You have two thousand channels that fit the campaign. Now you need the people behind them. So you open the first channel's About page and there is no address on it, just a button that says View email address, which wants a click, a captcha and a human being.
Once, that takes fifteen seconds. Two thousand times, it is a fortnight of somebody's working life. Which is why a YouTube creator contact list is something most teams buy rather than build, and why the pricing on it looks strange until you know what is actually being sold.
Two sources, sold as one product
The first kind of address is the one a creator publishes themselves: written into the description, or sitting on a linked site or Linktree. It is free to collect and it is public by intent. It is also thin. A minority of channels publish one in plain text, plenty of those route to a manager or a general business inbox nobody reads on a Tuesday, and a fair number have been stale since the channel changed hands.
Behind the button sits the other kind, held by YouTube rather than published by the creator. Most channels doing any kind of business have one, and it is usually the address the creator actually reads. It is gated on purpose: one click, one challenge, one address at a time, with a daily ceiling attached to whoever is doing the clicking.
Those two things get blended into a single spreadsheet and sold at a single price per row. The ratio between them is rarely stated, and it is the entire difference between a list you can send to and a list of manager inboxes.
The price per row is not the price per email
Most marketplace scrapers bill for rows processed rather than addresses returned. That distinction sounds like a technicality and it is not.
Run the arithmetic on your own numbers. Say a supplier charges a flat rate per thousand rows and returns an address for six in ten. Your real cost per address is already 1.67 times the sticker. Now assume a quarter of what came back is a dead manager inbox or bounces on first contact. You are at roughly double, and you have not yet paid for verification.
The number worth comparing between quotes is cost per address you can actually send to. Nobody advertises that number, because it is always worse than the one on the pricing page.
There is a second cost that doesn't appear on any invoice. Sending to unverified addresses drives your bounce rate up, and above about 3% you start teaching every filter in the chain that mail from your domain is not worth delivering. That reputation follows you into the next campaign. The full version of that argument is in why lead generation is an engineering problem, and it applies here with more force, because scraped creator data decays faster than B2B contact data does.
Why nobody delivers fifty thousand of these by Friday
YouTube rate-limits that gated address on purpose. It is meant to be read by one person with a reason to read it, so the ceiling is deliberately low and applies to whoever is asking. That makes supply, not code, the binding constraint.
Which means every honest supplier has a daily ceiling and can tell you what it is. Ours sits in the 200 to 300 range on a good day, and we say that upfront because it sets the delivery schedule for a campaign of any real size.
If a quote promises tens of thousands of addresses inside a week, one of two things is happening. Either the deliverable is description-scraped addresses relabelled as verified, or a large share of the rows will come back empty and you will be billed for the attempt anyway. Both look identical on the quote and very different on the invoice.
Five questions before you pay
Ask these of any supplier, including us:
- What proportion of the list is public-description addresses versus gated ones? If the answer is vague, assume the cheap kind.
- What fill rate do you get on a sample of my channels? Send twenty five you have already checked by hand. You know the right answer, so you can grade the reply.
- Am I charged for rows that return nothing? This one line changes the effective price more than any negotiated discount will.
- Is each address verified after extraction, or only extracted? Extraction tells you a string exists. Verification tells you a mailbox accepts mail.
- What is the realistic daily throughput, and does the quoted timeline match it? A timeline that contradicts the ceiling is a timeline that will slip or a quality claim that will not survive contact.
Then treat the list like it belongs to people
Creators are not B2B buyers and the outbound that works on operations directors reads badly to them. One address, one person, and a pitch count somewhere in the dozens per week. Relevance beats volume by a wider margin here than anywhere else in outbound, which is a good reason not to want fifty thousand rows in the first place.
Worth saying plainly: a published business address is not blanket permission. Rules on unsolicited mail vary by where the recipient sits, and the baseline everywhere is honest sender details, a one-click opt-out and honouring it quickly. That is the legal minimum in most places. It is also how you avoid being the fortieth ignored pitch of the week.
Where to start
Take fifty channels you already want to reach and check them by hand. Count how many show an address in the description, how many hide one behind the button, and how many have neither.
That ratio is your fill rate, and it is the number every quote should be judged against. It takes an afternoon, it costs nothing, and it turns a per-row price into something you can actually compare. Once you have it, the outbound work is the ordinary kind: verify, sequence, measure, cut what does not reply.
