AI recruitment agent cost tracking is the practice of measuring what AI screening, outreach, and matching agents cost per candidate, per placement, and per client — then using that data to calculate ROI and price recruitment services accurately.
AI recruitment agents can automate up to 90% of high-volume hiring workflows. A screening agent reviews a CV in seconds for £0.02–£0.10. An outreach agent sends personalised messages for £0.05–£0.30 per candidate. A matching agent evaluates job-candidate fit for £0.05–£0.20 per evaluation. These costs are tiny per candidate. But multiply them across thousands of candidates per month and they become a significant line item.
Without per-candidate and per-placement cost tracking, agencies cannot calculate agent ROI, price AI-assisted services, or prove the value of their technology investment to clients.
Key Takeaway: Track recruitment AI costs per candidate and per placement. The per-unit cost is small but scales fast.
Where Do Recruitment Agencies Use AI Agents?
AI agents now participate in every stage of the recruitment pipeline. Each stage has different cost characteristics.
Screening Agents
Screening agents are the highest-volume recruitment AI tool. They parse CVs, extract qualifications, match requirements against job specifications, and produce shortlists. A screening agent can review 1,000 CVs in the time a human recruiter reviews 20.
Typical actions: parse CV structure, extract skills and experience, compare against job requirements, score fit on a weighted scale, flag qualified candidates, reject unqualified ones.
Cost driver: token consumption for reading and analysing each CV. Simple CVs cost less to process than detailed ones.
Outreach Agents
Outreach agents craft personalised messages to candidates, manage follow-up sequences, and handle initial responses. They personalise each message based on the candidate’s profile, the role, and the client’s value proposition.
Typical actions: generate personalised first-touch message, send follow-up sequence, classify candidate responses (interested, not interested, questions), escalate warm leads to human recruiters.
Cost driver: message generation tokens plus any research tokens spent personalising the message (checking the candidate’s professional profile, recent activity, publications).
Matching Agents
Matching agents go beyond screening. They evaluate the depth of fit between candidates and roles — considering career trajectory, cultural signals, skill gaps, and growth potential. They are more expensive per evaluation than screening agents because they require deeper analysis.
Typical actions: multi-dimensional fit scoring, skill gap analysis, career trajectory assessment, compensation benchmark comparison, team fit evaluation.
Cost driver: reasoning complexity. Matching requires more inference compute than simple screening.
Interview Scheduling Agents
Scheduling agents coordinate availability between candidates, interviewers, and hiring managers. They handle confirmations, reminders, and rescheduling. These are the cheapest agents in the recruitment stack.
Voice Agents
Voice agents conduct initial phone screens — pre-qualification calls that verify availability, salary expectations, notice period, and basic eligibility. They operate via voice synthesis and speech recognition, which adds telephony costs on top of token costs.
Cost driver: call duration plus speech processing. A 5-minute voice screening call costs significantly more than a text-based screening.
Analytics Agents
Analytics agents generate pipeline reports, track time-to-fill metrics, analyse source effectiveness, and identify bottlenecks. They pull data from ATS systems and produce dashboards and summaries.
What Do Recruitment Agents Cost?
The table below shows cost benchmarks for recruitment AI agents in 2026.
| Agent Type | Action | Cost Per Action | Typical Volume/Month | Monthly Cost |
|---|---|---|---|---|
| Screening | CV review and score | £0.02–£0.10 | 2,000–10,000 | £40–£1,000 |
| Outreach | Personalised message | £0.05–£0.30 | 500–3,000 | £25–£900 |
| Matching | Deep fit evaluation | £0.05–£0.20 | 200–1,000 | £10–£200 |
| Scheduling | Interview coordination | £0.01–£0.05 | 100–500 | £1–£25 |
| Voice | Phone screen (5 min) | £0.50–£3.00 | 50–200 | £25–£600 |
| Analytics | Pipeline report | £1.00–£5.00 | 10–30 | £10–£150 |
Total monthly AI cost for a mid-sized recruitment agency: £110–£2,875
The wide range reflects differences in agency volume. A boutique agency placing 10 roles per month sits at the low end. A high-volume staffing firm processing 10,000+ candidates per month sits at the high end.
