Humanly

What it is

Humanly is a conversational-AI hiring platform for high-volume, hourly, and frontline roles. Candidates get screened, interviewed, and scheduled over chat, SMS, email, phone, and video, and the results write back into the employer’s ATS rather than into a second system of record. It was founded in 2019 in Seattle by Prem Kumar, Andrew Gardner, and Bryan Leptich.

Two things changed its position in the last year. First, the category leader stopped being independent: Workday closed its ~$1B acquisition of Paradox on October 1, 2025, and Olivia is now a Workday product. Second, on October 14, 2025 Humanly bought three companies in one announcement — Sprockets (sourcing, screening, and a frontline-manager app), Qualifi (automated phone and voice screening), and HourWork (post-hire onboarding, engagement, and retention for hourly workers). That took the product from a screening chatbot to a sourcing → screening → interview → hire → retain line. In May 2026 it raised a $25M Series B led by SEEK Investments, with MassMutual Catalyst Fund joining and Drive Capital and Zeal Capital Partners following on. Vendor-stated scale: 250,000+ candidates engaged per month and 5M+ interviews run to date.

Why it shows up in Recruiting/TA stacks

• It is the independent answer to a suite-owned category. If you run iCIMS, UKG, Dayforce, or SAP SuccessFactors and do not want your conversational-AI layer steered by Workday’s roadmap, Humanly is the standalone platform with the closest feature overlap to what Paradox sold before the deal. Buyers who were mid-cycle on Paradox in late 2025 are the single most common reason this entry gets pulled up.
• Voice screening is first-party, not a partner integration. Qualifi’s automated phone screen is now a native channel alongside chat, SMS, email, and video. That matters in frontline funnels, where applicants routinely answer a phone call from an unknown number but abandon a web form on a phone browser — so the channel mix, not the question set, is what moves screen-completion rate.
• Retention sits in the same system as hiring. HourWork extends the same conversational layer past the offer into onboarding check-ins and stay conversations. In hourly hiring the expensive failure is usually an early-tenure quit rather than an unfilled req, so running both from one candidate record is the argument for buying this instead of an interview point tool.
• Frontline HCM write-back. Published ATS integrations cover Workday, SAP SuccessFactors, Dayforce, Employ, UKG, Oracle Taleo, iCIMS, Greenhouse, and Avature — the set that actually appears under multi-site hourly employers, not just the startup ATS list. Named customers include MGM Resorts, Trinity Health, POOLCORP, Addus Homecare, and Domino’s franchise groups.

Pricing

No public price. The pricing page is a demo request, and every deal is an annual subscription scaled to interview or conversation volume plus recruiter seats. Third-party reporting puts mid-market deployments in a $15–50K/year band, with enterprise multi-brand rollouts above that and implementation billed separately. Treat that band as triangulated, not vendor-published. Humanly rates 4.8/5 on G2 across roughly 115 reviews, and price is the most frequently named drawback in them — which is the usual signature of volume-based pricing meeting buyers who modelled it per seat. Price it per completed screen against your real annual applicant count before you sign anything.

Best for

Talent-acquisition leaders at multi-site hourly and frontline employers — restaurants, retail, healthcare and home care, hospitality, field services, security — hiring 1,000+ people a year, who already run an enterprise HCM ATS and need screen-and-schedule throughput plus first-year retention from one vendor.

Do not buy Humanly if you hire in low volume or for salaried, senior, or executive roles (the contract floor outruns the recruiter hours saved, and an AI first round costs you scarce candidates); if you are already committed to Workday HCM end-to-end, where the first-party Paradox path will be tighter; or if what you actually need is a better interview transcript rather than funnel throughput — that is BrightHire and Metaview territory.

Versus the alternatives

• Paradox — still the deepest apply-by-chat and self-scheduling product, with 189M+ AI-assisted conversations behind it. Pick Paradox when you are a Workday HCM shop and want the first-party path, or when self-scheduling against manager calendars is the whole problem. Humanly wins when independence from Workday’s roadmap is a procurement requirement.
• Sense — pick it when your center of gravity is staffing or recruitment marketing: redeployment campaigns, talent-pool nurture, and SMS journeys against a Bullhorn or agency database. Humanly wins on the screen-and-interview core; Sense wins on engagement campaigns across an existing candidate base.
• Fountain — pick it when the bottleneck is downstream of the screen: document collection, I-9 and background-check orchestration, and location-by-location onboarding for hourly workers. Humanly wins on conversational screening depth; Fountain wins on high-volume onboarding logistics.
• Apriora — the fastest-growing entrant in AI interviewing, and the pick when live adaptive interviews and cheat detection matter more than frontline channel coverage. It does not do post-hire retention at all.

If none fit and you hire fewer than a few hundred people a year, a recruiting coordinator plus Calendly and your ATS’s own screening questions still costs less than any of these contracts.

Watch-outs

• A four-way roll-up is four codebases. Sprockets, Qualifi, and HourWork were separate products eleven months ago, and “one platform” in a press release is not one data model. Guard: in the demo, make them show a single candidate record moving from source to screen to voice interview to a 60-day stay conversation, in one UI, on your data — and put the integration milestones in the contract with an exit right if they slip.
• Automated-employment-decision rules land on you, not the vendor. Conversational screening that ranks or filters applicants falls under NYC Local Law 144, the Illinois AI Video Interview Act, Colorado’s SB 24-205, and the EU AI Act’s high-risk hiring category. Guard: run the annual bias audit and publish it, post the candidate notices, offer a documented non-AI alternative path, and get model documentation plus a DPA before launch. See the recruiting AI-use policy.
• Vendor-stated outcome numbers are marketing baselines. “8x faster time-to-hire” and “20+ recruiter hours saved weekly” come from selected customer stories, not an audited cohort. Guard: instrument your own pre-pilot baseline for screen-completion rate, time-to-first-interview, and 90-day retention on two req families, and make the renewal conditional on beating it.
• Channel automation can burn your employer brand at volume. SMS and voice outreach at 250,000 candidates a month means TCPA consent, quiet hours, and opt-out handling are live compliance surfaces, not settings. Guard: require documented consent capture at apply, cap message cadence per candidate, route every negative sentiment or accommodation request to a human within one business day, and track candidate experience scores by segment rather than in aggregate.