You're hiring again. Do you finally need proper tooling?

You've got a role open — maybe two — and the applications are piling up in your inbox. You've been here before: a spreadsheet with candidate names, a shared doc for interview notes, a few sticky messages in Slack. It works, sort of, until it doesn't. Someone good falls through the gaps. You follow up on the wrong version of a CV. A hiring manager swears they never saw a particular application.

At some point, every founder or hiring manager hits the same question: do I need an Applicant Tracking System? Or is there something lighter that solves the actual problem — sorting the pile and surfacing who's worth calling?

This article gives you a straight answer.

What an ATS actually does

An Applicant Tracking System is a workflow platform for the entire hiring process, from job posting to offer letter. The core capabilities of a mature ATS include:

  • Centralised candidate pipeline — every applicant in one place, moving through defined stages (applied, phone screen, interview, offer)
  • Job board distribution — post to Indeed, LinkedIn, and others from one interface
  • Careers page hosting — a branded page listing your open roles
  • Structured interview scorecards — interviewers log feedback against a consistent rubric
  • Collaboration tools — hiring managers, recruiters, and interviewers share notes and decisions
  • Reporting — time-to-fill, source quality, offer acceptance rates
  • Integrations — connects to HRIS, background check, e-signature, and payroll tools
  • Compliance features — EEOC tracking, GDPR data handling, audit trails

Platforms like Greenhouse, Lever, and Ashby are the best-known names in this space. Workable and JazzHR sit slightly lower in price and complexity and are aimed more squarely at smaller teams.

The case for a full ATS

A full ATS genuinely earns its cost when several conditions are true simultaneously:

  • You're filling 15–20+ roles per year consistently. At that volume, pipeline management is a part-time job on its own. Manually tracking fifty candidates across eight open roles in a spreadsheet is where real candidates start getting lost.
  • Multiple people are involved in every hire. When a recruiter, two interviewers, and a hiring manager all need to log feedback and see candidate history, a shared platform with structured scorecards stops decisions from being made on incomplete information.
  • You're posting the same roles repeatedly — seasonal, high-volume, or fast-growth hiring. The ROI on configuring job templates, screening question sets, and automated stage-progression is real when you run the same process dozens of times.
  • Compliance is a hard requirement. If you're in a regulated industry, hiring at scale in multiple jurisdictions, or subject to EEOC audit requirements, the data-handling and reporting features of a proper ATS are not optional.
  • You have a dedicated recruiter or HR function. ATS platforms are built to be operated by someone who owns recruiting. If hiring is one of ten things a founder or ops person does, a full ATS adds administration rather than removing it.

If several of those are true, the $6K–$25K/year Greenhouse price tag starts making sense. Workable at $5K–$12K/year is a more accessible entry point for teams under 50 people who check most of those boxes.

Where ATS falls short for small teams

For a 10- or 20-person company that hires three to eight people a year, a full ATS often creates more work than it removes.

Cost. Greenhouse runs $6K–$25K/year. Ashby starts at $400/month ($4,800/year) on its Foundations plan. Even Lever's starting tier is $4,000/year. If you're making six hires a year, you're spending $700+ per hire just on the platform — before any time investment.

Setup overhead. Mid-market platforms like Lever and Ashby typically take two to six weeks to configure properly — pipeline stages, scorecard templates, integrations, team permissions. Enterprise platforms can take four to twelve weeks. For a small team running their third hiring process ever, that setup cost is disproportionate to the actual problem.

Ongoing maintenance. An ATS reflects the hiring process you configure, not some idealised process. Poorly designed pipeline stages, irrelevant screening questions, or scorecards that interviewers never actually fill in degrade the system into an expensive version of the chaos it was supposed to fix. That configuration requires sustained attention from someone who knows what they're doing.

The core problem is still unsolved. Here's the thing most ATS vendors won't say clearly: an ATS organises your pipeline. It doesn't tell you which candidates are actually good. Screening resumes against a job description — the part that takes a hiring manager 10–16 hours per 200 applications — still falls on a human unless you separately configure AI screening on top of the ATS (an add-on many platforms charge extra for).

