Screening Screening completion rate: what it can and can't tell you about candidate experience. Metaview's figure counts candidates who started the call, excluding invited candidates who never began. Ask about the completion event, cohort, raw counts, and results split by the decision a human made.
Screening How to decide what to screen for and what to leave to the interview. Build a hiring map that records the instrument, delivery format, evidence, and limit for every criterion. Then connect Screening to the short structured conversation and keep the map current as recruiters refine the rubric.
research-insights Metaview-captured interviews peak at 1,000-5,000 employees by 1.5 minutes. Interview length is greatest in the band spanning 1,000 to 5,000 employees on both measures, with a 1.5-minute lead in Salesforce-matched customer data.
research-insights Detected language mix across 6,074,772 conversations captured on Metaview. The detected-language record is dominated by English, while its non-English side extends across 69 base languages.
research-insights The 1.24% debrief label: what it can and cannot tell you. A debrief label marks one bucket in the captured-conversation type split, not how often hiring teams hold debrief meetings.
research-insights Phone ranks ahead of Zoom in platform-labeled conversations captured on Metaview. Among conversations captured on Metaview that carry a platform label, phone accounts for 28.54%, second only to Google Meet at 30.64% and ahead of Zoom at 21.64%.
interviewing The recruiting scorecard playbook: one definition of the role, from intake to decision. Metaview’s product data contains 804,803 scorecard templates, most of them used for a single interview. The playbook shows how a TA team can preserve the same role criteria throughout a search.
Sourcing How to write a hiring brief you can delegate, to a person or an agent. A brief another person can run without a second intake meeting. Six elements, a worked example, and a test for whether the search is really specified.
Sourcing Why candidate replies get lost after outbound outreach. Reply rate only shows that a conversation started. Three habits that keep a positive reply from being buried, and how to tell when manual follow-up has run out.
Sourcing The searches you never run: an honest capacity audit for outbound recruiting. Outbound is the first work to slip once the week fills up. A working audit that shows which searches actually get hours, and a way to name the coverage model for every role.
Recruiting Metrics How much does interview order change the final hiring decision? Order can enter a hiring decision in several ways, and no study supports a universal number for how much it changes the result. What the research supports, what Metaview’s corpus can’t show, and how to keep sequence out of the rubric.
Interviews How AI-native startups interview differently. AI-native teams design the loop around agreed criteria and explicit human decisions, then let AI prepare the evidence. What the data supports, and where it stops.
Recruiting Metrics Do whiteboard interviews actually predict job performance? Whether a whiteboard interview is worth keeping depends on the role, the exercise, and the rubric. What the data can and cannot say, and how to make technical rounds produce evidence worth scoring.
Recruiting Metrics AI is baseline: what Shopify's memo means for your hiring process. Shopify made reflexive AI use a baseline expectation for its own teams. That is an operating policy rather than a universal hiring bar. What the memo actually asks of hiring teams, and how to assess AI fluency when a role needs it.
Recruiting Metrics Do the best interviewers ask fewer questions? Question counts vary far more between interviewers than most hiring teams expect. That spread is worth reading. It isn’t a ranking, and no interview record holds the outcome that would make it one.
Interviews SaaS hiring benchmarks: what Metaview's data can and can't tell you. Metaview's corpus has no validated industry field, so it cannot produce a SaaS hiring benchmark. What it can show is how often a hiring decision leaves a written record behind.
Automation The agentic talent acquisition strategy: 7 workflows and what the evidence supports. An agentic talent acquisition strategy puts an AI agent at each stage of hiring so the record one stage produces reaches the next. Seven workflows, the evidence behind two of them, and the limits nobody can talk past.
research-insights AI scorecard adoption across 5.5M conversations. Interviewers keep 77.4% of the scorecards Metaview drafts, and a generated first draft gets submitted far more often than one written from scratch. Where AI scorecard adoption is real, and what moves it.
research-insights Time to hire by role: what interview data can and cannot measure. An interview recording carries two timestamps, the start and the end of the conversation. It carries no hire date. Here’s what a corpus of 5.2 million recorded interviews can and cannot say about time to hire by role.
Interviews Skills-based hiring: how to interview for skills, not pedigree. Skills-based hiring is sold on what happens after the offer, which is the one thing a hiring loop never observes. Here’s the narrower check that’s actually available: whether the loop assessed the skills you named, and whether the record survived to the decision.
Application Review Inbound application overload: what an agent can actually fix. Learn why inbound application screening breaks at scale and how AI recruiting agents help teams review applications faster, identify top talent, and reduce manual work.
Interviews How to conduct a recruiter phone screen: questions, a script, and a screen scorecard. How to run a recruiter phone screen that leaves the next interviewer something to read: what to prepare, a six-question core, and how to calibrate the stage without over-reading conversion.
research-insights More than half of interviews run over. That does not mean they are too long. 54.3% of interviews run past their scheduled slot, and among those that do the median overrun is 11.9 minutes. What an overrun tells you about how a stage is scoped, and the conclusion this data will not support.
Hiring Strategy The bar raiser program: how to build one without being Amazon. What a bar raiser actually does, how to stand up a program with the people you already have, and why the test of one is the record it leaves behind rather than the hires it’s credited with.
AI Agents What an agentic recruiting platform actually means. One shared context layer makes the whole team faster, and it can make the whole team wrong in the same direction. What an agentic recruiting platform actually means, and what changes Monday morning.
