Current answers to the marketing questions consumer businesses ask. Sign in free to read them all.1,898 sources, at least 3 behind every answer.
The deadline moves, the terms don't
All week the countdown at the bottom said "the new terms start October 30." On Friday Meta moved the date. This is the week-in-review: Friday's news, what changed in the Answer Book, and the three stories worth keeping.
By Scott Stroud, Editor
A Meta spokesperson told Reuters that the updated Commercial Terms and Self-Serve Ad Terms now take effect November 20, not October 30, "to allow more time for us to explain to our partners and stakeholders what's changing and what's not." Meta didn't say it would change any wording, and says it will make clear that site owners can still use robots.txt. Hours later, Meta's UK preview page still said October 30 (PPC Land).
So the contract is the same, with three more weeks to read it. That's the crawling clause from last Saturday's issue and the AI liability clause from Friday's, both still in there. Four of our answers now say November 20, including our answer on Meta's creative enhancements and AI features and our answer on letting Meta's crawlers read your pages. The advice in them is unchanged: use the extra time to finish the unticking.
Also Friday
- Northbeam's Bryan Bumgardner says returns slipped as the Prime Deal Days rush settled, and that "in the first week of November, all of this will tank as spend skyrockets and conversion behavior freezes up" (Bumgardner). A vendor's forecast with no figures attached. Our answer on seasonal budgets still says to hold or raise spend through early November, so this one is worth watching.
Google joins the book
Prevailing now covers Google paid search as well as Meta. The first five Google answers went up this week:
- Performance Max alone or alongside: keep brand and exact-match Search plus Standard Shopping for the products you must control, and drop Shopping only if a Google Ads experiment shows a clear profit win.
- Choosing a Google bid strategy: bid for volume first, then test Maximize conversion value at about 50 conversions in 30 days.
- Spending on your own brand name: keep it to a small defensive floor. The answer finds Google reports about 4 to 10 times the sales brand search actually adds.
- Testing whether Google spend is incremental: run a geo holdout read on total store sales, and test brand, non-brand and Performance Max separately.
- Product feed changes that pay: rewrite product titles first, with brand, product type and the attributes people search for in the first 70 characters.
What changed in the book this week
Two calls changed. On Thursday, Meta's third quarter moved our answer on benchmarking your CPMs: a normal rise now tracks Meta's own price per ad, up 23% in the US and Canada, so look at your account only well above about 25%. On Wednesday, our answer on tying repeat orders to the first ad dropped an "about 30 customers" floor that no dated source supports.
New Meta answers this week, besides the Google five:
- Comparing results across 2026 reporting changes: treat January 12 and the March switch to link clicks as breaks, and compare on backend orders.
- Designing a lift or geo test: set the pass bar at your breakeven lift before launch, and hold out at least 15% of markets.
- Ramping spend on a new account: budget at your planned level, verify payment early, and expect very high CPMs at first.
- Funding ad spend as you scale: on monthly invoicing, keep the credit limit above your peak unpaid balance.
- Usage rights in creator contracts: license creator content instead of buying it, and name every platform and ad type.
- Vetting an agency or freelance buyer: give every candidate the same brief and month-three targets.
- What a Meta rep can do for you, a separate new-customer purchase event, and the crawler answer above round out the week.
The week in three
The price went up. Meta's US and Canada price per ad rose 23% on last year, and that growth has climbed every quarter of 2026: 14%, 20%, 23% (PPC Land). Thursday's issue had the details. If your CPMs feel worse, Meta's own numbers agree.
Most ads lose. Common Thread's 1.339 million ads had a median loss of about $3.20 each before production costs, with the top 1% producing 41% of the value (CTC). Friday's issue covered the catch: more ads only pays if each one is cheap to make. Our answer on monthly creative volume already leaned that way.
The terms story ran all week: last Saturday's read of the crawling clause, Monday's issue with Meta disputing parts of that reading, Friday's untick-every-box stories, and now a new date. The text hasn't changed, so the homework hasn't either.
Product browsing turns on October 28, 4 days from now. The new terms now start November 20, in 27. Black Friday is 34 days out.
There is no Sunday issue, so see you Monday.
Every fact links to its source. Opinions are the writer’s, built from the sources linked beside them.