Sponsored email pricing follows deliverability before it follows list size

A sponsor asks what they get for four thousand dollars. Your rate card says one dedicated email to the registration list, 30,000 recipients. They sign, the send goes out on the Tuesday before the show, and six weeks later their marketing manager asks how many of the 30,000 people actually received it.

Nobody on your side has the number to hand. When somebody finally pulls it, the answer is 24,600, and the conversation about sponsored email pricing that follows is much harder than the one you would have had if the rate card had said 24,600 in the first place.

CEIR's B2B Exhibition Sponsorship Playbook, Part 1, published in October 2019 from responses by more than 200 organiser executives and 728 exhibitors, found that 61 per cent of organisers offer a blast email to the registration list while 33 per cent of exhibitors buy one. That is a wide gap for an asset most organisers can produce at almost no marginal cost, and part of the reason is that the thing being sold has never been described in a unit the buyer can check.

What a 30,000 record list actually delivers

Start with the file. A registration list of 30,000 records is a count of rows in a database. It contains addresses that hard bounced last edition and were never removed, addresses that unsubscribed from marketing but stayed on the registration record, role accounts like info@ and registrations@, addresses typed into a mobile form with a missing dot, and people who registered three years ago at a company that no longer exists.

Your sending platform knows all of this. It suppresses most of it automatically, which is why the delivered count is always lower than the record count and why nobody notices until a sponsor asks.

Take 82 per cent deliverability on that file. The send delivers 24,600 messages and 5,400 records receive nothing. Neither party did anything wrong. The rate card simply described the wrong quantity, and once a sponsor has done that subtraction themselves they will discount everything else on the card by the same suspicion.

There is a second reason to price on delivered. It is the only count where your interests and the sponsor's point the same way. If the number on the invoice is the raw file, a decaying list costs you nothing until renewal. If the number on the invoice is delivered messages, list hygiene becomes revenue, and the person who finally removes the 4,000 addresses that have bounced twice is doing commercial work.

Why is the open rate no longer a denominator you can price on?

For years the sensible answer to "how many people saw it" was the open rate, measured by a tracking pixel. Apple ended that as a reliable measurement in 2021. Mail Privacy Protection, introduced with iOS 15, hides the recipient's IP address and downloads remote content in the background regardless of whether the person engages with the message, and Apple's own documentation states that it prevents senders from seeing whether a message was opened.

The practical effect is that every protected recipient registers as an open. Your reported open rate goes up, the underlying human behaviour does not change, and the two are now impossible to separate inside the aggregate.

This does not make opens useless. It makes them a number with an unknown machine component that varies with the mail client mix of your audience, which is different for a US industrial show and a European fashion show and different again year to year. You can still report opens. You cannot price a guarantee on them, and you should not quote them without saying what they now include.

Clicks survived better. A click needs somebody to move a cursor onto a specific link, and while prefetching and security scanners produce some machine clicks, the contamination is smaller and more stable. Unique clicks to the sponsor's destination is the closest thing to a countable outcome a dedicated send has.

The three CPMs, worked

Put the same send through three denominators and price it three ways.

The fee is 4,000 dollars. The raw file is 30,000 records. Deliverability is 82 per cent, so delivered is 30,000 times 0.82, which is 24,600. The open rate on delivered is 24 per cent, so recorded opens are 24,600 times 0.24, which is 5,904. The click rate on delivered is 1.8 per cent, so clicks are 24,600 times 0.018, which is 442.8, call it 443.

Cost per thousand records: 4,000 divided by 30,000, times 1,000, is 133.33 dollars.

Cost per thousand delivered: 4,000 divided by 24,600, times 1,000, is 162.60 dollars.

Cost per thousand opens: 4,000 divided by 5,904, times 1,000, is 677.51 dollars.

Cost per click: 4,000 divided by 443 is 9.03 dollars.

Those four numbers describe one transaction. The first is the one on most rate cards. The last is the one the sponsor's demand generation team will compute the moment they see the click report, because 9.03 dollars a click is a figure they can hold against their paid search costs without translating anything.