What AI CV screening does differently

AI screening tools address a specific, narrower problem: given a stack of CVs and a job description, which candidates should get a closer look?

Modern tools use large language models to read each CV semantically, not just keyword-match. They produce a fit score, identify gaps against the job requirements, and generate interview questions tailored to each candidate's specific background. The output is a ranked shortlist with reasoning, not just a filtered pile.

What AI screening does not do is manage your pipeline. There are no stage transitions, interview scheduling, scorecard collection, or offer workflows. It solves the first problem — who deserves your time — and leaves the rest to you.

This is the right trade-off for a team that:

  • Hires infrequently but gets flooded with applications when they do
  • Doesn't have a dedicated recruiter; the hiring manager screens CVs themselves
  • Already has a workable process for the rest of the hiring workflow — calendar invites, a shared doc for notes, Notion or a simple spreadsheet
  • Wants faster shortlisting without committing to a platform subscription they'll use three times a year

On cost, standalone AI screeners are a different category entirely. Pay-per-batch tools mean you pay when you hire, not every month regardless of hiring activity. For a team doing four hires a year, that's a meaningful difference.

One important caveat: AI screening, like any statistical tool, reflects patterns in its training. The best tools surface reasoning transparently so a human can spot when a recommendation doesn't look right. Human review of the shortlist — not just acceptance of it — remains essential.

ATS vs AI screener: side by side

Dimension Full ATS (Greenhouse, Ashby, Lever) AI CV screener (cvtally)
Primary problem solved Managing candidates through a structured hiring pipeline Shortlisting: who's worth your time from a pile of CVs
Typical cost $4,000–$25,000+/year (subscription) Pay per batch; no monthly commitment
Setup time Days to weeks; pipeline config, integrations, permissions Minutes; upload CVs + job description, done
Hiring volume fit 15–20+ roles/year; ongoing, structured programme Any volume; especially useful for infrequent hiring bursts
Team structure fit Dedicated recruiter or HR function Founder or hiring manager doing it themselves
Pipeline management Yes — stages, scorecards, collaboration, reporting No — you handle tracking separately
CV scoring / ranking Basic or add-on; quality varies Core function; fit score, gaps, interview questions per candidate
Job board posting Yes — distribute to multiple boards from one place No
Integrations HRIS, background checks, e-signature, payroll Standalone; exports results
Compliance tooling Yes — EEOC, GDPR, audit logs Minimal

Decision guide: which one fits your situation?

You probably need a full ATS if:

  • You're consistently filling 15 or more roles per year
  • You have a recruiter, HR lead, or ops person who owns the hiring process
  • Multiple stakeholders need to collaborate on every hire — interviewers, hiring managers, a panel
  • You need structured reporting: source quality, time-to-fill, diversity metrics
  • Compliance requirements mean you need an audit trail of every decision
  • You're growing fast enough that the setup cost is a one-time investment, not a recurring drag

In this case, Workable is the most practical starting point for companies under 50 people — it's significantly cheaper than Greenhouse or Ashby and doesn't require a sales process to evaluate.

An AI screener is probably the right fit if:

  • You hire sporadically — a few roles a year, not a continuous programme
  • The hiring manager screens CVs themselves, with limited recruiter support
  • Your current pain is the pile: too many CVs, not enough signal, not enough time to read them properly
  • You have some system for tracking candidates already — even a spreadsheet — and it works well enough once you have a shortlist
  • You want to pay per hire, not a flat annual fee that assumes a hiring cadence you don't have

The honest middle ground

A number of small teams run a lightweight combination: an AI screener to produce the shortlist, then a simple free-tier ATS or even a shared Notion board to manage the ten to fifteen candidates who make the cut. This is not a compromise — it's the right architecture for a team whose hiring activity doesn't justify a full-platform investment.

The moment you feel that combination fraying — candidates falling through cracks at the pipeline stage, not the screening stage — is the right moment to look at a full ATS.