Every active GB ad, where it lands, and the one sentence running underneath all of it
We pulled the complete active GB library from Meta's Ad Library, two independent queries converging on the same 30 entries. All 30 are video, all 30 carry a 40% code, and all 30 land on a single URL. What looks like a 30-ad creative programme is one primary text with 30 tracking codes, and the creative variation that does exist lives inside the videos, not in the angle.
28 of 30 ads run byte-identical primary text. The only variable across them is the code: GN40MBRS in the brand lane, GN40MCHI and GN40MLAP in the creator lane. That means the library is currently set up for attribution tracking, not angle testing, which is a big lever sitting unused.
Every active ad points at greenchef.co.uk/pages/resolution. Rendered live, it is a dressed plan-picker: goal tiles, a press strip, three one-line testimonials, then 40% pricing and a Build plan button. Roughly 490 words of visible text.
It is a good-looking page, and it is doing the job of a plan selector with a coupon: pick a goal, see the price, build the plan. What it does not do yet is build belief first, which is the job a presell layer takes on. There is no editorial step anywhere in the funnel, so the ad has to carry the entire argument in a video, and the page picks the reader up already at the pricing table.
That also means every angle shares one landing experience. A high-protein viewer, a calorie-conscious viewer and a lower-appetite viewer all arrive at the same tiles, so the specific reason each of them clicked is not carried through to the page.
We swept HelloFresh UK's GB library the same night as a cross-check: 56 active entries, 51 video and 5 static, every ad carrying a 50%-off first-box code, and 53 of 56 landing on hellofresh.co.uk/pages/resolution, the same /pages/resolution template Green Chef uses. Same landing handle, same coupon mechanics, across both brands.
| Signal | Green Chef UK | HelloFresh UK | Factor US (sibling) |
|---|---|---|---|
| Active ads mapped | 30 | 56 | 670 |
| Statics in the library | 0 | 5 | 267 |
| Discount-led | 100% (40% code) | 100% (50% code) | 67% |
| Single-destination concentration | 100% to /pages/resolution | 95% to /pages/resolution | 82% to /plans |
| Problem-aware creative | 0 ads | minimal | 11 ads |
| Evergreen creative | none past 28 days | none material | 655-day static, 42-day publisher lane |
The useful read: the sibling brand in the US has already proven that a durable static lane can carry volume for months. On the UK side that lane does not exist yet on either brand, which makes it the cleanest piece of open ground in the account.
Directional. The exact angles, the order, and the objections each page answers come from proper per-angle research.
The active GB library is 100% video, 0 statics, while Factor US runs 267 active statics as its durable volume layer. Cheap, always-on static volume is a channel nobody is buying here yet, and it is the fastest thing to stand up because it does not depend on anyone else's roadmap.
28 of 30 ads share one sentence, so the library is testing codes rather than angles. Organic, convenience, high-protein, appetite and results are all unbought positions, and the two ads that do carry a real angle are the two that have lasted longest.
Every angle currently shares one plan-picker. A congruent presell for each angle, each mapped to an awareness level, would carry the specific reason someone clicked all the way through to the plan, instead of handing every viewer the same tiles.
Zero ads currently open on a problem before the offer. That is the largest and cheapest audience for any of these angles, and it is the layer that lets the account grow past the people already shopping for a meal box.
