← all playbooks

Turn a conference speaker list into candidates worth approaching

Every speaker on the programme with their title, company and talk, the ones who fit your brief ranked on real work history, a personal email or LinkedIn profile for each, and a first message that starts from what they said on stage.

use when
an industry event has published its speakers and the people on stage look like the hire your client wants
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 turn the speaker list of <EVENT_NAME> into candidates for
my <ROLE_TITLE> search.

The speakers are listed here: <SPEAKER_PAGE_URL>

Pull every speaker: name, title, company and the title of their talk. If
the programme runs over several pages, find them all first.

The brief I am hiring against:
<PASTE THE BRIEF OR SPEC>

Then:
- Find each speaker's LinkedIn profile. Leave it blank when the match is
  not certain, do not guess.
- Pull the real work history and score each one against the brief. Tell
  me who is a practitioner and who is on stage to sell something.
- For the ones worth approaching, find a personal email. Never their work
  email, this is a candidate approach.
- Draft a short first message per person that starts from their talk.

Run five speakers end to end first and tell me what the full run costs
before you do the rest. Stage the messages, do not send anything.

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

What you need first

  • The public speaker or agenda page for the event, this year's or a past year's
  • A role brief, or a paragraph on who you are looking for
  • A Hyreflow workspace with credits
  • Optional: a connected sequencer with a paused campaign, if you want the messages staged

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 speaker pages

    credits

    Speakers are spread across an agenda, tracks and days. The agent maps the event site first to find every page that lists them, including past years if you ask for those.

  2. 2

    Extract the speakers

    credits

    Name, title, company and the title of each talk, as printed. The talk stays on the row because it is evidence of what the person knows and the reason to write to them.

  3. 3

    Resolve each LinkedIn profile

    credits

    A programme page rarely links to profiles. Each speaker is matched on name, company and title together, and the profile is left blank when the match is not certain. These lookups are charged whether or not they find the person.

  4. 4

    Pull the real work history

    credits

    A speaker bio is marketing copy. Dated career history is pulled for each resolved profile, from providers tried in a set order that stops at the first hit, so the score rests on what the person has done.

  5. 5

    Score against the brief

    credits

    Each speaker is judged against your must-haves with the evidence quoted back. Vendor evangelists, founders who will not move and people outside your geography are dropped here, before any contact details are bought.

  6. 6

    Find personal contact details

    credits

    Personal email for the survivors only, walked through the providers in a set order, then a deliverability check. Never the work address, because this is a candidate approach.

  7. 7

    Draft and stage the approach

    credits

    A short first message per person that opens on their talk. Drafts go into a paused campaign as per-person text, or come back as a file. Starting a send is a separate action that you take.

What a run costs

Credits are spent per candidate 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.

candidatesif the first provider answersif every lookup walks the chain
25$4.343 credits$13130 credits
100$17170 credits$51510 credits
500$85850 credits$2552550 credits
1,000$1701700 credits$5105100 credits

What moves the number

  • The channel. This play buys personal email, because these are candidates, and a work inbox is the wrong place to approach one. A personal address costs more to find than a work one.
  • Coverage on work history and personal email. The chain stops at the first provider that answers, and only that provider bills.
  • How many candidates 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 candidate, 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 committee has already run your first screen

Every industry event runs a selection process you do not pay for. Someone reads the submissions, asks around, and picks the people worth putting in front of a paying audience. The result sits on a public web page: a name, a title, a company and the title of a talk.

For a recruiter that page is two things a database row is not. It is a shortlist somebody else curated, and it is a statement, in the person's own words, of what they know. "Head of Data" tells you little. A talk on moving a bank's risk models off a mainframe tells you what the last three years of their job looked like.

It is the candidate-side twin of working the exhibitor list for business development. Same page, opposite purpose. There you want the company's hiring manager, on a work email, before the doors open. Here you want the person on stage, on a personal channel, whenever the search needs them.

What you get back

A table with one row per speaker: name, title, company, talk, the LinkedIn profile where the match was certain, the dated history behind the score, the score and the reason. For the ones who passed, a personal email where one exists and a drafted first message that opens on the talk. Drafts are held. Nothing is sent.

Variations worth knowing

Run several years at once. People who speak at the same event year after year are the recognised names in a niche. People who spoke once, four programmes ago, are often the ones nobody is calling.

Use the talks as a map. Group the programme by topic before scoring and you see which employers send people to talk about which problem. That shows where a skill sits in the market, and it feeds a full talent map.

Count panels and workshops. Keynote speakers are the most approached people at any event. Workshop leaders and panel members are closer to the work and get far less recruiter mail.

Where this goes wrong

Mistaking a stage for a fit. Vendors buy speaking slots. A good part of any programme is founders, evangelists and sales leaders who are there to sell. The score step separates practitioners from promoters, and it needs real must-haves to do it.

The right name on the wrong profile. Matching a name to a profile fails quietly. The play leaves the profile blank when company and title do not line up, because a wrong profile poisons the history, the score and the email that follow.

A generic message with the talk pasted in. "Loved your talk" from someone who was not in the room reads as what it is. Say what the talk was about and why it matters for the role, or leave it out.

Treating it as the whole search. A programme is a few dozen people. It supplements a proper search. It does not replace one.

Questions

Why would a speaker list beat a normal search?

It does not beat it, it does a different job. A search returns everyone who holds a title. A programme returns the few people an organising committee judged good enough to put in front of a paying audience, each with a public statement of what they know. That second part is what no database row carries: the talk tells you the person's real specialism in their own words.

Are speakers not the hardest people to move?

Often, yes. They are visible, senior and well looked after, and recruiters already write to them. That is why the play scores before it spends and why the message has to start from the talk. A speaker list suits a hard senior search where a handful of right names matters more than volume. It is a poor fit for a role you fill in numbers.

Can I use last year's programme?

Yes. Unlike the business development version of this play, nothing here expires when the event closes. Past programmes are an archive of who was credible on a topic in each year, and the work history step shows where each of them is today, so an old list still gives a current answer.

What about the attendee list?

Attendee lists normally sit behind a login and are not public, so they are out of scope. Speaker pages are published deliberately. Where a programme is an image or a PDF that cannot be read reliably, or the site says it does not want automated access, the agent stops and tells you instead of guessing names.

Why a personal email and not the address on their company site?

Because you are asking someone about leaving their employer, and their employer's mail server is the wrong place for that conversation. The play looks for a personal address and leads with LinkedIn. Where neither is found the row stays on the list with the profile only, and you decide whether a connection request is enough.