← all playbooks

Turn a list of company names into facts you can qualify on

Your company list back with the columns you chose filled in: size, industry, location, funding, technology and open roles. Every value names the source it came from, disagreements are shown side by side, and gaps are left as gaps, ready for the qualification step.

use when
you have a list of company names or websites and cannot tell which are worth finding people at until you know their size, sector, funding and whether they are hiring
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 add the company facts I need to the list I already have,
so I can decide which ones are worth finding people at.

The file is <FILE_PATH_OR_PASTED_ROWS>. Each row has
<NAME_ONLY / NAME_AND_WEBSITE>.

Columns I want: <SIZE / INDUSTRY / LOCATION / FUNDING / TECHNOLOGY / HIRING>.
Buy nothing I did not ask for. Technology only matters if <TECH_OR_NONE>.
For hiring, I care about roles in <DISCIPLINE>.

How I want it run:
- Resolve every name to a website first and show me the ones you are not
  sure of. Do not enrich a company you have not pinned down.
- Skip any column my file already holds.
- Put the source next to every value. If two sources disagree, show both.
  If nothing is found, leave the cell empty and say why.

Run five companies first and tell me what the full run costs, column by
column, before you do the rest. Give me back a copy of my file. Do not
look for people or contact anyone yet.

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

What you need first

  • The list as a CSV or pasted rows, with a company name or, better, a website on every row
  • The facts that decide fit for your desk, so only those columns are bought
  • 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

    Decide which facts decide fit

    free

    The agent reads your columns, counts the rows and asks which facts would make you keep or drop a company. Each fact is a separate lookup at its own rate, so the ones that do not decide anything are left out.

  2. 2

    Pin every name to a website

    credits

    A name is ambiguous and a domain is not, and every company source matches best on the domain. Rows without one get a web search, which is billed per search, and any match the agent is unsure of is put to you as a question.

  3. 3

    Pilot five companies and price the columns

    credits

    Five rows run end to end so you see real values and what each column really cost. The agent then states the most the full list could cost, column by column, and waits for your go-ahead.

  4. 4

    Fill size, industry and location

    credits

    One firmographic lookup per company returns headcount, industry and location, and from some sources a revenue band, batched where the source takes many companies in one call. A second source is asked only for the rows the first one missed.

  5. 5

    Add funding and technology only where they matter

    credits

    Funding history and the technology a company runs are separate lookups from separate sources, each at its own rate. They run only if you asked for them, and only on the rows still worth the spend after size and sector are in.

  6. 6

    Check whether they are hiring

    credits

    A company-signal source returns a company's active openings with their category and the date each was first seen, for one request per company, billed even when nothing is on file. For the few that matter most, the company's own careers site can be scraped, billed per role returned.

  7. 7

    Hand back the list, ready to qualify

    free

    A copy of your file with the added columns, your original untouched. Every value carries its source, every empty cell carries a reason, and the file goes straight into the qualification 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$10100 credits$10100 credits
100$41410 credits$41410 credits
500$2052050 credits$2052050 credits
1,000$4104100 credits$4104100 credits

Free before anything is charged

  • Decide which facts decide fit
  • Hand back the list, ready to qualify

What moves the number

  • How many companies survive the free filters. Everything dropped before the paid steps costs nothing.
  • 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 list of names is not yet a list you can judge

An exhibitor list, a supplier directory, a spreadsheet a colleague left behind. It holds two hundred company names and tells you nothing about which twenty belong on your desk. Before you can qualify it you need facts, and before you pay to find people you need to qualify it. This play is the first of those three steps.

Company facts are sold the way contact details are: column by column, at different rates, by different providers. Size, industry and location usually arrive together in one firmographic lookup. Funding is a separate lookup. Technology is another. Whether they are hiring is another again. Asking for everything on every row is the expensive way to learn that most of the list was never a fit.

So the play starts with a question: which facts decide fit for you? A desk that places finance staff in mid-sized firms needs size, sector and location and has no use for a technology column. A desk built around one software platform needs the technology first. Only the deciding columns are bought.

The other half of the work is identity. A name is not a company. A domain is, so every row is pinned to a website before anything is looked up.

What you get back

A copy of your file, same rows and order, with the chosen columns added. Every value names its source. Where two sources disagree, both are shown. Where nothing was found, the cell is empty and a status says why: not resolved, not found, or skipped because you already held it. Hiring comes back as the active roles in your discipline, with the date the latest one was first seen.

No people are searched for and nobody is contacted. The file is ready for qualify a company list against your ICP, which turns the facts into a pass or a drop with a reason.

Variations worth knowing

Hiring only. If the list is already qualified, run the hiring column alone to decide who to ring first.

Go deep on the few. For the companies you are going to call, walk into a business development call with a one-page account brief does the full research this play deliberately skips.

Where this goes wrong

The wrong company behind the name. Names collide, and groups have subsidiaries with their own sites. Unsure matches are put to you, not guessed.

Headcount lags. Database size figures trail reality and come as bands. Treat them as bands.

Technology from a single advert. A detection is often inferred from the wording of one job advert. One detection is a hint, several are evidence.

No roles found is not "not hiring". Roles given to agencies or never advertised do not show.

Small private firms are thin. Little funding data, few records. Expect more empty cells, each labelled with the reason.

Questions

Why not buy every column for every company?

Because most of a raw list is not a fit, and the first two columns tell you so. Size and sector usually remove most of it. Funding, technology and hiring are dearer lookups that only tell you something about the companies left. Buying them in that order means the costly columns run on a fraction of the rows.

Do I pay for companies that come back empty?

It depends on the source, and the agent says which is which before you approve. Some firmographic lookups bill only when they return a record. One bills per call whatever comes back. The company-signal source bills per request even when it holds nothing on the company, and a web search bills per search. The five-company pilot shows what each row was really charged.

How does it know whether a company is hiring?

From adverts the company has published. A company-signal source lists active openings with a category and a first-seen date, which answers the question for a whole list at one request per company. A scrape of the company's own careers site is more current and is billed per role, so it is kept for the short list. Neither sees roles given to an agency or never advertised, so no roles found is recorded as exactly that, not as not hiring.

Two sources disagree on headcount. Which one is right?

Possibly neither. Each database counts the people it can see, on its own schedule, and reports a band. The file shows both values with their sources and does not average them. For a decision such as over or under two hundred people, two bands that agree on the side of the line are enough. Where they straddle it, the company's own site or a quick call settles it.

Is this the same as qualifying the list?

No, it is the step before. This play produces facts and leaves verdicts alone. Qualification reads those facts against your ideal client and gives each company a pass or a drop with a reason, and that judgement is free. Keeping the two apart means you can change your mind about what a good client looks like without paying for the data twice.