- Blog
- Screening
Screening
High-volume hiring: how to plan capacity and run the process
Alfie Whattam, Co-founder and CEO, Alfa AI ·

What is high-volume hiring?
High-volume hiring is recruiting many people for the same or similar roles in a short period, so applicant numbers outrun the team's time to read, reply to and interview each person individually. There is no fixed headcount threshold: the openings, applicant load and deadline determine whether your team needs a different process.
I'd start with capacity, before buying tools or increasing advertising. Work out what the team must process, then decide which tasks to automate and which decisions need a person.
How high-volume hiring differs from single-role hiring
| Aspect | Single-role hiring | High-volume hiring |
|---|---|---|
| Roles | A specific vacancy | Many openings with shared requirements |
| Applicants per opening | Plan around the vacancy's applicant pool | Model the pool needed across all openings |
| The bottleneck | Moving individual candidates forward | Keeping screening, replies and interviews within capacity |
| Where quality is decided | Individual evidence and assessment | Shared rules, consistent assessment and exception handling |
| What goes wrong first | A candidate's progress stalls | A queue builds and candidates lose contact |
Keep accountability at campaign level: someone must own the whole journey, including candidates waiting between stages.
Build a high-volume hiring plan backwards from the start date
Use rates from comparable past campaigns, with the same stage definitions. Where evidence is missing, label the assumption and update it as applications move through.
Hires needed (H) = ___
Weeks until start date (W) = ___
Offer acceptance rate, as a fraction (A) = ___
Offers per interview, as a fraction (O) = ___
Invited candidates who attend the interview, as a fraction (T) = ___
Applicants who pass screening, as a fraction (S) = ___
Minutes per first-round interview (M) = ___
Offers needed = H / A
First-round interviews needed = H / (A * O)
Candidates to invite = H / (A * O * T)
Applicants needed = H / (A * O * T * S)
Interviewer hours = (H / (A * O)) * M / 60
Applicants needed per week = (H / (A * O * T * S)) / W
Example assumptions, replace these with your own past data: a contact center needs 12 customer service advisers in 4 weeks. Assume 3 in 4 offers are accepted, 1 in 2 interviews leads to an offer, 2 in 3 invited candidates attend, and 1 in 2 applicants passes screening.
It needs 16 offers, 32 first-round interviews, 48 invitations and 96 applicants, about 24 applicants a week. Those 32 interviews of 30 minutes take 16 interviewer hours.
Round candidate requirements up to whole people. Interviewer hours cover interviews held; add time for evidence review, scheduling, missed appointments and offers. If you add a later interview stage, budget its capacity separately.
Set the last viable dates for interviews, offers and pre-start checks. Track withdrawals before the start date separately and update the plan when they happen.
If applicants are short, revise your candidate sourcing plan around the specific gap. If processing capacity is short, assign more reviewers or consider recruitment process outsourcing for the hiring peak, with clear ownership of decisions and messages.
Write screening rules that hold at volume
An application knockout asks directly about an essential condition. A hard resume requirement checks evidence of a non-negotiable qualification. A scored signal helps rank candidates without excluding them.
Each knockout should cover something the law or job truly requires. Preferences belong in scored signals. Use this warehouse operative rule table as a starting point, subject to the actual duties and available adjustments.
| Rule | Type | How to check it | Why |
|---|---|---|---|
| Right to work in the country | Knockout application question | Ask about work authorization and sponsorship needs; review unclear answers | Establish whether the employment can proceed |
| Able to work the advertised shift pattern | Knockout application question | State the shifts; ask about availability and review adjustment requests | Resumes rarely mention availability |
| Forklift license where the role needs one | Knockout application question | Ask whether the applicant holds the required license; verify it before appointment | Check a qualification the actual duties require |
| Previous warehouse, logistics or other physical work | Scored signal, not a requirement | Assess relevant tasks and transferable experience | Recognize useful evidence without excluding newcomers |
| Experience with handheld scanners or a warehouse management system | Scored signal | Look for examples in the resume or application | Inform ranking without making familiarity essential |
| A gap in employment | Do not use as a rule | Leave it out of screening criteria | Can screen out carers, parents and people with a disability, risking indirect discrimination |
That employment-gap filter risks indirect discrimination under UK law. Assess the ability to do the work instead.
