A healthcare staffing case study filling 45 nursing roles in 30 days
In this healthcare staffing case study, a Toronto agency placing nurses in Ontario, BC and Alberta hospitals moved first interviews to The Cognitive's live AI video interview. Customer-reported results: time to first interview fell from 5 days to 3 hours, drop-off from 58% to 12%, and it filled 45 positions in 30 days.
What changed for the Canadian nurse staffing agency?
The first 3 rows are the agency's own figures. The last 2 are our arithmetic on them.
| Metric | Before | After | Change |
|---|---|---|---|
| Positions filled (customer-reported) | 8 a month | 45 in 30 days | About 5.6 times as many |
| Time to first interview (customer-reported) | 5 days | 3 hours | 117 hours sooner |
| Candidate drop-off (customer-reported) | 58% | 12% | Down 46 points |
| Candidates still in process, per 100 (our arithmetic) | 42 | 88 | About 2.1 times as many |
| Hours from applying to first interview (our arithmetic) | 120 | 3 | 40 times faster |
Customer figures from our case study record dated 4 April 2026. Per-100 and hour figures are our arithmetic on them.
Why were nurses dropping out before the first interview?
The customer is a healthcare staffing group based in Toronto, named in our record only as Healthcare Staffing Group. It places nurses with hospitals in Ontario, British Columbia and Alberta, and before this change it filled about 8 positions a month.
The candidates were the problem nobody could schedule around. Nurses applied after night shifts and could not take a call during office hours. The first interview took 5 days to arrange, and by then most had accepted another agency's offer. 58% of candidates dropped out of the process, most of them during that wait.
Nurses work 12-hour shifts. They cannot take a screening call at 2 PM on a Tuesday. VP Talent Acquisition, Healthcare Staffing Group, Toronto
Why does speed decide who wins a nurse?
Nursing is a candidate's market. The same nurse is usually talking to several agencies at once, and the agency that reaches a decision first tends to get the placement. A 5-day wait is not an inconvenience in that market. It is the loss itself.
Before The Cognitive, the agency tried faster manual phone screens, more calendar chasing and standard hiring software. None of it fitted people coming off a night shift. Every option still needed the nurse to be awake and reachable during office hours, so applicants who looked strong on paper kept vanishing before a first conversation.
Statistics Canada counted 170,800 job vacancies in Ontario and 73,100 in Alberta in July 2026, Alberta's highest since September 2024. In the second quarter, 25.9% of Canadian vacancies had been open 90 days or more, and health occupation vacancies fell 5,500 (7.9%) on the year. Both read 5 October 2026. Fewer openings still leave agencies racing each other for every nurse who applies.
How does a nurse interview after a night shift?
A live, two-way AI video interview: the candidate checks camera and microphone, the AI asks and follows up.
Nurses book from the invitation. Today the booking page offers start times on the hour and half hour across the whole day, weekends included, within the dates the recruiter opens, starting at least 2 hours after booking. No recruiter needs to be awake. The AI runs the interview and the report is waiting when the team logs in.
Which situational questions suit a nurse screening interview?
The record says the interviews were behavioural and situational, tailored to healthcare. These are illustrations of that style, not the agency's own questions.
Early in a shift, two of your patients need you at the same moment. Talk me through what you do first and why.
A clear reason for who comes first, a request for help named out loud, and a plan to document what happened.
What would make you change that order halfway through?
A patient's condition is changing and the physician on call dismisses your concern. What do you do?
Restates the observations in specific terms, escalates through the proper chain, and records the conversation.
How do you raise it again without damaging the working relationship?
You are placed at a hospital you have never worked in, on a unit with different charting. How do you get safe on day one?
Asks for an orientation, finds the charge nurse early, and checks local policy before acting on habit.
Tell me about a time a new site's process surprised you.
What should a first interview for an agency nurse probe, and why?
