Newsletter Sponsorships Bought on Clicks, Not Open Rate

Open rate has been unreliable since Apple Mail Privacy Protection began prefetching images in September 2021, and on lists heavy in Apple Mail the reported figure can run twenty to forty points above reality. We buy on unique click rate against the sponsor link and on what those clicks do afterwards. A 38,000-subscriber list producing 240 unique clicks on a sponsor slot is a worse buy than a 9,000-subscriber list producing 400, and the second one usually costs less. Vetting a list before we place anything: A screenshot of the beehiiv, Kit or Ghost dashboard showing subscribers active in the last 90 days, not total subscribers. A growth source breakdown, because a list built on giveaway co-registration and paid recommendation swaps behaves nothing like one built on the writing. The last six sponsors and what they paid, which good writers will tell you. And deliverability: SPF, DKIM, a DMARC policy, one-click unsubscribe, and a Google Postmaster spam rate under 0.3 percent, per the bulk sender requirements Gmail and Yahoo began enforcing in February 2024. A list failing those loses inbox placement mid-campaign, and your slot goes with it. The slot: A primary sponsorship above the first section break, 90 to 140 words, one link. We draft it, the writer rewrites it in their own voice, and we sign off on the claim rather than on the sentences. Text first, because image-only ads are still blocked by default in enough clients to matter, and because a graphic in a text letter reads as an advertisement before it reads as anything else. No dedicated sends unless the writer has run them before and their readers have not been trained to ignore them. Where this works: Developer tools placed with writers whose readers ship code, and kids-and-family products placed with parenting writers whose readers already have the problem you solve. Both categories have a small number of lists that matter and a long tail that does not. The work is knowing which is which, and being willing to tell you that a list you asked about is not worth buying. Cadence: Single slots underperform in this channel. We plan three placements in the same letter across six to eight weeks, because click rate on the second appearance runs above the first in our own placements, and the third is where readers start using the product's name unprompted in replies. If your budget covers one slot, buy a different channel. What we refuse: CPM priced against total subscribers. Sponsorships in letters that do not label sponsored sections. Lists whose growth is mostly paid recommendations. Any placement where the writer has not used the product. Requests to reuse the subscriber list afterwards, because you are buying attention inside someone else's letter and not their audience. Newsletters written for children, which is not a category we will place into at all. Reporting: Unique clicks and click rate measured against the writer's own trailing average rather than an industry benchmark. Sessions and their behaviour on the landing route. Signups and revenue attributed through a destination URL unique to the placement. And the writer's unsubscribe rate for that send against their baseline, which tells you whether the placement cost them something, which in turn decides whether you get invited back next quarter. Not a fit: Anyone who wants a spreadsheet of every newsletter in their category with rates attached. Anyone who needs the placement to look like a display ad.