HR manager manual

Welcome. As an HR manager you run hiring end to end: you post a role, let HeadHunter parse and score every application, build a shortlist you can defend to any donor, and close the loop with every candidate respectfully. This manual walks the platform in the order a real hiring run happens — from the first CV to the final decision — so you can operate it unaided.

It covers your surface. HeadHunter only shows you actions your role can perform, so a few things live with your organization's admin instead: adding staff and setting their roles, the pool subscription and billing, the AI/data opt-in, and the audit retention policy. Where those touch your work, this manual points you to them.

A few rules that hold everywhere

These are true across every screen, and they're worth stating once:

• The interface is bilingual (English / العربية) and right-to-left aware. You can switch language at any time; the platform keeps your place.
• Dates are shown as YYYY-MM-DD — unambiguous across languages.
• Scoring is explainable, never a black box. Every rank is backed by evidence pulled from the CV, and you can point a donor to exactly why one applicant ranked above another. HeadHunter never gives you a mystery number.
• Records are audited, not silently deleted. Consequential actions — decisions, consent, comms — leave a trail you can export.
• The shared talent pool needs each candidate's explicit consent. Nothing about a candidate crosses to another organization without it. Our privacy promise: EU-hosted, never sold to third parties.

Getting started

Signing in

Go to your organization's HeadHunter address and sign in with your email and password on the login page. If your organization requires two-step verification (recommended), you'll be asked for a one-time code from your authenticator app right after your password.

The first time you sign in with a password an admin set for you, HeadHunter asks you to choose a new one before you can continue. That's expected — it means only you know your password.

Finding your way around

Once you're in, the console shell is the same on every page: a navy sidebar down the side (a rail on a computer, a drawer you open with the menu button on a phone). The sidebar lists only the areas your role can reach — Dashboard, Postings, Candidates, Pool, Reviews, Reports, Audit — and it's the same wherever you are. On a narrow screen you can collapse it to icons to save room.

At the bottom of the sidebar are three account controls:

• User manual — opens this manual for your role, in your current language, in a new tab.
• Language toggle (English ⇄ العربية) — switches the whole interface. It keeps you on the same page, so you never lose your place.
• Sign out.

Your account settings

Two self-service pages keep your account yours:

• Account → Password (account/password) — change your password whenever you like.
• Account → Security (account/security) — turn on two-step verification, or manage it if it's already on. We strongly recommend enabling it: you're handling real people's personal data.

The dashboard — what needs you now

Your day starts on the Dashboard. It isn't a wall of charts — it's a short, action-first to-do list of the work that's actually waiting on you.

At the top sits your headline — the one number that runs your day as an HR manager: how many postings are open, and how many items across them are awaiting your action. It's the pipeline at a glance, before you drop into the detail.

Below it comes What needs you now — each item a card with an exact count that jumps you straight to the right screen:

• New applicants to review, and candidates already in screening — your live triage load.
• Shortlists awaiting your confirmation.
• Offers awaiting your decision — candidates who've reached the offer stage in the interview pipeline.
• Postings closing within a week — a nudge to extend or act before the date passes.
• Unsent rejections — candidates you've decided on but haven't yet closed the loop with.
• Reference checks still awaiting a reply, and the overdue subset (waiting over a week).
• Hires missing a required declaration — a compliance gate still open on a hire in progress.
• Pool referral requests in your inbox.

Only work you're allowed to act on appears, and the counts are exact. When there's nothing outstanding, you'll see a calm "all caught up" — not an empty error. Think of the dashboard as the question "what should I do next?" answered every morning.

Below the cards, a This month panel gives you the hiring funnel at a glance — applied → shortlisted → hired for the current month — with a link straight to the full Reports.

Posting a job

Create and edit a posting

Open Postings for the list of your organization's roles, then New to create one, or a row's title to edit it. A posting holds the job's bilingual title and description and — the part that makes HeadHunter work for you — its scoring criteria.

Defining scoring criteria

Criteria are how you tell HeadHunter what a strong applicant looks like for this role: the skills, experience, languages, and qualifications that matter, and how much each weighs. This is the rubric every applicant is measured against, and it's why the ranking is defensible — you set the standard up front, in the open, before you've seen a single CV. Spend a moment here; a good rubric is the difference between a ranked list you trust and one you second-guess.