An example rubric for agency nursing, with weights adding to 100. It shows the kind of setup that fits this case, not the agency's own wording.
| Example criterion (weight) | What the AI listens for | Why it matters for agency placements |
|---|---|---|
| Clinical judgment and prioritising (30) | A clear order of actions, with the reason for each, when two patients need care at once | Agency nurses join units mid-stream and have to triage without knowing the team |
| Escalation and safety (25) | Specific observations, the escalation chain used, and what was documented | Hospital clients judge an agency on whether its nurses raise concerns early |
| Adapting to a new site (20) | How they learn local policy, charting and who to ask on day one | Every placement is a new building, new systems and new colleagues |
| Communication with patients and teams (15) | Plain explanations, checking understanding, a calm handover | Short placements leave little time to build trust slowly |
| Availability and shift reality (10) | Shifts, units and provinces they can actually work | A strong nurse who cannot work the client's shifts is not a placement |
Registration and licence status are asked about but checked by a recruiter with the provincial regulator. See the nurse interview questions for more prompts.
How did the agency run it, step by step?
The steps below use today's product. The agency's setup dates from April 2026, and the product has changed since.
- Write the nursing criteria Clinical judgment, prioritising, communication with patients and teams, and fit for agency work across sites. Put what a strong answer sounds like into each criterion's description. Set the role to 20 minutes so the AI has time for a follow-up on each clinical scenario. Use 10 minutes for a fast availability-first screen.
- Agree what ready to place means The record says recruiters agreed this before launch. Turn it into the score at which a report gets read first, not a rule that rejects anyone.
- Open booking dates wide Set the role's booking window so nurses can pick any start time, including the hours right after a shift ends. Run a French version of the role alongside the English one if you recruit bilingual nurses.
- Invite every applicant the day they apply Import from your ATS through one of 60+ ATS integrations, or upload a CSV, then bulk invite the batch in one go. The agency's time to first interview fell to 3 hours once nobody had to coordinate a calendar.
- Read reports at the start of the day Each report gives a 1 to 5 score per criterion, overall written feedback, a weighted score out of 100, a transcript and a recording. A workable rule: call anyone scoring 4 or 5 on clinical judgment the same morning, and watch the recording when escalation scores 2 or lower.
- Check licences, then call the shortlist A recruiter confirms registration with the provincial regulator and calls the nurses the reports point to. The AI suggests; the recruiter decides who goes to a hospital client. To widen the pool, [AI Sourcing](/ai-sourcing-tool) can search ~900M public profiles for nurses who have not applied and reveal verified emails and phone numbers.
How did three provinces get one standard?
An example rubric: what a weak, a middling and a strong answer looks like, fixed for every candidate.
A phone screen in Toronto and a phone screen for an Alberta hospital used to depend on who made the call. With one written rubric per role, every nurse was scored against the same criteria, whichever province the placement was in. The agency reported the AI evaluation was more consistent than its phone screens.
The arithmetic behind the headline numbers
Inputs are customer-reported. Everything after an equals sign is our arithmetic.
PLACEMENTS Before: 8 a month. After: 45 in 30 days. 45 / 8 = 5.6 times the monthly rate. TIME TO FIRST INTERVIEW Before: 5 days x 24 = 120 hours. After: 3 hours. 120 minus 3 = 117 hours sooner. 120 / 3 = 40 times faster. CANDIDATE DROP-OFF, PER 100 CANDIDATES Before: 58 dropped, 42 stayed. After: 12 dropped, 88 stayed. 88 / 42 = 2.1 times as many nurses still in the process. 46 more of every 100 stayed long enough to be considered. APPLICANTS NEEDED TO KEEP 45 NURSES IN PLAY Before: 45 / 0.42 = about 107 applicants. After: 45 / 0.88 = about 51 applicants. Roughly half the applicant flow for the same live pipeline. PLACEMENTS PER WEEK Before: 8 a month is about 2 a week. After: 45 in 30 days is about 10.5 a week. AGAINST SCHEDULING ALONE Ashby's 2026 data: manual scheduling takes a median of 5 hours, just to get an interview scheduled. This agency: 3 hours from applying to the interview itself. COVERAGE 3 provinces: Ontario, British Columbia, Alberta.
