A contact centre hiring 100 agents per month at 40% first-90-day attrition is losing 40 agents before they become productive. Each replacement costs R8,000–R15,000 in sourcing, assessment and training. That's R320,000–R600,000 per month — on hires that could have been screened out before training started.
Sourcing, background check, onboarding and the first three weeks of training — before the agent even reaches a live queue. For a 500-seat floor with 40% attrition, this is a fixed monthly cost that most ops teams have stopped questioning.
Industry-wide figure for South African contact centres. The primary cause is not candidate quality — it is the absence of predictive screening. CVs and unstructured interviews select for articulateness, not role readiness.
Structured, AI-scored assessments that measure the dimensions actually predictive of contact-centre performance: resilience under pressure, communication quality, CRM aptitude, compliance orientation. Before training. Before a rand is spent.
The two tools most employers rely on were designed in an era before we understood cognitive science, behavioural psychology, or predictive validity. They tell you what someone wants you to believe — not how they'll actually perform under real work conditions.
A CV proves someone held a job — not that they were good at it, not that they'll thrive in yours, not that they won't leave in 90 days. Tenure and titles are lagging indicators dressed as forward signals.
Skilled interviewers are rare. Most interviews are unstructured conversations that correlate weakly with job performance — and strongly with how well someone performs under low-stakes social pressure.
Structured assessments measure the dimensions that actually predict performance: how someone responds to pressure, communicates under stress, processes new information, and adapts their behaviour — before you spend a rand on training.
Every assessment we build is grounded in occupational psychology, predictive validity research, and real South African hiring outcome data. We don't ask candidates what they'd do. We put them in situations that reveal how they actually respond.
Behavioural dimensions — not personality types
We measure observable behaviours relevant to work performance, not pseudo-scientific personality categories.
Scenario-based questions — not self-report
Candidates respond to realistic situations drawn from the actual working environments they'll enter.
AI scoring — not human interpretation
Our models evaluate consistency, quality, and response patterns without interviewer bias or fatigue.
Outcomes-informed — not static weights
Employer feedback at 30, 60, and 90 days post-placement feeds back into the model and informs scoring calibration as the dataset grows.
Behavioural Dimensions
Resilience, Emotional Reg., Compliance, Communication, Ethics, Orientation, Retention Fit
Scoring Signals
Intelligibility, Pacing, Presence, Structure, Naturalness — for every voice recording
Competency Dimensions
Communication, Problem-Solving, EQ, Professionalism, Motivation, Adaptability (SVI)
Validation Checkpoints
30, 60, and 90-day post-placement performance ratings fed back into scoring models
Each layer builds on the last. Deploy just the first layer or stack the full journey — every layer delivers results in under 60 seconds, works on any mobile device, and requires zero candidate registration.
Eligibility & Suitability
Behavioural Intelligence
Voice & Communication
Skills & Systems
Role Simulation
Readiness & Recommendations
“How someone thinks under pressure — before you ever meet them.”
Our flagship L1 Screener measures 7 core dimensions that predict real-world work performance: resilience under rejection, emotional regulation, compliance instinct, communication clarity, target pressure & ethics, customer orientation, and retention fit. CVs can't tell you any of this. One 30-minute assessment can.
Sectors
What It Measures
Tap to Start
We'll walk you through it
Every campaign type has its own knowledge layer, scoring weights, and role-specific simulation. Financial services and insurance are contact-centre campaign types — not separate sectors.
Full five-layer journey — suitability, CRM simulation, voice readiness, customer service knowledge, and structured interview. South Africa's highest-volume contact-centre campaign type.
Sales aptitude, target pressure, and voice energy scored separately. International voice track (UK / USA) for cross-border campaigns. Custom weighting per client profile available.
Negotiation readiness, compliance awareness, and resilience under pressure weighted more heavily. Regulatory language and debt-collection conduct basics tested in the knowledge layer.
Digital literacy, logical troubleshooting, and system navigation scored alongside the core behavioural screener. CRM simulation uses a technical scenario variant.
Financial services and insurance are contact-centre campaign types — not separate industries. Assessed using the same core framework with domain-specific knowledge layers for each.
We build custom question banks, scoring profiles, and validation studies for campaign types not yet in our standard suite. Tell us your hiring profile — we'll build the science.
One scenario from our L1 behavioural screener. Select any answer — then see exactly how our scoring engine reads it and which dimension it signals.
Global occupational psychology firms charge R2,000–R8,000 per candidate for enterprise assessments not built for South African contact-centre role types. ShiftMate delivers structured, AI-scored, science-grounded assessments at a fraction of that cost — and the accuracy compounds with every candidate.
Every assessment scored feeds outcome data back into the models — 30, 60 and 90-day employer performance ratings feed back into the model to inform scoring calibration as the dataset grows. The longer the platform runs, the richer the outcome data underpinning each assessment decision. No new entrant can replicate this without years of SA-specific hiring outcome data.
Assessment is the acquisition surface. Once a BPO's standards are configured on ShiftMate, moving them is a compliance and continuity risk — not an inconvenience. Retention is structural. And assessment clients are the cohort most likely to adopt AI Co-pilot, Workspaces and Training Simulator.
Assessment-as-infrastructure
ShiftMate assessments are not a screening tool you buy once — they are the infrastructure every hiring decision in a contact centre runs through. Once a BPO's assessment standards are configured on the platform, moving them is a compliance and continuity risk. Retention is structural.
Data flywheel: every candidate improves prediction
Every scored candidate generates outcome data — 30, 60, and 90-day performance ratings from employers feed back into the model to inform deliberate calibration releases. Scoring weights are reviewed and updated as the hiring outcome dataset grows. No new entrant can replicate this without years of SA-specific hiring outcome data.
SA psychometric market is wide open
Global occupational psychology firms (SHL, Hogan, Korn Ferry) charge R2,000–R8,000 per candidate for enterprise assessments. ShiftMate delivers AI-scored, science-grounded assessments at a fraction of that cost — and built specifically for SA contact-centre role types. The addressable market is every BPO, bank, insurer and retailer hiring at volume.
Attrition reduction = measurable ROI
A contact centre with 40% first-90-day attrition, hiring 100 agents per month, spends R400,000–R600,000 per month on attrition-driven rehiring and retraining. A 10-point attrition improvement is R40,000–R60,000 per month in recoverable cost — for every 100 hires. That math makes the sale for us.
Modular architecture scales the business model
Assessment is layer one. Clients who start with assessments get AI Co-pilot, Custom Workspaces and Training Simulator at incremental cost. Each module deepens the data advantage and increases the switching cost. Assessment is the acquisition surface; the platform is the retention engine.