JD-assist and the AI job-description draft

Writing a job description from a blank page is slow. JD-assist helps, and you can go further with the AI JD draft: give HeadHunter a few inputs about the role and it produces an editable, bilingual draft for you. Read it, fix it, make it yours — the AI writes a first pass, you stay the author. Once you're happy, HeadHunter feeds your description through the same criteria extractor the rest of the platform uses, so your rubric and your JD stay in step.

The value gate

When you publish, HeadHunter shows the value/billing gate — a clear note about what publishing this posting involves under your organization's plan. Publishing is a real, outward-facing action, so it's deliberate, not a stray click. If your organization's plan or subscription needs attention, that's your org admin's area (billing lives with them, not with you); the gate simply tells you where things stand.

Publishing to the public board

Publishing puts the role on your organization's public job board at jobs/[org] — your own branded page listing only your published, open roles. From there candidates can read the posting and apply. A published role also becomes discoverable on the aggregated cross-org board (/jobs) — "every open job in one place" — which is where most candidates will find it (see Where CVs come from below).

Where CVs come from

HeadHunter gathers applications from several channels and funnels them into one place, so you never chase CVs across inboxes and phones.

• Email — CVs sent to your organization's intake address are ingested automatically.
• WhatsApp — the channel your network already uses; messages to your connected WhatsApp number come straight in.
• The public apply form — the Apply button on a posting (both your own board jobs/[org]/[id] and any direct apply link) collects the CV, the applicant's details, and their explicit consent.
• An authenticated apply from a candidate account — new since the candidate realm launched. A job-seeker with a HeadHunter account can apply to a role on the aggregated board with one tap, using the profile and CV they built. From your side this applicant looks and screens exactly like any other — same intake, same dedupe, the same candidate record in your organization. The only difference is behind the scenes: their application is linked to their account so they can track its status themselves (more on why that matters under Rejecting respectfully).

The scan-gate — why a photographed CV is turned away

HeadHunter's edge is that it reads CVs — including Arabic and scanned documents — to score them against your criteria. To read a CV, it needs actual text, not a photo of text. So at intake a scan-gate checks each file: a genuine digital document goes through, but a photograph or image-only scan is turned away with a short, constructive message pointing the sender to the candidate CV guide (cv-guide) — bilingual do's and don'ts on sending a CV that parses well. This isn't us being fussy; a CV we can't read can't be scored fairly, and turning it away with guidance is more respectful than silently ranking it last.

Triage

New arrivals land in Intake → Triage (intake/triage), where you sort what came in — confirm matches, resolve duplicates, and route each CV to the right posting before it enters screening.

Screening applicants

This is the heart of your day, and where HeadHunter earns its keep.

The ranked list — and the board ⇄ list toggle

Open a posting's Applicants for every application, ranked against the criteria you set. You can view it two ways, and a toggle (top of the list) switches between them; your choice is remembered on that device:

• List — the ranked table, best for reading scores in order (the default on a phone).
• Board — columns by application stage with lightweight cards, best for seeing where everyone sits and moving people along (the default on a wide screen). You can restage a candidate by dragging their card or with the status buttons — both go through the same audited action.

"Why this score" — the breakdown with CV evidence

This is the differentiator, so use it well. Open any applicant's Breakdown and you'll see the score taken apart criterion by criterion — and for each one, the actual evidence from their CV that earned (or didn't earn) the points. Not "8/10 for experience," but the sentence in the CV that shows it. This is what makes a shortlist donor-defensible: you can point to the CV text behind every rank, and so can anyone who audits your decision. HeadHunter never asks you to trust a number you can't explain.

The CV view

Every applicant's CV is one click away, so you can read the whole document in context alongside its score.

Compare, side by side

When two or three applicants are close, pick 2–4 of them and open Compare. HeadHunter lays them out in a side-by-side criterion matrix — the same evidence per cell as the breakdown — so you can weigh them directly. Compare reflects the scoring order; it never re-ranks anyone.

Semantic match — an optional recall aid

Your criteria are keyword-based, which is precise but can miss a candidate who describes the right experience in unexpected words. Semantic match is an optional aid for exactly that: for the criteria the keyword rubric didn't hit, it looks for possible matches in the CV and surfaces them in a separate "possible matches" section of the breakdown.

Two things to know. First, it never changes the score or the ranking — it's a recall aid to make sure you don't overlook someone, not a re-ranker. Second, it only runs if your organization has deliberately opted in to AI/data processing (your org admin controls that toggle, and it's off by default). If you don't see it, semantic match simply isn't enabled for your org — nothing is broken.

