A specialized recruiting firm built a full-stack recruitment automation system on Make.com and documented a 207% ROI at 18 months. The return came from four sources: sourcing labor reduction, screening time elimination, scheduling automation, and candidate nurturing pipeline reactivation — each measured against pre-automation baselines with documented before-and-after metrics.

ROI claims in recruitment technology are easy to make and hard to verify. This case study documents actual numbers — the specific automations built, the labor time eliminated, the revenue generated from pipeline improvements, and the methodology used to calculate the 207% return. The goal is a model you can apply to your own stack to project realistic returns before committing to a build.

The automation infrastructure behind these results is the same stack covered in the Make.com HR Integrations to Automate Workflows — Complete 2026 Guide. This implementation covered every major automation layer described there, built over 18 months in sequenced phases.

Company Profile

The firm is a mid-sized recruiting operation specializing in finance, accounting, and operations roles across the mid-market. At project start, 12 full-time recruiters handled approximately 180–200 active requisitions per month. The core operational problem was scale: demand from existing clients was growing, but the firm was hitting a ceiling on requisitions-per-recruiter. Adding a recruiter carried substantial fully-loaded annual compensation before that recruiter reached full productivity at 12–18 months. The question leadership asked was whether automation could extend existing recruiter capacity before adding headcount.

Phase 1: Sourcing Automation (Months 1–4)

What Was Built

A Make.com-based sourcing system accepted structured requisition parameters from the ATS, executed searches against Apollo’s API, enriched returned profiles, scored candidates against qualification criteria, and created ATS records for qualified candidates — all without recruiter involvement until the qualified list was ready for review.

Pre-Automation Baseline

Each recruiter spent an average of 12 hours per week on sourcing tasks: running searches, reviewing profiles, enriching contact data, and adding qualified candidates to the ATS. Across 12 recruiters, sourcing consumed the single largest share of total weekly recruiter capacity.

Post-Automation Results

Recruiter sourcing time dropped to an average of 2.5 hours per week — reviewing the qualified lists produced by the automation and making judgment calls on edge cases. The 9.5 hours per recruiter per week reclaimed translated to a 79% reduction in sourcing time across the team.

Qualified candidate volume increased 42% because the automation ran searches continuously rather than only when a recruiter had time to run them. More qualified candidates in the pipeline earlier meant more active placements per recruiter.

Contribution to Total ROI

The savings calculation focused on the revenue impact of reclaimed time, not labor cost reduction in isolation. The 9.5 hours per recruiter per week reclaimed from sourcing were redirected to revenue-generating activities: candidate relationship management, client development, and offer negotiation. Incremental placements generated from increased recruiter capacity were the largest single contributor to the documented annual return — approximately 60% of the total.

Phase 2: Screening and Scheduling Automation (Months 3–7)

What Was Built

AI-assisted first-pass screening scored inbound applications against role requirements and ranked candidates before any recruiter review. Automated interview scheduling triggered calendar invitations and handled rescheduling without recruiter coordination. Automated candidate status communications fired at every stage transition.

Pre-Automation Baseline

First-pass screening consumed an average of 6 hours per recruiter per week. Interview scheduling consumed another 4 hours per recruiter per week — primarily back-and-forth coordination between candidates and hiring managers. Status communications were largely inconsistent, resulting in candidate drop-off and reputation damage.

Post-Automation Results

Screening time dropped to under 1 hour per week per recruiter. Scheduling time dropped to near zero for standard interview types; complex scheduling (multi-interviewer panels, mixed-format interviews) still required 30–60 minutes of recruiter involvement per role.

Candidate drop-off between application and first screen dropped 28% because the automated acknowledgment and status communication sequence kept candidates engaged. Higher retention translated directly to a larger qualified pool available at the offer stage.

Contribution to Total ROI

The 8.5 hours per recruiter per week reclaimed from screening and scheduling were redirected to offer stage management and candidate preparation — activities that had received less recruiter attention due to time constraints. Improved offer stage management contributed to a 7-percentage-point increase in offer acceptance rate, representing roughly 22% of the total documented annual return at this firm’s placement volume.

Phase 3: Candidate Nurturing Pipeline (Months 6–12)

What Was Built

A Keap-based candidate nurturing system connected to the ATS via Make.com. Silver medalists and declined-offer candidates entered segmented nurture sequences automatically when their ATS stage changed. Behavioral triggers in Keap flagged warm candidates — those engaging with nurture content — for recruiter outreach. The system ran without manual recruiter input once a candidate entered the pipeline.

Pre-Automation Baseline

There was no formal candidate nurturing system. Silver medalists received a manual keep-in-touch call from the placing recruiter approximately 30% of the time. Declined-offer candidates were essentially lost to the pipeline. The firm estimated it was losing 15–20 qualified, already-vetted candidates per month to this gap.

