positions even when budgets are frozen.

Meet Phil: The Algorithm Personified

ZipRecruiter’s matching system has a name: Phil. Phil is an AI that sends job recommendations to users, often multiple times per day. Users report receiving 8-10 emails daily from Phil.

The quality of these matches reveals something fundamental about how the platform is optimized. A CPA reports being sent bookkeeping positions and jobs “totally out of my wheelhouse.” An educator receives suggestions for rendering truck driver positions. Multiple users in customer service report being flooded with commission-based sales jobs they’ve explicitly indicated no interest in.

The pattern suggests Phil isn’t optimized for match quality—he’s optimized for engagement. More emails mean more clicks mean better metrics to show advertisers and investors. Whether those clicks lead to actual employment is a secondary concern.

The Resume Black Hole

Users report a peculiar phenomenon: ZipRecruiter’s system indicates their application has been “viewed” by employers—sometimes 2-3 times—yet they receive no response. One user reported 50 applications over a week, with most marked as “viewed” multiple times, resulting in zero callbacks.

This creates the appearance of activity without actual progress. The platform can show engagement metrics (applications submitted, profiles viewed) that justify its value to both users and investors, even if the fundamental purpose—connecting people to jobs—isn’t being fulfilled.

The Recruiter Swarm

Multiple users report a specific pattern: applying to a position on ZipRecruiter, then immediately receiving calls or texts from third-party recruiters about different positions. Some report these contacts have Indian accents and ask them to call back about opportunities, then never follow through. Others report the jobs these recruiters pitch bear no resemblance to what they applied for.

Some ghost jobs appear to be posted by staffing agencies specifically to harvest resumes and contact information. The position listing serves as bait; the actual product is the candidate data itself. One user reported that after using ZipRecruiter, they began receiving multiple daily scam calls, suggesting their information had been sold or leaked to bad actors.

The One-Click Paradox

ZipRecruiter’s marquee feature is “one-click apply”—upload your resume once, then apply to multiple jobs with a single click. Users cite this as the platform’s main advantage: it eliminates the tedious process of re-entering the same information for every application.

But this convenience creates its own problem. The ease of application means employers are flooded with candidates, many of whom haven’t carefully considered whether they’re actually qualified or interested. This drowns out serious applicants and incentivizes employers to use increasingly arbitrary filtering mechanisms. The efficiency of applying is offset by the inefficiency of the signal.

From a platform economics perspective, one-click apply is brilliant: it generates massive application volume (a metric ZipRecruiter can tout to employers) without requiring actual job seekers to invest much effort (keeping them engaged with the platform). Whether these applications result in hires is, again, secondary to the appearance of activity.

Why It Persists

Given all these problems, why does ZipRecruiter continue to attract both job seekers and employers?

For job seekers: desperation. When you’re unemployed, you can’t afford not to use every available tool, no matter how ineffective. The variable ratio reinforcement of occasional success stories keeps hope alive.

For employers: they’re buying reach. ZipRecruiter distributes postings across multiple job boards, creating the appearance of a wide search. Whether these postings attract quality candidates is less important than satisfying internal or legal requirements to conduct a “thorough” search.

For third-party recruiters: it’s a lead generation tool. The platform provides access to a constant stream of job seekers actively looking for work—a valuable commodity regardless of whether any specific posting is legitimate.

For ZipRecruiter itself: the business model works. Employers pay to post (or to access candidates). Job seekers provide free labor in the form of their data and attention. The platform can show engagement metrics without needing to prove actual hiring outcomes. As long as venture capital funding or revenue from postings continues, the incentive is to optimize for volume and engagement, not match quality.

What It Reveals

ZipRecruiter isn’t unique in these problems—Indeed, LinkedIn, and other job boards face similar criticism. But the platform’s heavy advertising presence creates a particular dissonance between promise and reality.

The broader issue is that platform economics have created a system where everyone’s incentives are misaligned with the ostensible purpose. Job boards make money from posting volume and user engagement. Employers want to appear to be hiring, or to harvest data about the labor market. Third-party recruiters want leads. Job seekers want actual jobs.

Only one of these groups is unable to get what they want from the current system.

The ghost job phenomenon—now documented by Congress and multiple research organizations—represents a kind of market failure. A system designed to match labor supply with labor demand has instead created an elaborate theater of employment, where positions are posted, applications are submitted, profiles are viewed, and very little actual hiring occurs.

When a Congressional Research Service report has to explain that official government labor statistics (JOLTS) likely don’t suffer from the ghost job problem because “if employers answer the JOLTS questionnaire honestly, the job openings measured in JOLTS are unlikely to be ghost jobs,” it’s an admission that dishonesty is the assumed baseline for how employers interact with public-facing job platforms.

ZipRecruiter’s massive advertising spend makes sense in this context. The platform doesn’t need to work well to be profitable—it needs to maintain the appearance of working well enough to keep both sides of the marketplace engaged. The gap between the marketing and the user experience isn’t a bug. It’s the business model.

Tags:

No responses yet

Leave a Reply

Your email address will not be published. Required fields are marked *