Building a shortlist

When you've decided who advances, build the Shortlist from the posting. From there you can:

• Share it with a read-only link (s/[token]) — send a hiring lead or director the shortlist to look at without giving them a login. The link is read-only and scoped to that one shortlist.
• Export to PDF or print it — a clean, presentable document for a meeting or a donor file.
• Invite reviewers (hiring managers). For a colleague who should weigh in properly — comment on candidates, approve, request changes — invite them as a hiring-manager reviewer on the posting. HeadHunter sends them a set-password link; once in, they see only the shortlist you assigned them, nothing else in your platform. You can see each reviewer's status and remove them at any time.

Rejecting the rest, respectfully

Closing the loop is part of hiring well, and HeadHunter makes the respectful path the easy one. From a posting's Reject flow you send the candidates you're not advancing a clear, kind message in their own language, in bulk, honoring the consent they gave when they applied.

This matters more than it used to. For anyone who applied with a candidate account, the outcome now also appears in their own applications tracker (/me/applications) — the person sees the result in-account, not just in an email. The "close the loop" promise is now candidate-visible, which means the tone of a rejection is seen and remembered. Be brief, be human, and never leave someone wondering. HeadHunter won't let a candidate silently vanish off the bottom of a list.

Compliance and finalizing a hire

Before a hire is final, the Compliance panel on an applicant (postings/[id]/applicants/[appId]/compliance) captures what a donor will ask about:

• Declarations — the PSEA/safeguarding checklist. This is the gate: the declarations must be recorded before the hire can be finalized.
• Reference checks — record your reference requests and their outcomes.
• Finalize the hire — once declarations are in and you're ready, finalize. This is the moment the decision is committed, and it's audited.

In some organizations a separate compliance officer records the declarations and reference evidence while you make the final hiring call — the platform keeps those responsibilities distinct on purpose (separation of duties). Either way, the compliance panel is where the two meet.

Interviews

For candidates who advance past the shortlist, drive the interview stage from the Reviews area's per-application interview panel (reviews/[id]/[appId]). You move a candidate through the stages — shortlisted → interview 1 → interview 2 → offer — and can add a scheduling note.

Every interviewer fills a structured scorecard, and HeadHunter aggregates the ratings across interviewers so you see the panel's combined view, not a single opinion. Every stage change, schedule, and submitted scorecard is audited.

Candidates and the talent pool

Your organization's directory

Candidates is your organization's own directory — search and filter every candidate you've seen, and open a profile to view their history across all your postings. This is your memory: a strong applicant who wasn't right this time is easy to find next time.

The shared talent pool — the moat

The Pool is the productized version of the inter-NGO talent sharing that used to happen informally over WhatsApp: a cross-organization pool you can browse and search to find people beyond your own applicants.

Two guardrails make it trustworthy, and they're features, not friction:

• Explicit candidate consent. A candidate only appears in the shared pool if they consented to be shared across organizations. No silent cross-org sharing, ever.
• An active subscription. Pool access is part of your organization's plan. If it's active, you can browse; if it lapses, you'll see a respectful lock notice rather than a broken page. (Managing the subscription is your org admin's job.)

The referral loop

Found someone promising in the pool? Use Request to connect on their profile. The other organization sees your request in their referrals inbox (pool/referrals) and can approve or decline; contact details are revealed only on approval. Your own inbox works the same way for requests coming to you, and you can withdraw a request you sent. Every step is audited. This is the network effect that makes HeadHunter more valuable the more NGOs join.

Reports

Reports gives you the donor-facing view: the hiring funnel (how applicants flowed from application to hire) and a localization / diversity dashboard, over a date window and posting filter you choose.

Everything here is aggregate and anonymized — with small-cell suppression, so a figure can never be small enough to identify one person. That's deliberate: donor reporting should demonstrate your process without exposing any individual candidate.

Audit

Audit is the read-only trail of consequential actions across your hiring — decisions, consent, comms, stage changes. You can view it and export it to CSV for a donor review or your own records. Because HeadHunter audits rather than silently deletes, this trail is the honest, defensible record of how a hire was made.

The audit retention policy — how long the trail is kept — is set by your org admin, not here; this page is where you read and export what's been recorded.

Quick reference

You run the whole hiring wedge from these screens: post → screen → shortlist → reject → comply → hire, with the pool, reports, and interviews around it. When in doubt, start on the Dashboard — it always knows what's next.