· 6 min read
Newsletter advertising in the week Google changed Search again
Search sent fewer clicks again this month, which makes an inbox look safer than it is. What a newsletter slot is worth now, and how to price one honestly.
Newsletter advertising has spent this year being sold as the answer to a question search kept asking. Every time referral traffic drops, somebody points at the inbox and says: you own that list, nobody can change the algorithm underneath it. The first half of that is true and the second half is not quite, and this month is a good week to separate them, because the search side moved again and the arguments got louder in response.
What moved
Two changes landed in the same few days. Google began auto-migrating Broad Match and automatically-created-asset Search campaigns into AI Max from 1 September, tracked by Search Engine Land, and the click picture underneath it kept deteriorating: about 42% fewer search clicks against baseline, a 61% click-through drop on queries carrying an AI Overview, and zero-click behaviour near 69% of queries, per Search Engine Land's analysis.
Press Gazette's framing is the more useful one for publishers: search is not dead, it is fragmenting. Traffic is not disappearing so much as arriving from more places in smaller amounts, which is a harder operational problem than a straightforward decline. You cannot fix fragmentation by working harder on one channel.
That is the backdrop against which newsletter inventory is being bought and sold right now.
Why sponsors moved money toward inboxes
The argument is genuinely good, and it is not new — it just got more persuasive.
A newsletter slot is a known denominator. The publisher tells you the list size and typically the open rate before you pay. Delivery does not depend on a ranking, a bidding system or a summary box. Somebody chose to receive this, and the sponsor borrows a little of that choice. It is the closest thing in small-scale advertising to buying a defined quantity of attention at a defined price.
Set against display, priced per real visitor, the two land closer than most buyers expect once you count who actually looks rather than who was technically served. That comparison, with the arithmetic, is in newsletter sponsorship vs display ads.
The part the enthusiasm skips
Three things, and a buyer should have all three before writing a cheque.
Rates rise when a channel becomes the consensus answer. If every advertiser displaced from search moves into newsletters at once, sponsor prices go up without the audience improving. Some 2026 rate cards have moved for reasons that have nothing to do with the list and everything to do with demand from people fleeing somewhere else. That is a bad time to be the marginal buyer.
Open rate stopped being a measurement. Automated inbox protections fetch images on the recipient's behalf, which registers as an open by somebody who never read the issue. A quoted open rate is now an upper bound with an unknown amount of machinery inside it. The number to ask for is clicks on the sponsor link in the last three issues — and be prepared for the publisher not to have it.
A list is a channel you do not own either. You own the addresses; you do not own delivery. Inbox placement is decided by somebody else's classifier, and it can change quietly. That is a smaller version of the same dependency search advertisers just discovered, not an escape from it.
None of this makes newsletters a bad buy. It makes them a normal one, which is a better basis for spending money than a refuge.
Four questions before you buy a slot
Ask these of any publisher, and treat a refusal to answer as an answer.
- How many people received the last issue? Not subscribers. Received.
- How many clicked the sponsor link in each of the last three issues? Three numbers, not an average — the variance tells you more than the mean.
- How many sponsor slots are in one issue, and where is mine? A single mid-issue placement is a different product from the fourth item in a block of six.
- What happens if it underperforms? Most honest answers are "nothing", and that is fine as long as it is said out loud before the invoice.
If you are comparing a newsletter against a podcast read, the same four questions transfer with one substitution: downloads for sends, and a promo code for the click count. Both are slot purchases. The differences are in how much of the host's voice comes with the placement, and how long the buying window stays open.
The wider map — cross-promotions, recommendation networks, directories, and where paid fits among them — is in where can you actually advertise a newsletter. Most of the cheapest growth for a newsletter is still not advertising at all.
What we sell, and how it compares
Disclosure: we run BidSurvivor, so this section is about our own product and should be weighed accordingly.
It is the same purchase shape as a newsletter sponsorship — a defined slot, bought in advance, for a fixed window — with two differences worth naming. The board has two slots a day, twelve hours each. The first brand into an empty slot pays $0, no account and no card required, and every visitor starts with $100 of house credit before they sign up for anything, so a first test can cost nothing out of pocket. Taking a slot somebody already holds starts at a $1.00 floor, and no bid can exceed $5,000.
The first difference is that the performance is published afterwards rather than reported privately: opens and click-throughs are live on the front page and on each brand's own page, and every slot's clearing price is visible on the board, in the Hall of Survivors and on the brand page. The Hall runs three rolling boards — 7 days, 30 days and a year — ranked by amount paid. What you paid and what it sent are on the record either way, including when the answer is unflattering.
The second difference is that a slot can be taken from you. Get outbid and you lose the slot and the money; there is no refund. A bid that fails to beat the standing bid costs nothing, so trying is cheap and holding is not. A newsletter sponsorship, once booked, is yours.
Which of those two is better depends entirely on what you are buying. A sponsorship buys a voice the reader trusts, and that is not something an auction can manufacture. A time-boxed slot buys a published number. The unit costs of the latter, in every denomination, are worked out in what a twelve-hour ad slot costs.
Our audience is small and young, which we would rather state plainly than let a rate card imply otherwise. If you need volume this quarter, buy the newsletter. If you need to know what a message does before you pay for volume, buy the cheapest thing that publishes its numbers — ours or anyone else's.