What changed, number by number?
Time to first interview: 5 days to 3 hours. That is the change that drove the rest. The agency's record says the binding constraint was never the supply of nurses. It was the delay between a nurse showing interest and a nurse being interviewed.
Drop-off: 58% to 12%. Out of every 100 nurses who entered the process, 88 stayed in it instead of 42. The agency did not need more applicants. It needed to stop losing the ones it already had, and most of the 45 placements came from that recovered pipeline.
Placements: 8 a month to 45 in 30 days, across 3 provinces. That is about 10.5 placements a week against roughly 2 before. Hospital clients also told the agency they were more satisfied with candidate quality, which matters as much as volume: a client that trusts the shortlist books the next nurse from the same agency.
What did a recruiter's day look like before and after?
A recruiter's day at the agency, built from the figures in the record. The per-100 figures are our arithmetic.
| Part of the day | Before: phone screens | After: AI first interviews |
|---|---|---|
| First thing in the morning | Leave voicemails for nurses who applied after their night shift | Read the reports from interviews nurses booked and completed overnight |
| Getting to a first interview | About 5 days of calendar chasing per nurse | About 3 hours, booked by the nurse from the invitation |
| Who is still there | 42 of every 100 applicants | 88 of every 100 applicants |
| Calls the recruiter makes | Screening calls with everyone who answers | Calls to the nurses the reports point to, plus licence checks |
| Consistency across provinces | Depends on who made the call | One rubric per role for Ontario, BC and Alberta |
| Placements at the end of the month | About 8 | 45 in 30 days |
What does this case study not prove?
- One agency, one 30-day period. Use it for planning, not as a forecast.
- The speed gain depends on inviting nurses the day they apply; a backlog brings the wait back.
- Licence checks, references and unit matching still take recruiter time on every placement.
- Interview quality depends on the rubric. Vague clinical criteria give vague reports.
- The setup dates from April 2026. The product has changed since; this page describes today's workflow.
Which healthcare hiring teams should copy this?
Do
- Agencies losing nurses to faster competitors during the scheduling wait.
- Shift workers who can only talk outside office hours.
- Placements spread across sites or provinces that need one standard.
- Recruiters who spend their days chasing calendars.
Do not
- Replacing licence or reference checks. Those stay with your team.
- Senior clinical leadership hires, where a panel should meet people early.
- Teams that will not read the reports before calling candidates.
What still needs a human in nurse recruitment?
Credentials. The AI can ask a nurse about their registration and recent clinical experience, and the answer lands in the transcript. Confirming that a licence is current is a check against the provincial regulator, done by a person before any placement. The record notes some staff at first re-checked the AI's notes on credentials; that eased as they read the conversations in context.
The placement itself. Matching a nurse to a hospital unit, agreeing shifts and handling references stayed with the agency's recruiters. The AI interview took the scheduling and the first conversation off their plate, and the reports told them whom to call first.
Try it on one nursing role
Open a nursing role with wide booking dates, invite this week's applicants the day they apply and read the reports each morning. Start free at https://app.thecognitive.io/signup.
Where to go next
AI recruiting for healthcare · Nurse interview questions · AI interviewer for nurses · For staffing agencies · Pricing · All case studies
Sources and notes
Figures and quote reported by the customer in The Cognitive's case study record for Healthcare Staffing Group, dated 4 April 2026. Setup dated April 2026. Ratios and per-100 figures are our arithmetic. The example questions are illustrations. Booking behaviour describes the product as checked on 4 October 2026.
Market context: Statistics Canada, "Payroll employment, earnings and hours, and job vacancies, July 2026", The Daily, 24 September 2026, and "Job vacancies, second quarter 2026", The Daily, 15 September 2026. Scheduling benchmark: Ashby, "Recruiting Operations Benchmarks", 2026 Talent Trends Report, 7 May 2026, ashbyhq.com. All read 5 October 2026.
