← all playbooks

Know when a target account opens a role in your discipline

Every role in your discipline that has appeared at your target accounts since the last check, the person who owns each hire with verified work email, and a draft note per role waiting for your edit.

use when
you already have a list of accounts and want to know when one of them opens a role you recruit
starts from
A list

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 check my target accounts for roles in my discipline that
were not there last time.

The accounts are in <ACCOUNT_LIST_FILE_OR_PASTED_NAMES>. Ask me for any
company website you are unsure of, do not guess it.

For each company:
- Read its own career page and job data for <ROLE_TITLES> in <REGION>.
- Compare against <LAST_RUN_FILE_OR_DATE> and show me only what is new. If
  there is no last run, treat this one as the baseline.
- Drop agency adverts and reposts of a role I have already seen.
- For each new role, find the person who owns the hire and get their work
  email. This is business development.
- Draft a short note per role that is about that role, not a template.
- List the accounts you checked that had nothing new.

Run one company end to end first so I can see the output, and tell me what
the full list costs before you do the rest.

Stage the outreach, do not send anything.

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

What you need first

  • A list of named accounts, ideally with each company's website
  • The job titles you recruit, and the region
  • A Hyreflow workspace with credits
  • Optional: the output of your last run, so only what is new comes back

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

    Confirm the accounts and their websites

    free

    Each company on your list is matched to its own website before anything runs. Where the agent is unsure it asks, because a guessed address can point at a different company with a similar name.

  2. 2

    Read each career page

    credits

    The company's own site is mapped and followed through to the applicant tracking system behind it, filtered to the titles you recruit. Launching a read is free. You pay per role that comes back, so an account with nothing open costs nothing.

  3. 3

    Add first-seen dates from job data

    credits

    A company signals source holds each company's openings with the date every role was first seen, and can be asked only for roles first seen since a date you give. That separates a fresh role from one that has sat there all year.

  4. 4

    Keep only what is new

    free

    This run is compared with the last one, posting by posting, keyed on the link to the advert. Reposts, agency adverts and roles outside your discipline are dropped, so what is left is what changed.

  5. 5

    Find who owns each new role

    credits

    Where the advert names a reporting line, that person is found directly. Where it does not, a company-scoped search looks for the head of the function, sized to the company.

  6. 6

    Get and verify the work email

    credits

    The enrichment waterfall runs in a fixed order and stops at the first hit, then the address is deliverability-checked. Work email, because this is business development.

  7. 7

    Draft a note per new role

    credits

    One short draft per role, written about that role and your history with the account. It is a starting point for your edit, not a finished email.

  8. 8

    Stage, do not send

    free

    Drafts sit beside each role and contact in the output. Nothing is loaded into a sequencer or sent unless you ask for that as a separate step.

What a run costs

Credits are spent per company 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.

companiesif the first provider answersif every lookup walks the chain
25$3.939 credits$12120 credits
100$16160 credits$50500 credits
500$78780 credits$2482480 credits
1,000$1551550 credits$4954950 credits

Free before anything is charged

  • Confirm the accounts and their websites
  • Keep only what is new
  • Stage, do not send

What moves the number

  • The channel. This play buys work email, because these are people you are approaching about work you want to win. A work address is the cheaper of the two to find.
  • Coverage on people search and work email. The chain stops at the first provider that answers, and only that provider bills.
  • How many companies 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 company, 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.

A short list you already believe in beats a wide one

You already know which companies you want to work with: current clients, lapsed ones, the targets you have been circling. What you do not know is which of them opened a role in your discipline since you last looked. Checking forty career pages by hand is the task that never gets done.

This play does that check. It differs from find every company hiring in your market in one respect that changes everything downstream: the list is yours. Nothing that comes back needs qualifying, because you qualified the company when you put it on the list. A company's own career page carries its own roles, so there is little agency noise to clear. Forty accounts checked properly return fewer results than a market sweep and lead to more conversations.

The second idea is memory. A career page read once is a snapshot. Read against the last run, it becomes a signal: this role was not there before. That is what you act on, and it is what stops the same twelve vacancies landing in front of you every week.

What you get back

A table of what changed: company, role title, location, the link to the advert, the date the role was first seen where job data carries it, and whether it is new since the last run. For each new role, the person most likely to own the hire with title and verified work email, and a short draft note about that role. Accounts with nothing new are listed as checked, so silence reads as a result and not as a gap.

Variations worth knowing

Watch the same list for other signals. Funding rounds and news events can be pulled for a named company, so the accounts can also be checked for a raise or a leadership change.

Put it on a schedule. Once a hand run looks right, the check converts into a saved workflow. The daily business development brief is the play that covers scheduling.

Where this goes wrong

The first run has nothing to compare against. Everything looks new because nothing was recorded before. Treat run one as the baseline, or use the first-seen dates to pick out the recent roles.

A wrong website. A guessed address can map a different company with a similar name, and every later step inherits the mistake. When the agent asks which site is right, answer it.

An empty result is ambiguous. It can mean no open roles, or a jobs page the read did not find. Give the agent the direct address of that company's jobs page and run it again.

One advert, thirty agencies. A message about a single advert is the least differentiated outreach there is. What the others lack is your history with the account. Put it in the draft.

Questions

How is this different from searching the whole market for who is hiring?

The market-wide play starts from a niche and finds every employer advertising in it, so most of the work is cleaning: removing agencies, collapsing duplicates, deciding who is worth a call. This one starts from companies you have already decided are worth a call and only asks what changed. It returns fewer results, and nearly all of them are usable.

How does it know a role is new?

Two ways. The output of each run is kept, and the next run is compared against it advert by advert, so anything not seen before is flagged. Job data also carries the date a role was first seen, which helps on the first run when there is nothing to compare against.

What am I charged for when I keep checking the same companies?

Launching a career page read is free and you pay per role returned, which is why the read is narrowed to your titles and roles outside them are left out. The job data lookup is charged per company asked, whatever it returns. The agent runs one company, shows you what it cost, estimates the full list, and waits for your go-ahead.

Can I run this automatically every week?

Yes, once you have run it manually and trust the shape of the output. It converts naturally into a scheduled workflow that checks the list for roles first seen since the last check and puts them in front of you. Prove it by hand first, because a watch running unattended on a wrong website or a loose title filter produces confident nonsense every week.

Does drafting a note mean anything gets sent?

No. The drafts wait in the output for you to read and change. Loading contacts into a campaign is a separate step you ask for, and starting a send is always yours.