Applicant Tracking System Development: Cost, Features, and When Custom Beats Greenhouse

A staffing agency processing 500 candidates per month across 20 active roles pays Greenhouse $6,000-$30,000 per year for a system they do not own. A client asks for white-label branding. Greenhouse says no. The hiring workflow needs custom approval stages for contract roles versus permanent placements. The agency builds workarounds in spreadsheets. When they grow to 50 clients with different workflows, one shared Greenhouse instance becomes unmanageable.

That is the wall most agencies hit before they call us about applicant tracking system development.

A custom build is a one-time capital expense. Here is how the cost breaks down:

According to SHRM, the average cost-per-hire in the US is $4,700 and hiring top performers takes an average of 36 days. Time lost to manual handoffs between an ATS and an HRIS sits inside those numbers. Owning your platform removes that cost entirely.

Greenhouse, Lever, and Workday vs. custom ATS software

Before committing to applicant tracking system development, you need to know exactly where the off-the-shelf tools fail. Most companies do not hit those limits. The ones that do waste months trying to work around them.

Greenhouse is well built for in-house HR teams at single companies. It handles standard commercial hiring pipelines, structured interviews, and offer management. Pricing runs $6,000-$30,000 per year for mid-market and $50,000-$80,000 for enterprise. The failure points are specific: no white-label branding for agencies serving external clients, no per-client workflow isolation, and a data model that cannot express genuinely different approval chains per country. Greenhouse also does not offer per-placement pricing, which makes the economics wrong for agencies.

Lever sits in the $3,000-$8,000 per year range and is strong for candidate relationship management. It is better than Greenhouse at tracking passive candidates over time. The same limits apply: no white-label, no per-client isolation, no niche screening workflows. If your hiring process includes clinical credential verification, embedded coding assessments, or multi-decade executive relationship tracking, Lever cannot model it.

Workday Recruiting is the enterprise option. Pricing starts around $100,000 per year for large organizations and compounds with each module. It handles multi-country compliance and complex approval chains. But implementation takes 6-18 months and requires dedicated consultants. Companies that want to ship in 14 weeks with a system their team owns outright choose custom ATS development over a Workday rollout.

Custom wins in three specific situations. First, when you serve multiple clients who each need their own isolated workspace and branding. No off-the-shelf ATS offers this cleanly. Second, when your hiring workflow is a proprietary methodology you are selling, not just using. Competitors using Greenhouse cannot replicate what you own. Third, when compliance requirements per jurisdiction mean you need different offer templates, background check integrations, and approval chains per country. That is not a configuration problem. It requires a data model designed for it from the start.

"The biggest driver of poor candidate experience is process failure, not employer brand. Slow scheduling, missing calendar events, and inconsistent scorecard feedback account for 60% of candidate drop-off after the first interview." - LinkedIn, Global Talent Trends Report

Who actually builds a custom ATS

Applicant tracking system development is not for every company. Here are the four operator types where the economics clearly work.

Staffing agencies with 500+ candidates per month. An agency running recruiting across multiple client companies needs to manage candidates across different workflows, track placements, and show clients a branded platform. Greenhouse is built for a single company's in-house HR team. It is the wrong shape for an agency. At $18,000 per year for Greenhouse, a custom build at $120,000 reaches payback in licensing savings alone in about seven years. But the real case is not the licensing fee. It is charging clients for platform access, adding clients with near-zero marginal cost, and building workflow features competitors on Greenhouse cannot replicate.

HR tech founders targeting niche hiring segments. A startup building for technical hiring wants embedded coding assessments in the pipeline. An executive search firm needs relationship tracking across years of candidate touchpoints, not just active applicants. A healthcare staffing firm needs clinical credential verification as a required stage. These workflows are different enough from commercial hiring that a specialized tool commands a price premium and earns it. Greenhouse is not a starting point for this. It is a ceiling.

Large employers hiring across 10 or more countries. Different approval chains, different offer templates, different background check providers, and different compliance requirements per jurisdiction. Workday handles some of this but costs $100,000 per year and takes 18 months to implement. A custom build at $140,000 in 16 weeks with a team that owns the codebase is often a better option for mid-market companies that do not want to be on a Workday contract.

Recruitment process outsourcing firms with proprietary methodology. If your firm has spent years refining a screening and matching process that outperforms the market, running it on a generic ATS means competitors can observe and copy the surface layer. Owning the platform means the workflow itself is protected. The ATS is not a tool. It is the product.

V1, V2, and V3 features and what each phase costs

Separating features by phase is the most useful thing a build guide can do. Shipping everything in the first build is how teams end up with a 24-week project that runs 18 months.

V1 (8-10 weeks, $55,000-$80,000): what you need to start taking candidates

Configurable pipeline per job requisition. Every other feature depends on candidates moving through a defined stage sequence. Without a per-job pipeline, nothing else works correctly. This is the data foundation.

