The Weekly Marketing Report

A capsule by designer Megan O'Cain, launched with a runway film and a two-day pop-up at Null Object — a lime green bag reading air-mess, a zhee-vahn-shee jacket. What makes it work is the research, not the clothes: a YouGov study found 62% were confident they could pronounce Hermès and 15% could. That gap is Duolingo's entire product category, worn. Two mechanics worth stealing — entry was bidding your Duolingo streak, which makes the activation's currency the brand's own retention metric; and it is fashion-adjacent, not fashion-competitive, so the pieces are props for an argument rather than clothes to be judged.

A campaign built on the people who make things by hand and by craft, nine days after Snickers' AI spoof took Muse by Clio's Ad of the Week. It responds to AI fatigue rather than to AI — the audience's feeling about the tool, not the tool — and it is the second named case, which is what sharpens the tracker's anti-polish entry into something specific: anti-AI, not anti-slick. Note who is saying it: a company that sells design software, positioning against the automation of its own customers' work.

An episodic workplace series with Jack Quaid as a “weekday warrior consultant”, running across digital and social all season and built around the Weekday Warrior Pant. CMO Joe Berean: “Jack brings such a unique personality and is able to represent that person who can be calm in the midst of the storm.” This is the second new brand to enter the micro-drama format in four weeks — Nuuly on 19 August, Bonobos now, Cadbury the next day — and dated properly the three are a sequence, not a cluster. No audience figure was published.

A multiyear generosity platform launched into fashion week with five objects: a Swarovski-studded insulated clutch for XL bars, debuting at NYFW on 14 September, only five made, entry to win on the 15th. Around it a 30-second hero directed by Tom Dream, an episodic social comedy following a “Department of More” intern, and partners Luar and Nylon. The object is not the product; it is the reason the press has a photograph. Hershey projects a 30% year-on-year increase in brand investment for 2026–27 — a plan, not a spend, and logged as such on page eight.

A mascot invented from nothing for a sports franchise — the fourth case on the tracker's mascot entry, and the first that is not a retail brand. It moves the entry to Building on count, but the test that matters is still The UPS Store's Blu (No. 36): a character with an agency of record and a stated job. Mascots are cheap to launch and expensive to sustain; the twelve-month check is logged.

Amazon advertisers can extend campaigns into ChatGPT, with ChatGPT Ads inventory available through Amazon's DSP and Delta Vacations among the first US pilots. ChatGPT ads reached a ~$1bn annualised run rate late last month. Read against the FTC complaint in No. 36: in eleven days Amazon's advertising business was accused of running an undisclosed auction floor and opened a new inventory surface inside the fastest-growing consumer AI product. One counterparty, largely unobservable from outside.

Nuuly on 19 August, Bonobos on 10 September, Cadbury on 11 September — fashion rental, menswear, confectionery. In the original catch-up edition all three were placed in mid-August and read as one undifferentiated cluster. Dated properly they are a sequence: a new brand entering the format roughly every fortnight. That is a different finding, and the only one of the two that is true. The format is an ordinary line item now — and still not one of the three has published an audience figure.

Duolingo into fashion (8 Sep). OLAPLEX's first hairspray — and note how the claim had to be welded to the brand's only story: 96-hour hold and 60% less breakage, a styling benefit and a repair benefit sold as one line, because OLAPLEX cannot sell a benefit that is not repair. MAKE UP FOR EVER's first lip balm, a pro-artistry brand arriving in a category it would have dismissed five years ago. Vaseline's reactive rebrand. None of these is about a gap in the market. All of them are about a cult asset being asked to carry more products.

Four cases in five weeks — Crocs’ Niles and Stanley’s toolbox in W33, The UPS Store’s Blu in No. 36, and now the Portland Fire’s Rose (9 Sep), a sports mascot rather than a retail one. Building on count. But the real test is Blu’s: the only one with an agency of record and a stated function. Mascots are cheap to launch and expensive to sustain; the twelve-month check is logged for 2027-09-11.

The entry sat at Emerging on a single brand for six weeks while the tracker asked whether a second would appear. The answer arrived from the other direction: WONDR discontinued its packaging-free waves and bars (11 Sep). Meanwhile Mondelēz did the thing that actually scales — not removing the packaging, replacing the material, with FSC-certified paper. Paper substitution ships; format abolition does not.