The Unit Economics Are Compelling
Even at the high end, the per-candidate AI cost is a fraction of human recruiter cost. Consider a high-volume scenario:
- 5,000 candidates screened at £0.05 each: £250
- 1,000 outreach messages at £0.15 each: £150
- 300 match evaluations at £0.10 each: £30
- 100 interview schedules at £0.03 each: £3
- 50 voice screens at £1.50 each: £75
- Total AI cost: £508
- Placements made: 25
- AI cost per placement: £20.32
A human recruiter spending 40 hours to make those 25 placements at £30/hour costs £1,200 in time alone — plus the recruiter handled far fewer candidates.
For guidance on tracking AI agent time alongside costs, see our AI agent time tracking guide.
How Do You Measure Cost Per Candidate vs Cost Per Placement?
These are two different metrics and both matter.
Cost Per Candidate Screened
This is the most granular metric. Divide total screening agent costs by the number of candidates processed.
Formula: Cost per candidate = total screening costs / candidates screened
This metric is useful for:
- Comparing AI screening cost vs human screening cost per candidate
- Benchmarking across different screening agent configurations
- Identifying whether agent costs are rising (model changes, prompt complexity)
Typical range: £0.02–£0.15 per candidate (screening only).
Cost Per Placement
This is the business metric. Divide total AI costs across the entire pipeline by the number of successful placements.
Formula: Cost per placement = total AI pipeline costs / successful placements
This metric tells you the true AI cost of each placement — the number you need for pricing and ROI calculations.
The Funnel Effect
The relationship between cost per candidate and cost per placement depends on your funnel conversion rate.
| Hiring Type | Candidates Screened | Placements | Conversion | AI Cost Per Placement |
|---|---|---|---|---|
| High-volume retail | 500 | 1 | 0.2% | £15–£30 |
| Mid-volume professional | 200 | 1 | 0.5% | £20–£40 |
| Specialist/executive | 50 | 1 | 2% | £25–£50 |
High-volume roles screen more candidates per placement but use cheaper per-candidate screening (simpler evaluation). Specialist roles screen fewer candidates but spend more per screen (deeper analysis, matching, and multiple rounds).
Tracking by Client
Different clients have different funnel shapes. A retail client fills 50 roles per month with high-volume screening. A technology client fills 3 specialist roles with deep matching. Per-placement AI costs differ significantly.
Track costs by client to ensure pricing reflects actual AI consumption.
For the foundational guide on AI cost per task, see our AI agent cost per task guide.
What Is the ROI of AI Recruitment Agents?
ROI calculation requires comparing the time saved against the agent cost.
The ROI Formula
ROI = (human time saved x hourly rate - agent cost) / agent cost x 100
Worked Example: High-Volume Staffing
A recruitment agency processing 5,000 candidates per month for 25 placements.
Without AI agents:
- 3 recruiters × 160 hours/month × £25/hour = £12,000/month
- Placements: 25
- Human cost per placement: £480
With AI agents:
- AI agent costs: £508/month
- 1 recruiter (handling exceptions and relationships) × 160 hours × £25/hour = £4,000/month
- Total cost: £4,508/month
- Cost per placement: £180.32
Savings: £7,492/month. ROI: 1,475%.
The AI agents did not just reduce cost. They freed two recruiters to handle additional requisitions, increasing the agency’s capacity without adding headcount.
Worked Example: Specialist Recruitment
A boutique agency placing 8 specialist roles per month.
Without AI agents:
- 2 recruiters × 160 hours/month × £35/hour = £11,200/month
- Placements: 8
- Human cost per placement: £1,400
With AI agents:
- AI agent costs: £280/month
- 2 recruiters still needed (relationship-driven process) × 120 hours × £35/hour = £8,400/month
- Total cost: £8,680/month
- Cost per placement: £1,085
Savings: £2,520/month. ROI: 900%.
The ROI is lower for specialist recruitment because human involvement remains high. Agents handle screening and outreach but recruiters still manage relationships, negotiations, and closing.