Candidate profiles with resume parsing. Applications arrive as PDF files. A parsed profile means the recruiter reviews structured data within minutes. At 500 applicants per month, resume parsing saves roughly 80 recruiter-hours per month.

Basic job posting with source tracking. Post to a careers page and track where each applicant came from. This data drives every recruiting budget decision you make in the second year.

Email notifications. Candidates need confirmation their application arrived. Interviewers need reminders 24 hours before their interview. Missing either is visible and damages the candidate's experience from day one.

V2 (weeks 11-18, additional $45,000-$80,000 on top of V1): once you have real hiring volume

Interview scheduling with calendar integration. This is the single most expensive feature in the build. Calendar API connections to Google Calendar and Outlook, availability polling, scheduling link generation, and calendar event creation for all participants add $25,000-$40,000 and 4-6 weeks. It is also the feature that produces the most visible failures when something goes wrong.

Structured scorecards. Google's Project Oxygen research found that structured interviews with consistent scoring rubrics predict job performance 2x better than unstructured conversations. Scorecards per pipeline stage, assigned at scheduling time and submitted within hours, aggregate for the hiring manager before the debrief.

Offer management with e-signature. Offer letter generation from templates, internal approval routing, and DocuSign or HelloSign integration for the candidate signature. Do not build native e-signature. The legal requirements for a valid electronic signature take 3-6 months to get right and candidates trust DocuSign more than a custom implementation.

HRIS integration. When a candidate accepts an offer, their data should push to BambooHR, Workday, or ADP automatically. An HR coordinator manually keying new hire data takes 20-30 minutes per hire. At 50 hires per year, that is 25 hours of manual data entry and a non-trivial error rate on day-one employee records.

V3 (triggered by scale, not a schedule): only build when the threshold arrives

AI resume scoring and candidate ranking ($30,000-$60,000). An AI model trained on past hires scores incoming resumes against job requirements. Useful at 1,000+ applications per month per role. Below that threshold, a recruiter can review applications faster than the model adds value.

Advanced reporting and analytics ($20,000-$30,000). Source attribution, interviewer load, offer acceptance rate, and time-to-hire by department. Build these when the hiring team is large enough to act on the data.

Multi-tenant white-label infrastructure ($30,000-$50,000). Subdomain routing, per-client branding, role isolation between client organizations. Only relevant for agencies and platforms serving multiple separate businesses. Building this before you have paying clients is money spent early.

Where applicant tracking system development projects fail

Underestimating interview scheduling. Most teams plan two weeks for it and discover the edge cases in production. The core requirement is connecting to both Google Calendar and Outlook to read interviewer availability. Both require OAuth authentication from each interviewer. Both return free/busy data, not event details. Availability queries cover the next two weeks of working hours. When a recruiter schedules, the system shows slots where all required interviewers are free simultaneously.

When the interview is rescheduled, every calendar event must update synchronously. Every participant gets a notification. The pipeline stage timestamp reflects the new time. We have seen rescheduling bugs leave candidates in confirmation-loop emails for 48 hours before anyone noticed. That kind of failure is visible and costly. Building this correctly costs $25,000-$40,000 and 4-6 weeks. Teams that cut this to two weeks find the real cost in production.

Building a global pipeline instead of a per-job configuration. Teams think "we can add per-job configuration later" and ship a global pipeline. Three months in, the engineering team is creating fake stages they do not use just to match the sales team's workflow. Rebuilding the pipeline data model mid-project costs $20,000-$30,000 and delays launch by 4-6 weeks. The per-job pipeline must be designed in week one. It is the decision that cannot be undone cheaply.

A third failure mode worth naming: deduplication logic built as an afterthought. A candidate who applies to three roles over two years should have one profile with a full history. Teams that build deduplication after launch discover thousands of duplicate profiles with no clean way to merge them. The data model for single-profile-multiple-applications must be designed before any application data is written.

How RaftLabs builds custom ATS platforms

We have built recruiting platforms and HR tech tools across multiple markets: staffing agencies, technical hiring platforms, and employers with multi-country workflows. The pattern across these builds is consistent. Interview scheduling takes longer than teams expect. The pipeline data model is the decision that cannot be undone.

The first thing we do is define the pipeline configuration model before any code is written. The second is design the calendar integration state machine, including all rescheduling scenarios. Both decisions save 6-10 weeks of rework later.

For companies where SaaS application development is the goal and the ATS is the core product, the architecture decisions above apply directly. For companies that need to validate a market before committing to a full build, MVP development with a focused feature set gets you to market in 8-10 weeks at roughly half the full build cost.

If you are scoping a custom ATS, tell us your candidate volume, your current hiring workflow, and the specific gaps your current tool cannot handle. We will give you a realistic cost and timeline in 48 hours. If an off-the-shelf tool is the right answer for your situation, we will tell you that too.

Book the 30-minute scoping call