Do not ask a resume to prove future availability, shifts, travel or equipment. Put necessary questions on the application form and route ambiguous answers for review. Document acceptable evidence and exceptions using a screening rule template.
Replace the first phone screen with a structured interview
The same core questions and the same interview scorecard make interviews comparable at volume. Define what a strong answer demonstrates before candidates start, and keep the criteria consistent when providing adjustments.
| Format | Who asks the questions | When the candidate takes it | Best for |
|---|---|---|---|
| AI interview | An AI interviewer | On demand | Screening conversations without live interviewer scheduling |
| One-way video interview | Prerecorded or written prompts | On demand | A fixed question set with recorded answers for review |
| Structured phone screen | A recruiter | At an agreed time | Candidates needing a live exchange or an alternative format |
For a warehouse role, ask: “What would you do if a scanner count did not match the items?” Score the evidence of checking the discrepancy, following the stated process and seeking help when needed. Avoid treating confidence or polished delivery as evidence of warehouse ability.
Write descriptions of clear, partial and insufficient evidence, then have reviewers compare how they apply them. Use the AI interviewer setup guide to prepare questions and candidate instructions. Assign a reviewer to each completed interview and record the evidence supporting the next step.
Keep every candidate informed
Measure screening response times from application and post-interview replies from interview completion. Treat these timings as internal service commitments. Assign an owner to failed messages and candidates who cannot use the booking or interview format.
| Moment | What the candidate hears | Sent by | Send by |
|---|---|---|---|
| Application received | Automatic confirmation | Recruiting system | Same day |
| Screening outcome for candidates moving on | Invitation to the first-round interview | Recruiting system after approval | Within 2 working days |
| Screening outcome for candidates not moving on | A short, polite decline after a person has reviewed the decision | Recruiter or approved automated message | Within 5 working days |
| Interview invitation | Booking link with the times on offer and the adjustments question | Interview coordinator or recruiting system | With the invitation |
| After the interview | Next step or decline | Named recruiter | Within 3 working days |
| Offer | Call, then written offer | Hiring manager or recruiter | Same day as the decision |
For an on-demand interview, replace the booking instructions with access instructions. If a decision slips, send a holding message with the next update point rather than leaving the applicant waiting without an explanation.
Check fairness before you scale
Run an adverse impact check at each automated stage
The US Uniform Guidelines' four-fifths rule generally treats a selection rate for any race, sex or ethnic group below four-fifths, or 80 percent, of the highest group's rate as evidence of adverse impact.
Stage checked (for example, passed screening) = ___
For each group:
Applicants at the stage (N) = ___
Applicants who passed (P) = ___
Selection rate (R) = P / N
Highest selection rate (Rmax) = ___
Impact ratio for each group = R / Rmax
Flag any impact ratio below 0.8 for review
Run this check weekly on each automated stage: screening, knockout questions and first-round interviews. Compare people entering the same stage, job and reporting period. Use appropriate monitoring data; do not infer protected characteristics from names or resumes.
A low ratio prompts review of the rule behind it; it is not proof of discrimination. Small groups need care, and clearing the threshold is not a guarantee: the same guideline recognizes smaller differences that are statistically and practically significant. Inspect the underlying counts and individual decisions before changing a rule.
This is a US guideline. In the UK, section 19 of the Equality Act 2010 treats a rule that puts people sharing a protected characteristic at a particular disadvantage as indirect discrimination unless it is a proportionate means of achieving a legitimate aim. Apply that test alongside the operational checks in the UK AI recruitment guide.
Keep meaningful human involvement in rejection decisions
Under UK GDPR Articles 22A to 22D, a decision is solely automated when there is no meaningful human involvement in taking it. Rejecting an applicant can qualify as significant: the test is a legal or similarly significant effect on the person. For meaningful review, give the reviewer the original evidence, time to assess it and authority to change the outcome.
For significant decisions based solely on automated processing, Article 22C requires safeguards that provide:
- Information about the decision.
- An opportunity to make representations about it.
- Human intervention.
- A way to contest the decision.
I'd have a person review screened-out applicants before anyone is rejected, recording the reason and correcting misapplied rules.