The entry closed in No. 34 as unmeasurable — cases without numbers. This week produced three more cases and three more absences: Bonobos published nothing, Cadbury published nothing, and Glossy reported fashion brands launching podcasts “to add more depth to their communities” with no download figure attached. A closure that keeps being confirmed by new cases is a stronger result than one confirmed by silence. Reopening requires a number.

Inditex: revenue +7.6% to €19.8bn, gross margin a record 58.7%, autumn trading +9% — and the shares fell on a profit miss. Follows Lululemon and PVH. The gap between operating performance and market reaction is the cleanest illustration this series has of what margin-over-growth costs a business that has already optimised both. Building, on three named cases.

The replacement entry is deliberately narrower and has exactly one case and exactly one number: Maybelline's Lifter Gel grew roughly 10× in weekly units between Love Island USA's 2 June premiere and its 12 July finale, measured by commerce-intelligence firm Daash. No new case this week; the entry stays at Emerging on one case and one figure. Borrowing an audience is measurable because somebody else already counted it. Building one is not — which is precisely why nobody publishes.

Opened in No. 34 on four founders publicly seeking a buyer in one trade roundup, and has produced nothing in three weeks. It closes as a single roundup — the shape of a story that was real for one day. Highstock's $30m Series A for a surplus-beauty-stock marketplace is adjacent evidence that the underlying pressure exists, and it is deliberately not used to rescue the entry: a funding round for a liquidation marketplace is not the same claim as founders soliciting buyers in public.

No. 36 found that five of eight Reklamombudsmannen rulings failed on the distance between a categorical headline and its qualification, across five media. No new RO decisions fall inside this window, so the promotion condition is not met and the entry is carried at Emerging. A pattern inside one caseload can be an artefact of that caseload; the next batch decides it.

No prestige fragrance house response has been detected in seven weeks. W33 said the question should stop being asked; this issue stops asking. A tracker that only adds entries is a list, not an instrument.

The default assumption is that social is an interruption channel: you buy attention from someone who was doing something else. The Grand Prix entry describes the opposite premise — that for this audience and this subject, social is where people go to look things up, and the job was to be findable there rather than visible there. Built from a deep read of the target group's media behaviour, for an audience described as advertising-sceptical, on a personal and sensitive subject. (ACO Spotless is a blemish range; the sensitivity follows.) Three things worth stealing: a behaviour as the insight rather than a demographic; the audience's scepticism treated as correct and designed around; and a success measure of being found at the moment of a private question — a search metric applied to a social budget. The most useful Swedish campaign this series has surfaced: in Ellen's category, in her market, this year, agency named.

The EU's TTPA transparency rules took effect in October 2025, requiring every political ad to carry its payer, cost, election and targeting. Meta and Google both concluded compliance was not worth building and left the category across the EU. So the 2026 Swedish campaign — the first fought under the rule — was fought without paid social. What replaced it was old media that still sells reach, and organic distribution off a follower base that has to be built years ahead. Vänsterpartiet alone bought 1,500 outdoor boards for 2.9 MSEK, plus 798,062 SEK of metro and bus space for 7–13 September. Socialdemokraterna is reported to have spent around 115 MSEK on advertising space across 2026 — roughly 29 MSEK outdoor and 24 MSEK television — single-source, Kantar Sifo net figures via Nyheter24.

Gold at StockholmMediaAward for a media-led campaign from the market's largest fast-food advertiser, with the creative agency and the production partner both credited. Less instructive than the Grand Prix — the citation says less about the mechanism — but it is a named Swedish campaign with an agency attached, which is the thing this series had failed to find for six editions.

A Swedish beauty retailer moving into gaming with an esports partner, and taking Silver for it. The single most job-relevant item in this edition. Kicks is in Ellen's exact category — and it is also the brand Matas flagged as trading materially worse late in 2025, revenue −4%. A business under commercial pressure doing award-winning work is a specific and unusual thing to be able to talk about, and the kind of observation that shows a candidate reads the market rather than the press releases.

