Recruitment automation that knows where to stop
Recruitment automation means letting software run the repetitive, rule based parts of hiring (sourcing, first reads of CVs, outreach, reminders, scheduling) while people keep the judgment calls. RecruitingBot automates nine of the eleven steps and leaves approvals to you.
- GDPR built in
- Human approves every shortlist
- No hiring fees
Why most hiring time goes into the wrong work
Ask any recruiter where a week disappears and the answer is rarely interviews. It goes into searching profile after profile, reading CVs that were never close, writing the same first message thirty times, chasing people who said maybe, and trading calendar emails with a hiring manager who is in meetings all day. None of that work needs a senior recruiter, yet it lands on one because nobody else has the time or the context to do it.
The point of automating recruitment is not to remove people from hiring. It is to move the expensive human hours to the places where a person actually changes the outcome: deciding what the role really needs, judging trade offs between two good candidates, selling the job to someone who is unsure, and making the offer. Everything that follows a rule and repeats per candidate is a candidate for a machine.
The eleven hiring steps and who should own each
The table below walks through the full pipeline RecruitingBot runs for every role. The middle column describes how heavy each step feels when a person does it by hand, in plain terms rather than invented hour counts, because the effort varies a lot with seniority and market. The right column shows who owns the step once the role is live.
Notice the pattern. The steps that stay with a person are the ones where context and accountability matter most: defining the role at the start and approving the shortlist at the end. The middle of the funnel, where volume lives, is where automation pays for itself. A person can review twenty profiles carefully, but reviewing five hundred with the same care is not realistic, and that gap is exactly where consistent scoring against a written rubric beats tired eyes late on a Thursday.
| Step | Done by hand it feels like | Owner with RecruitingBot |
|---|---|---|
| 1. Role intake | A short but important conversation with the hiring manager | You, with a guided form for must haves, nice to haves and deal breakers |
| 2. Job post and rubric | An hour of drafting, then rounds of edits | Automated draft, you edit the rubric weights |
| 3. Sourcing | The biggest time sink, endless searching and copying | Automated across public profiles, your database and inbound applicants |
| 4. CV and profile analysis | Hours of reading with fading attention by the hundredth CV | Automated scoring per criterion with evidence |
| 5. Outreach | Repetitive writing that gets less personal as the day goes on | Automated sequences from your own mailbox |
| 6. Replies | Constant inbox checking and context switching | Automated classification, you can take over any thread |
| 7. Screening | Phone calls that often end in a knockout in minute two | Automated knockout and pre-screening questions |
| 8. Skills test | Sending, chasing and grading tests by hand | Automated, auto scored |
| 9. Scheduling | Email ping pong across two or three calendars | Automated slot picking from the connected calendar |
| 10. Shortlist | Compiling notes into something a manager will read | Assembled automatically, approved by a person |
| 11. Follow ups | Reminders and rejections that often get forgotten | Automated reminders, rejections and retention clock |
What should never be fully automated
Two decisions stay human in RecruitingBot by design, and we think they should stay human in any setup. The first is the intake. Software can structure the conversation and suggest criteria, but only the hiring manager knows that the team really needs someone who has shipped a payments product, or that a certain degree is irrelevant despite what the old job ad said. A rubric built on a weak intake will score the wrong people very consistently.
The second is the decision about who moves forward to interviews and offers. Every shortlist waits for a person to approve or reject each candidate, and nothing in the product rejects someone automatically on the strength of a score alone. That keeps the final call accountable, keeps you on the right side of Article 22 of the GDPR on automated decision making, and protects candidates from a model being wrong about them without anyone noticing.
There is a softer third area worth keeping close: the moment a strong candidate shows real interest. The reply handler can answer factual questions about salary range, location or the interview process, but when someone is weighing a move, a personal note from the hiring manager is still worth more than any perfectly worded template. You can take over any conversation in one click.
Guardrails that keep automation from backfiring
Badly configured recruiting automation does real damage. Candidates get the same message three times, people who opted out get contacted again, and a hiring manager receives forty interview invites in an afternoon. The harm is not only internal, since candidates talk to each other and an employer brand built over years can be dented by one careless campaign. Guardrails matter more than features, so these are built into the pipeline rather than offered as optional settings you have to remember to switch on.
- Human approval on every shortlist. Scores rank candidates, a person decides who advances.
- Editable rubric per role. Each criterion has a weight you control, and every score shows the evidence behind it.
- Global suppression list per workspace. Anyone who opts out from any role is never contacted again from that workspace.
- One click opt out and an Article 14 notice in every message. Candidates always know who is writing, why, and where their details came from.
- Retention clock. Sourced people who never reply are deleted automatically after 30 days by default, adjustable from 30 to 180.
- Human takeover. Any reply thread can be paused and answered by you, and the sequence stops for that person.