Check local tool requirements and interview access
Employers and agencies using covered tools under New York City's Local Law 144 need a bias audit within one year of use, publicly available audit information and candidate notices. Give the required notice 10 business days before using the tool.
For UK interviews, ask every invited candidate about reasonable adjustments, including for AI and one-way formats. Acas explains that this is a legal requirement. Keep online interviews consistent with the in-person process, arrange a connection backup and obtain permission before recording.
Metrics to review every week
Keep stage definitions and reporting cohorts consistent. Compare observed rates with your capacity assumptions and investigate changes before tightening screening.
| Measure | How to calculate it | What a change tells you |
|---|---|---|
| Applicants per opening | Applicants / openings in the campaign | Whether applicant supply supports the plan |
| Share passing screening | Applicants passing / applicants screened | Whether sourcing, rules or applicant mix changed |
| Impact ratio at screening | Group selection rate / highest group selection rate | Where a rule needs a fairness review |
| Interview attendance | Attended interviews / invited candidates whose interview window has closed | Whether access, scheduling or communication needs attention |
| Interview-to-offer | Offers / completed interviews from the same cohort | Whether screening and interview judgments align |
| Offer acceptance | Accepted offers / offers made to a cohort whose response window has closed | Whether the offer matches candidates' expectations |
| Time from application to first response | First response timestamp minus application timestamp | Whether acknowledgment or follow-up is slowing |
| Time from application to offer | Offer timestamp minus application timestamp | Where the overall process is slowing |
| New-starter retention at the end of the first month | Starters still employed at that point / all starters in a cohort whose first month has elapsed | Whether to investigate expectations, selection or onboarding |
Separate automatic acknowledgments from substantive replies so an instant receipt does not conceal a stalled application.
How Alfa handles high-volume hiring
Alfa screens every applicant against positive and negative signals, hard requirements, hard disqualifiers and overrides. A hard requirement is met only when a resume or matching application answer shows it; silence means it is not met. Use application questions for future commitments such as shifts or travel.
Lisa puts suitable applicants in Lisa's Picks and the rest in Applied. That placement in Applied is not a rejection and sends no email. Spot-check Applied on every new job, correct faulty rules and move candidates when appropriate; a score or column should not decide a hire.
The AI Interview is a short first-round audio or video conversation in the candidate's browser, taken when convenient. Suitable applicants are invited automatically. A person on the hiring team reviews every recording and decides the next step, and candidates are told this before starting.
Frequently asked questions
- What is high-volume hiring?
- High-volume hiring is recruiting many people for the same or similar roles in a short period, so that the number of applicants outruns the time a hiring team has to read, reply to and interview each one individually. Define the campaign around a shared role, location and start window so candidates are assessed against the right requirements.
- What is considered high-volume recruiting?
- There is no fixed threshold: hiring counts as high-volume recruiting when the openings, applicants per opening or deadline mean the team cannot give every applicant individual attention, a situation common in warehouse, retail, contact center, care, hospitality, seasonal and graduate hiring. A small recruiting team can therefore need a volume process even when the headcount request looks modest.
- How do you handle high-volume recruitment?
- Handle high-volume recruitment by planning applicant and interview numbers backwards from the hire date, screening against a short set of written rules, replacing the first phone screen with a structured first-round interview, telling every candidate where they stand, and checking outcomes for adverse impact every week. Name a deputy for each operational owner so absences do not leave applications unattended.
- Can AI screen applicants for high-volume roles?
- Yes, AI can read and score every application against the job's rules, but a person should review its results and make hiring decisions, and in the UK a significant decision made solely by automated processing needs [safeguards including human intervention](https://www.legislation.gov.uk/ukpga/2025/18/section/80). Before expanding its use, agree who can pause screening when a rule produces questionable outcomes.
- What is the four-fifths rule in hiring?
- The four-fifths rule is the [US federal guideline](https://www.law.cornell.edu/cfr/text/29/1607.4) that treats a selection rate for any race, sex or ethnic group below four-fifths (80 percent) of the highest group's rate as general evidence of adverse impact. Keep a dated record of the rule version and any corrective action, so a later review can distinguish a changed process from a changed applicant pool.