The spread between 133 and 678 is the whole argument. A sponsor who was sold on 133 and later works out 678 has not been overcharged. They have been told a number that answers a different question, and the discovery costs you more credibility than the discount you were protecting.

What should the rate card promise?

Write the line as a guarantee plus a report, and keep the guarantee on the count you control.

Guarantee the delivered floor. Something like "one dedicated send to a named segment, minimum 22,000 delivered messages", where the floor sits comfortably under your recent actual delivery. Guaranteeing 24,600 exactly invites an argument about 24,400 in a bad week.

Report opens, clicks, unique clicks and unsubscribes, with the open figure carrying a one-line note about privacy proxies. Do not guarantee any of them.

Name the segment. "The full registration list" and "registrants from the last two editions who opted into partner messages" are different products with different sizes and different deliverability, and a sponsor who thought they bought the first and got the second has a real complaint.

State the send window and the exclusivity, because a dedicated send that lands two hours after another sponsor's dedicated send is worth measurably less and the sponsor cannot see the schedule.

Deliverability is a shared commercial risk, so write it down

Google's email sender guidelines took effect on 1 February 2024 and apply to anyone sending more than 5,000 messages a day to Gmail addresses. They require SPF, DKIM and DMARC authentication, a working one-click unsubscribe on marketing mail, and spam complaint rates reported in Postmaster Tools kept below 0.30 per cent, with 0.10 per cent recommended as the safer operating level.

Read that as a commercial constraint on your sponsorship inventory. If a sponsor supplies creative that produces a complaint spike, the damage lands on your sending domain and degrades every subsequent send, including the ones other sponsors have already paid for. A rate card that sells twelve dedicated sends a year into a list of 30,000 is selling a shared resource that can be exhausted.

Two contract terms follow. First, you approve the creative and the subject line, and you say so on the card rather than negotiating it in the week of the send. Second, there is a stated annual cap on dedicated sends per list, and the cap exists because of the complaint threshold, which is a reason a sponsor can verify rather than an arbitrary scarcity play.

The frequency cap is also what protects your pricing. Sponsored email that lands weekly stops being worth 4,000 dollars long before the deliverability data shows it, because the audience learns to ignore the format.

What to put in the fulfilment line

One row per send, with the segment name, the segment size at the time of contracting, the scheduled window, the actual send timestamp, delivered, bounced, opens, unique clicks, unsubscribes and the destination URL. That row is the evidence for the invoice and the input to next year's price.

Two derived figures are worth keeping across editions. Delivered as a share of the segment size at contract, which tells you whether your list is decaying faster than you are cleaning it. And unique clicks per thousand delivered, which is the closest you have to a stable quality measure for the audience you are renting out. Comparing a send against a signage package or a lanyard on one scale is a different exercise, and putting several assets on a common CPM needs its own denominator discipline.

Where this stops

Delivered messages measure delivery. They do not measure attention, and no arithmetic in this post gets you to attention.

A dedicated send delivered to 24,600 inboxes at nine in the morning on a Tuesday two weeks out is a genuinely different product from the same send delivered on the Friday of a public holiday, and the delivered count is identical. Send timing, subject line quality and the sponsor's own offer probably account for more of the variation in clicks than the size of your list does, and you control only one of those three.

There is also a floor under how precise this can get. If your registration platform and your sending platform hold different suppression states, which is common where an unsubscribe is captured in one system and not synchronised to the other, then your delivered count is right and your addressable segment size is wrong, and the ratio between them will move for reasons that have nothing to do with list health. Reconciling those two suppression states is unglamorous work and it has to happen before any of the pricing above is worth defending.

Pricing an in-app placement runs into a related counting problem, and what a push notification is worth turns on annoyance cost in a way email does not. The same question about whether an upgrade earns its price appears again in the enhanced directory listing, where the measurable outcome is a profile view instead of a click. All of it feeds the same commercial picture that the rest of exhibitor and sponsor analytics is trying to hold together.

This week, take the last three sponsored sends you ran and write down four numbers for each: records in the segment at contract, delivered, unique clicks, and the fee. Divide the fee by delivered and by unique clicks. If either figure surprises you, it will surprise the sponsor too, and it is better that you get there first.