Frequently asked questions
How do you recruit nurses faster?
Shrink the gap between application and first interview. This Canadian agency cut it from 5 days to 3 hours by letting nurses book a live AI video interview themselves, at a time that suited their shifts, instead of waiting for a recruiter's call.
Why did candidate drop-off fall so sharply?
Most drop-off in nurse recruitment happens in the wait, not in the interview. When the first interview moved from days away to hours away, the agency reported drop-off falling from 58% to 12%.
How do staffing agencies reduce candidate drop-off?
Cut the wait between applying and the first conversation, because that is where most candidates go. Invite every applicant the day they apply, let them book their own slot and read reports first thing each day. This agency reported drop-off falling from 58% to 12%, so 88 of every 100 nurses stayed in the process instead of 42.
Are nursing job vacancies in Canada going up or down?
Statistics Canada reported that vacancies in health occupations fell by 5,500 (7.9%) year over year in the second quarter of 2026. Openings remain spread across provinces, with 170,800 vacancies of all kinds in Ontario and 73,100 in Alberta in July 2026. For an agency, fewer openings mean more competition for each placement, not less.
What should a nurse screening interview assess?
Clinical judgment and prioritising, escalation and patient safety, how the nurse adapts to a new site, communication, and the shifts they can actually work. Weight clinical judgment highest. Credentials are asked about in the interview but checked by a recruiter with the provincial regulator.
How long does it take to hire a nurse through an agency?
It depends mostly on how long the first interview takes to arrange. This agency's wait was 5 days before the change and 3 hours after, and it went from about 8 placements a month to 45 in 30 days. Licence checks and client matching then set the remaining pace.
Can nurses do an AI interview after a night shift?
Yes. The booking page offers start times on the hour and half hour across the whole day, weekends included, within the dates the recruiter opens. The earliest slot is 2 hours after booking. No recruiter needs to be online for the interview to run.
Does the AI verify nursing licences?
No. The AI can ask about registration and experience, and the answers appear in the transcript. Checking that a licence is valid in Ontario, British Columbia or Alberta stays with your recruiters and the provincial regulator.
What do healthcare recruiters see after each interview?
A report with a 1 to 5 score per criterion, overall written feedback, a weighted score out of 100, a suggested verdict, the full transcript and the recording. A recruiter decides who goes forward; nothing is rejected automatically.
Can the interview run in French?
Yes. A recruiter can set a role's interview language to French, one of 9 languages available. That lets an agency run French and English versions of the same nursing role.
What should healthcare recruiting software do differently?
Work around shifts, keep one standard across sites and leave clinical checks with people. In this case study the agency needed interviews nurses could book outside office hours, scoring that held steady across 3 provinces, and recruiters who still confirmed credentials.
Is this a healthcare staffing automation tool?
It automates the first interview and the scheduling around it. The Cognitive is not an ATS or a vendor management system, so shift booking, timesheets and hospital contracts stay in your existing tools.
Can The Cognitive also find nurses?
Yes, through AI Sourcing, which searches ~900M public profiles from a plain-English brief and can reveal verified emails and phone numbers. This agency's case study covers interviewing its own applicants, not sourcing.
Does it connect to our ATS?
Yes, through 60+ ATS integrations, including Bullhorn, JobAdder and Crelate, or by CSV upload. ATS-imported candidates get a note with score, summary and report link. Stages in the ATS never move on their own.
How much does it cost?
AI Interview starts at $99/month, and AI Sourcing starts at $49/month if you also want to find nurses who have not applied. Plans are monthly and cancel anytime. The pricing page lists what each interview plan includes, so an agency can size a plan to its applicant volume.
Explore: Healthcare recruiting software · AI interviews in healthcare · Interview scheduling software · What is candidate experience? · Hire a nurse