← all playbooks

Work both sides of a job ad: the hiring manager and the candidates

For each live role: the hiring manager with a work email, three to five real candidates with personal contact details, and two drafted messages, one for each side. All of it staged, none of it sent.

use when
roles are being advertised in your niche and you want to reach the hiring manager with real candidates already in hand
starts from
Just your niche

The prompt

Paste it into Claude Code or the Claude desktop app with Hyreflow connected. The first line tells your agent to use Hyreflow, so it reads the play, asks before it spends anything, and hands the work back to you.

paste this into Claude
Use hyreflow and work both sides of the live job ads in my niche: the hiring
manager behind each role, and real candidates for it.

My niche: <SPECIALISM> roles in <REGION>, posted in the last <DAYS> days.

- Find live roles from direct employers. Drop agency-posted ads and
  duplicates. Keep the best <N> roles, not the whole market.
- For each role, find the person who owns the hire, sized to the company,
  and get their work email.
- For each role, shortlist <CANDIDATES_PER_ROLE> real candidates who fit
  the ad, and nobody who works at the hiring company. Score them on work
  history, then get personal email and LinkedIn. Never their work email.
- Draft two messages per role. One to the hiring manager that leads with
  the candidates, anonymised: discipline, seniority, achievements, no names
  and no employers. One to the candidates about the role, without naming
  the company, because I do not have terms with them yet.

Run one role end to end first and tell me what the full run costs before
you do the rest. Stage it, do not send anything.

Replace every <PLACEHOLDER> with your own detail. Everything else can stay as written.

What you need first

  • A description of your niche: the role titles you place and the region you place them in
  • A cap on how many roles to work in one run
  • A Hyreflow workspace with credits

Tools it can reach for

The agent picks per step from what your workspace has. Nothing here is required by name.

What happens when you run it

Free steps are marked free. Anything that spends credits is marked, and the agent asks before the first paid run of any size.

  1. 1

    Find the live roles

    credits

    Open roles for your titles and region are pulled from LinkedIn and Indeed, with the German federal job board added for a German search. The scrape is sized to the number of roles you asked for, and you pay per role returned.

  2. 2

    Strip agency ads and duplicates

    free

    A role posted by another agency is a mandate somebody else holds, so those are removed. The same role across several boards collapses into one row.

  3. 3

    Cap the list

    free

    What is left is ranked against your niche and cut to the number of roles you set. Every role kept costs a hiring-manager search and a candidate search, so the cap is the budget.

  4. 4

    Find who owns each hire

    credits

    A company-scoped search with a title hierarchy sized to the company: the founder at a small business, the function head or talent lead at a large one. Where the ad names the reporting line, that person wins.

  5. 5

    Get the manager's work email

    credits

    This side is business development, so the channel is work email. The enrichment waterfall runs in a set order and stops at the first hit, and the address can be deliverability-checked.

  6. 6

    Shortlist candidates against the ad

    credits

    The ad is the spec. People are sourced on its titles and location, the hiring company's own staff are excluded, and dated work history is pulled so each person is scored on what they have done. Anyone who fails a must-have is dropped here.

  7. 7

    Get personal contact details for the survivors

    credits

    Personal email and LinkedIn for the three to five who made each shortlist. Never a work address: these are employed people being approached about leaving.

  8. 8

    Draft both messages and stage them

    credits

    One message to the hiring manager that opens with the anonymised candidates, and one to the candidates about the role. Both are delivered as drafts beside the two contact tables. Nothing is loaded into a sequencer and nothing is sent.

What a run costs

Credits are spent per role the play actually works, and a lookup that finds nothing usually costs nothing. The two figures are the run where the first provider answers and the run where every lookup walks its full chain.

rolesif the first provider answersif every lookup walks the chain
25$20200 credits$70700 credits
100$81810 credits$2802800 credits
500$4054050 credits$1,40014000 credits
1,000$8108100 credits$2,80028000 credits

Free before anything is charged

  • Strip agency ads and duplicates
  • Cap the list

What moves the number

  • The channel. This play buys both: work email for the client side and personal email for candidates. The candidate side is the dearer of the two, so the balance between them moves the total.
  • Coverage on people search, work email, work history and personal email. The chain stops at the first provider that answers, and only that provider bills.
  • How many roles survive the free filters. Everything dropped before the paid steps costs nothing.
  • The scoring and drafting steps run on the metered agent, charged on what they read and write rather than per role, so they sit outside this table.
  • Providers you connect with your own key. Those calls bill your account, not your credits.

An estimate, not a quote, priced at the volume credit rate. Your agent sizes the run against your own workspace and tells you what it will cost before it spends anything.

One run, two ways to get a reply

A job ad is a company saying in public that it has a problem. Most recruiters answer it with half a desk. Either they write to the hiring manager with a capability statement and no people, or they go and find people and hope the role is still open by the time they have a client to show them to.

This play does both for a small number of roles. The message to the hiring manager leads with three to five people who fit the ad, described but not named. The message to the candidates describes a real role that is open this week. A reply from either side gives you something to work: a manager who wants to see the people, or a candidate who wants to hear more and whom you can take to a named manager.

The two sides run on two channels. The manager is business development, so work email. The candidates are being approached about leaving their employer, so personal email and LinkedIn. The play keeps those as two tables and never mixes them.

What you get back

Per role: the ad, the hiring manager with title, LinkedIn and work email, and a candidate table with the score, the reason, the personal email where one was found and the LinkedIn profile. Then the two drafts. Above it all, a short note on which roles look strongest and why.

Nothing is sent, nothing is loaded into a sequencer, and nothing is written to your CRM.

Variations worth knowing

Start from one target account. Point it at a single company's career page and it works every open role there that fits your niche.

Look in your own database first. If the people you need may already be in your ATS, run find the candidates already sitting in your own ATS before buying anyone.

Send it properly. When the drafts read well, build an outreach sequence covers loading a campaign and keeping the start as your decision. Build two campaigns, one per side.

Where this goes wrong

Working the whole market. Cost scales with roles multiplied by candidates per role. A cap of a handful of roles, piloted on one, tells you whether the output is worth scaling.

Implying a relationship you do not have. The candidates have not agreed to be represented. If the manager draft reads as though they have, change it before it goes anywhere.

Only advertised roles are visible. Roles filled through networks or retained firms never appear, and agency-posted ads are removed on purpose. The list is real demand, not all demand.

Thin data at small employers. At a very small company the owner of the hire is usually the founder, and sometimes nobody is in the data at all. Those rows come back flagged, not guessed.

Questions

Why approach both sides at once?

Because a reply from either side gives you something to work. If the hiring manager answers, you already hold people who fit and can be on the phone to them that day. If a candidate answers, you hold a live role and the name of the person who owns it. Working one side at a time means every reply starts a second search.

Are the candidates real people or placeholders?

Real. They are sourced against the ad, scored on dated work history and delivered with contact details. They have not agreed to anything yet, and the draft to the hiring manager is written that way: people you can introduce, not people you represent. The first call after a manager replies is to the candidates.

Why are the candidates anonymous in the message to the hiring manager?

Naming them tells the company it can go and find them without you, and it exposes people who are employed and have not asked to be put forward. Discipline, seniority and what they have achieved is enough for a manager to decide whether to reply.

Why does the message to candidates not name the company?

You do not hold the mandate yet. Describing the role, the level, the sector and the location is honest and is enough to start a conversation. Naming an employer you have no terms with invites the candidate to apply direct, and it overstates your position.

Does anything get sent?

No. You get two drafts and two contact tables per role. Loading either list into a sequencer is a separate step that you approve, and the two lists stay separate because they use different channels.