Post-Automation Results

At 12 months of operation, 19% of placements originated from the nurture pipeline — candidates who had been in a Keap sequence before the role that placed them opened. The average time-to-fill for nurture pipeline candidates was 8 days shorter than for cold-sourced candidates, because the relationship and qualification work was already done.

Contribution to Total ROI

At full placement volume, 19% of placements from the nurture pipeline represented approximately 14–15 additional placements per month that either would not have happened or would have required full cold-sourcing cycles. The nurture system contributed roughly 18% of the total documented annual return — a conservative estimate of incremental placement revenue from pipeline reactivation versus sourcing those same candidates from scratch.

The Full ROI Calculation

The 207% ROI figure used a straightforward methodology:

Total investment: Build cost (external development plus internal time) plus first-year platform costs (Make.com, Apollo API, Keap) plus ongoing maintenance estimate.

Total return: Incremental revenue from sourcing capacity reclaimed (approximately 60% of total) plus incremental revenue from screening and scheduling capacity and improved offer acceptance (approximately 22%) plus incremental revenue from nurture pipeline placements (approximately 18%) equals the total documented return across all three phases.

ROI: (Total return minus total investment) divided by total investment equals 207%.

The leadership team validated each component against actual placement data — not projections. Every figure included was from automations that were operational and measured for at least six months before inclusion in the ROI calculation.

What Didn’t Work (And What Was Fixed)

Three setbacks from this build are worth documenting for anyone planning a similar project:

Sourcing deduplication gap: The initial sourcing automation did not check for existing ATS contacts before creating new records. By month 3, the ATS had approximately 800 duplicate records. Fixing this required building the deduplication check into the scenario (a 2-day build) and a one-time cleanup pass on existing data (a 3-day project). Build deduplication on day one.

Nurture sequence unsubscribe rate: The initial silver medalist sequence had a 4.2% unsubscribe rate by month 2. Audit revealed the sequence was sending too frequently (weekly) and content was not differentiated enough by role type. Frequency dropped to bi-weekly and content was segmented by role family. Unsubscribe rate fell to 1.1%.

API instability in the scheduling integration: One benefits vendor used in the scheduling integration had an undocumented rate limit causing intermittent failures. The error handler was catching failures, but the retry volume was creating a queue backlog during peak periods. The fix: exponential backoff retry logic plus a daily reconciliation check to catch records that failed out of the retry queue.

Each problem was solvable. None required rebuilding the system from scratch. Build error handling and reconciliation checks into the initial design — not as an afterthought after the first failure.

Applying This Model to Your Firm

The numbers in this case study are specific to one firm’s placement fees, recruiter compensation, and requisition volume. The methodology — calculate pre-automation labor cost per task, measure post-automation time, quantify reclaimed capacity in revenue terms — applies to any recruiting firm regardless of size.

Start with the baseline calculation: how many hours per recruiter per week are spent on sourcing, screening, and scheduling? What is your average placement fee contribution margin? What would a 10–15% increase in placement volume mean at your current fee levels? Those three inputs tell you whether the build economics work before you commit to the project.

Frequently Asked Questions

What is a realistic ROI for recruitment automation?

ROI varies based on placement fee levels, recruiter compensation, and requisition volume. The firm in this case achieved 207% ROI at 18 months. Smaller firms with lower fee volumes should expect longer payback periods but the same directional ROI if automation frees meaningful recruiter time for revenue-generating activities. Run the baseline calculation — hours saved multiplied by recruiter hourly cost, plus incremental revenue from reclaimed capacity — against your own numbers before projecting ROI.

How long does it take to build a full recruitment automation stack?

A full three-phase build takes 18 months when executed sequentially. Phase 1 (sourcing automation) is operational in 4 months. Phase 2 (screening and scheduling) adds 3–4 months. Phase 3 (nurture pipeline) adds another 4–6 months to reach full operation. Teams that attempt all phases simultaneously extend total project time because testing and calibration requirements compound.

What percentage of placements should come from a nurture pipeline?

The firm in this case reached 19% of placements from their nurture pipeline at 12 months of operation. Firms with mature nurture programs range from 15–25%. Reaching these levels requires 8–12 months of pipeline build-up — nurture candidates do not convert in the first 60 days, so the ROI compounds over time rather than appearing immediately.

Does recruitment automation reduce the need for recruiters?

In this case, automation eliminated the need to add headcount to handle growth — the firm handled significantly higher requisition volume with the same recruiter count. It did not reduce existing headcount. Reclaimed time was redirected to relationship management, client development, and offer negotiation — activities where recruiter judgment creates direct revenue value.

What is the biggest mistake in recruitment automation ROI calculations?

The most common mistake is calculating ROI based on projected labor savings alone without accounting for what recruiters do with the time they reclaim. If reclaimed hours go to administrative tasks rather than revenue-generating activities, the ROI falls below the labor cost calculation. The correct methodology explicitly redirects reclaimed time to specific high-value activities and measures the revenue impact — not just the hours saved.