Sending caps and the health of your domain
Outreach is where recruitment automation most often hurts the company running it. Blasting hundreds of cold emails from a fresh address, or from a shared sending service, gets messages filtered as spam and can drag down the reputation of the domain your whole team uses for client and candidate email. Once mailbox providers start distrusting a domain, recovering takes weeks of careful sending.
RecruitingBot sends from your own connected Gmail or Outlook and Microsoft 365 mailbox, so replies land in a real inbox and candidates see a real person. Each mailbox has a daily sending cap, and messages only go out between 8:00 and 20:00 in the recipient's local time, which reads like a human recruiter working through a list rather than a script firing at midnight. If you need more volume, the sensible move is connecting another mailbox rather than raising the cap on one, and higher plans include more mailboxes for exactly that reason.
Personalization also protects deliverability. A sequence that references the candidate's actual experience and the specific role tends to earn more replies and fewer spam reports than a generic template, and replies are one of the strongest signals mailbox providers use to decide your mail is wanted. Keep follow ups short, stop the sequence the moment someone answers, and never send a third nudge to a person who has already said not now. Those habits are defaults in the product, and they are good practice for any recruiter writing by hand too.
-
Day 1
First messageSent
Personal note on their payments work, role facts, salary range, opt out link
-
Day 4
Follow upSent
Short nudge with one new detail about the team
-
Day 9
Last noteScheduled
Polite close, door left open
A realistic rollout for your first automated role
Teams that get value quickly tend to start with one role they understand well, usually a repeat hire such as a mid level engineer or an account executive, rather than the hardest search on the board. That gives you a baseline: you already know what a good candidate looks like, so you can judge whether the scores and the shortlist make sense.
In a typical example run the funnel looks like this: around 500 people found, 80 selected after scoring, outreach sent to that group, 25 replies, screening questions to the interested ones, and 5 candidates on the shortlist. Treat those numbers as an illustration of the shape of the funnel, not a forecast, since your market, salary range and location drive the real figures. Once the first role behaves the way you expect, add the next two.
If you want the step by step mechanics before you start, the how it works page walks through a role from intake to shortlist, and the recruiting automation software page covers triggers, reminders and settings inside the product. Whichever route you take, write down what a good shortlist looked like for your last manual hire of the same role. That short note becomes the yardstick you hold the automated pipeline against during the first weeks.
Where recruitment automation is a poor fit
Automation works best for knowledge work roles with a clear profile and a reasonably large pool: engineering, data, product, design, sales, marketing, customer success, operations, finance, HR and support. It is a weaker fit for confidential executive searches, where the approach itself is sensitive, and for very small niches where the whole market is a few dozen people a specialist already knows by name.
High volume hourly hiring, where candidates apply on mobile and expect a reply by text within minutes, is also a different problem with different tools. Be honest about which of your roles fit before you automate, and keep the rest on the process that already works. If you are weighing an outside specialist for senior searches, our page on the ai headhunter compares the two approaches.
Questions buyers ask
Anything else goes to support@recruitingbot.com or the help panel inside your dashboard.
What is recruitment automation in simple terms?
It is software that performs the repeatable steps of hiring on your behalf, such as finding candidates, scoring CVs against criteria, sending personalized outreach, sorting replies, sending screening questions and booking interviews. People still define the role and decide who moves forward. Good automation shortens the busy middle of the funnel without taking accountability away from the hiring team.
Which recruiting tasks are safest to automate first?
Start with the tasks that follow clear rules and repeat for every candidate. Scheduling, reminders, rejection emails and knockout questions are low risk and noticeable wins. Sourcing and CV scoring come next, ideally on a role you know well so you can check whether the ranking matches your own judgment before you trust it on harder searches.
Will automated outreach hurt my email domain?
It can if it is done carelessly. RecruitingBot reduces that risk by sending from your own connected mailbox with a daily cap per mailbox, a sending window of 8:00 to 20:00 in the recipient's time zone, personalized copy, and an opt out link in every message. Opted out addresses go on a workspace wide suppression list so they are never contacted again.
Does automation make the final hiring decision?
No. Scores and rankings help you prioritize, but every shortlist needs a person to approve or reject each candidate, and nobody is rejected automatically on a score alone. That keeps decisions accountable and avoids solely automated decisions with significant effects on people, which GDPR Article 22 restricts.
How long does it take to automate a role?
Setting up a role takes a few minutes: fill in the intake, review the generated job post, adjust the rubric weights, check the outreach sequence and connect your mailbox and calendar. After launch the pipeline runs on its own, and you come back to replies that need you and to the shortlist when it is ready for approval.
Can I keep using my ATS?
Yes. RecruitingBot does not replace your applicant tracking system. You can import existing candidates by CSV on every plan, import from Greenhouse, Lever or Workable on Growth and Agency, and keep your ATS as the system of record for offers and onboarding.
Keep reading
Give it a role and see the shortlist it builds
Type a job title on the homepage. The demo runs instantly without an account and shows an example pipeline, job post and shortlist for that role.