When ROI Turns Negative
AI recruitment agents lose money when:
- False-positive rates are high: If the screening agent passes 200 unqualified candidates, human recruiters waste time reviewing them. The agent cost is £10 but the wasted recruiter time is £500.
- Role complexity exceeds agent capability: For C-suite or highly specialised roles, agent screening may be counterproductive. Human judgement is required from the start.
- Candidate experience suffers: If automated outreach damages the agency’s reputation with candidates, the long-term cost exceeds the short-term saving.
Monitor these risks alongside cost metrics.
How Do You Integrate AI Costs With ATS and CRM Billing?
AI cost data is most useful when it lives alongside candidate and client data in your existing systems.
Push Costs Into Your ATS
Add AI cost as metadata on each candidate record. When a recruiter opens a candidate profile, they see the AI cost incurred — screening cost, outreach cost, matching cost. This makes cost tracking visible without requiring a separate dashboard.
Implementation: most modern ATS platforms support custom fields via API. Push agent cost data into these fields after each agent action.
Roll Up by Job Requisition
Aggregate candidate-level AI costs by job requisition. This gives you the total AI cost per role, which feeds into client billing.
A job requisition that screened 300 candidates, sent 80 outreach messages, and scheduled 12 interviews has a total AI cost that is the sum of all those actions.
Build AI Cost Into Placement Fees
Two approaches for pricing:
Absorb and profit. If AI saves £300 per placement in recruiter time but costs £20 in agent fees, keep your fees the same and improve margins by £280 per placement.
Pass through and discount. Show clients the AI efficiency gain and offer a modest fee reduction. This builds client loyalty and positions the agency as technology-forward.
Hybrid. Absorb agent costs but offer clients faster time-to-fill — which has its own value.
Client Reporting
Show clients:
- Number of candidates screened (AI-powered volume that manual processes could not match)
- Time-to-fill improvement
- Quality metrics (percentage of shortlisted candidates who reach interview stage)
- Optional: AI cost per placement, demonstrating operational efficiency
For the broader framework on tracking AI costs in professional services, see our AI agent cost tracking guide.
Frequently Asked Questions
How much do AI recruitment agents cost per candidate?
CV screening agents cost £0.02–£0.10 per candidate. Outreach agents cost £0.05–£0.30 per personalised message. Matching agents cost £0.05–£0.20 per evaluation. Voice screening agents cost £0.50–£3.00 per call. Total AI cost per candidate through the full pipeline ranges from £0.10 to £3.50 depending on how many stages the candidate passes through.
What is the cost per placement for AI recruitment agents?
The total AI cost per placement typically ranges from £15 to £50. High-volume roles cost less per placement (simpler screening, more candidates processed). Specialist roles cost more per placement (deeper analysis, voice screening, multi-stage matching). These figures include all AI agent costs across the pipeline.
How do you calculate ROI for AI recruitment agents?
ROI = (human time saved x hourly rate - agent cost) / agent cost x 100. For high-volume staffing, typical ROI is 1,000–2,000%. For specialist recruitment, typical ROI is 200–500%. The ROI includes both direct cost savings and capacity gains — agents let recruiters handle 3–5x more requisitions simultaneously.
What AI agents do recruitment agencies use?
Six main types: screening agents (CV parsing and qualification matching), outreach agents (personalised candidate messaging), matching agents (deep fit evaluation), scheduling agents (interview coordination), voice agents (phone pre-screening), and analytics agents (pipeline reporting and metric tracking). Most agencies start with screening and outreach.
How do you integrate AI costs with ATS billing?
Push AI cost data into your ATS as candidate-level metadata via API. Aggregate by job requisition for per-role cost tracking. Roll up to client accounts for billing. Most modern ATS platforms support custom fields that can hold agent cost data. Build AI cost into placement fee structures — either absorb for margin improvement or pass through with discounts.
Keito tracks AI recruitment agent costs per candidate and placement, integrated with ATS and billing systems automatically. Start tracking recruitment AI costs.