Case Study: 0 to 120 Agents in 90 Days — Launching a BPO from Scratch

Greenfield BPO launches are the hardest staffing engagements we run. There is no employer brand, no candidate familiarity with the company name, no existing tenured population to refer in friends, and the operations team is usually fighting an IT punchlist and a lease build-out at the same time. The recruiting work has to start while everything else is still being assembled.

This is a real engagement from 2025, anonymized for confidentiality. The client name and a couple of non-load-bearing details have been changed; the timeline, the playbook, and the outcomes are real. We are publishing it because BPO launches are a recurring engagement type and the pattern below is what reliably works.

The client

Beacon Outsourcing, a multi-client fintech BPO, was standing up its first US-fintech contract delivering customer support, fraud-flag review, and account-recovery handling for a US neobank. The contract had been signed three months earlier; the launch date for first SLA delivery was 90 days out when we were brought in.

Beacon had run prior BPO programs in Manila for non-fintech clients, but this was their first US-financial-services contract and the regulatory and quality bar was higher than their existing operations. They needed 120 agents — across customer support and fraud lanes — on a Manila-and-Cebu split, hitting first-week SLA inside the contractual ramp window.

The constraint stack

When we engaged, the constraints were stacked:

• Lease was signed on the Manila floor; build-out was complete but IT punchlist (badge access, agent workstations, CRM provisioning, call-recording config) was still open at week -4.
• Cebu floor was sub-leased space inside an existing partner facility, available from week -2.
• No employer brand for the new fintech program — Beacon's existing reputation was in non-financial work, and candidates who saw "fintech" on the listing did not associate it with Beacon.
• The client (the US neobank) had specified a fluency and quality bar above what Beacon's prior programs had run, with a specific written QA rubric and CEFR floor.
• Compliance overlay: candidates needed to clear pre-employment screens including financial-services-specific background checks and a US sanctions-list verification, which Beacon had not run before.
• 90 days from kickoff to first SLA hit. The contract had penalty clauses for missed ramp.

The playbook

The engagement ran on three principles, all running in parallel from week 1:

• Pre-built bench pull from Manila and Cebu before badging was even possible. We had a standing fintech-aligned bench in both cities (built up across other engagements) — voice-screened, CEFR-graded, and recently checked-in. Rather than wait for the IT punchlist to close, we started screening and shortlisting against the role spec from day one.
• Embedded recruiters with the Beacon ops team, on the same Slack and the same standups. The screening rubric was calibrated against the US neobank's QA scorecard before week 1 and recalibrated between waves based on early data.
• Cohort waves of 20, starting week 4 and continuing every 2 weeks. The wave cadence was designed around the IT readiness curve — wave 1 starting the moment 20 workstations were live, with subsequent waves staged against IT punchlist completion.

For the compliance overlay, we ran financial-services background and sanctions-list verification at candidate submission rather than at offer — which surfaced 4 disqualifications inside the first 60 candidates that would otherwise have shown up at offer-time and forced rework. /services/bpo-recruitment walks through how we structure compliance-at-submission for financial-services BPOs.

Week-by-week timeline

A reconstruction of the 90-day engagement:

• Weeks 1 to 2: kickoff, rubric calibration with US neobank QA team, bench pull from Manila and Cebu, first 40 candidates submitted with voice samples and CEFR scores. Compliance verification begun at submission.
• Weeks 3 to 4: interview loops running; offers extended for wave 1 (target: 25 offers to land 20 starts); IT punchlist closing on Manila floor for first 20 workstations.
• Week 4: wave 1 (20 agents) starts training in Manila.
• Weeks 5 to 6: sourcing for waves 2 and 3 in parallel; second-round interviews; embedded recruiter on daily standups with Beacon ops; rubric recalibrated based on wave 1 day-3 training observations.
• Week 6: wave 2 (20 agents) starts training. Cebu floor accessible; first wave 3 sourcing pivoted to Cebu pool.
• Weeks 7 to 8: continued sourcing waves 3, 4 and 5; wave 1 completing training, entering nesting on first live calls under supervision.
• Week 8: wave 3 (20 agents) starts training, split Manila/Cebu.
• Weeks 9 to 10: waves 1 and 2 nesting; wave 1 hitting first AHT and CSAT benchmarks against the US neobank's scorecard. Wave 4 starts training (week 10).
• Weeks 11 to 12: wave 1 productive on the floor against full SLA; wave 5 (final 20 agents) starts training.
• Week 13 (day 90): first contractual SLA window. Wave 1 hitting full SLA; waves 2 and 3 hitting nesting-stage benchmarks; full 120-agent footprint deployed across training, nesting, and floor.

The outcomes

The engagement closed inside the contractual ramp window with the headcount and the SLAs both hit:

• Headcount: 120 of 120 placed across 5 cohort waves over 12 weeks. Final wave landed week 11; 90-day mark had full footprint deployed.
• First contractual SLA window hit by wave 1 inside the ramp window. The penalty clause never triggered.
• 90-day attrition across all 5 waves: 11 percent. Industry average for greenfield fintech BPO launches typically runs 20 to 25 percent at 90 days. The lift came primarily from compliance-at-submission (eliminating offer-stage rescissions) and from CEFR-rubric-aligned screening.
• CSAT against US neobank scorecard: hitting program target by wave 2, exceeding by wave 3. Wave 1 ran below target through nesting, calibrated up by week 12 with rubric and coaching adjustments.
• Quality of hire: validated by the US neobank in their first quarterly business review with Beacon, which concluded with the contract being expanded by an additional 30 seats for Q3.

David Cho, COO of Beacon Outsourcing, summarized the engagement: "Their recruiters embedded with our ops team and we hit our first SLA inside the ramp window." The first-SLA-inside-ramp-window outcome was the contractual milestone that mattered most; everything else was downstream of that.

Lessons for future BPO launches

Three things we have generalized from this and similar engagements:

• The pre-built bench is the launch enabler, period. Without a standing pool in the geography that fits the role spec, the timeline cannot compress. Operators planning a launch should be evaluating partners on bench depth and freshness, not on placement-fee math.
• Cohort waves staged against IT readiness, not against a recruiting calendar, prevent the "we have offers but no workstations" mess that kills most launch timelines. The recruiting and IT critical paths must be on the same Gantt chart.
• Compliance-at-submission for financial-services and healthcare BPO launches is not optional. The cost of late-stage offer rescission compounds in a launch where every cohort feeds the next, and 4 to 8 percent of candidates get knocked out at compliance even with strong screening upstream. Pull it forward.

Operators planning a BPO launch in Manila, Cebu, Mexico, or any other delivery geography: the playbook above is repeatable. /locations walks through the regions we recruit in and what the bench looks like for English-language work. /solutions/launching-new-call-center walks through how we typically scope greenfield engagements end-to-end. Other launch case studies are at /case-studies.