A production office in Casa de les Punxes, Avinguda Diagonal 420 — H&M's thirteenth, focused on trend identification and new product development, opened two days before Inditex reported a record half. A production office that exists to spot trends faster is not a supply-chain decision dressed as a marketing one; it is a speed-to-trend play, sited inside the competitor's talent pool.

W33 left "does the mid-price squeeze hold for a third month?" as a W37–38 question. It is a W38 question: the official plan puts August 2026 data on 16 September. And the figures search returns for "Stilindex augusti" — clothing +6.3%, shoes +11.9%, third consecutive month — are the August 2025 release of 15 September 2025. This is the second time the same trap has been caught, both times by the same fix: read the primary release before quoting it.

Announced 26 August (No. 35); this is the week it actually opened — 600+ stores and Target.com, 1,600+ products, 90 brands, more than two-thirds of them new to Target, in the roughly 1,000 square feet that 610 Ulta shop-in-shops had occupied since 2021. The largest single change to American mass beauty distribution this year. The tracker's retailer as the audience owner entry stays at Emerging: this is the announcement's execution, not a second case.

Revenue +7.6% to €19.8bn, net profit +6.8% to about €3bn, gross margin 58.7% (+40bp), autumn/winter trading 1 August – 7 September +9% year on year — and the shares fell on a profit miss. The number to carry into an interview: the gap between operating performance and market reaction is what “margin over growth” costs a business that has already optimised both. H&M's answer, two days earlier, was a product-development office in Barcelona — inside its competitor's talent pool (page six).

The first US sunscreen with bemotrizinol (BEMT) — the first new UV filter approved in the US in more than twenty-five years. Regulatory news wearing a product's clothes: the interesting fact is not the SPF but that a category frozen by regulation for a generation has just moved. Expect every prestige suncare brand to have a BEMT product in development; first-mover advantage here is measured in months. Also on the list: MAKE UP FOR EVER's first lip balm at $26, La Prairie's $320 serum, Bubble's $13 hypochlorous spray.

An immersive takeover of the Saatchi Gallery for London Fashion Week. The pattern to note is that the venue is doing the repositioning — a waxed-jacket brand buys contemporary-art adjacency for a week. Too early to read; no attendance or coverage figure yet.

Clothing +6.3%, shoes +11.9%, third consecutive month of growth. Real figures — the August 2025 release. And a second decoy surfaced this week: a Svensk Handel press release of 12 September 2024 carrying August 2024 data — clothing +2.1%, e-commerce +10.8%, stores +1.8%, complete with a Sofia Larsen quote. August 2026 publishes on 16 September 2026. Two distinct wrong years now circulate under “Stilindex augusti”; anyone quoting either is a week early and a year or two late. Third catch, same fix: read the publiceringsplan first.

The idea is well evidenced as an idea — a Grand Prix at a juried award, the entry description published. Its effect is not. No performance figures, budget or methodology have been made public, and an award entry is a document written to win an award. That makes it the fourth campaign of that shape in five weeks: celebrated, and unmeasured in public. Watch-list item four asks whether anything comes out.

Hershey projects a 30% year-on-year increase in brand investment to fuel Cadbury's 2026–27 platform — in a period when most of the evidence on this board points at cost discipline, a company betting the other way. A stated investment increase is a plan, not a spend. Watch whether it holds through the next two quarterlies.

Around 115 MSEK of advertising space across 2026 — roughly 29 MSEK outdoor and 24 MSEK television — per Kantar Sifo net figures as reported by Nyheter24. Single-source, and carried as such on page one. Vänsterpartiet's figures, by contrast, come from the party's own campaign page and are primary. The displacement argument built on both — that political spend crowded commercial outdoor inventory in the run-up to 13 September — is reasoned from party spend, not sourced from rate-card or occupancy data, and is watch-list item three.

Items resold through Vinted are said to generate five times the waste of charity-shop resale. No methodology seen, single source, and the source competes with the platform it measured. If it holds up it is one of the more consequential claims in resale; until the method is published it is a press line. And it is campaign collateral: Oxfam launched Second Hand September on 1 September (No. 36), and this claim arrives ten days into it from the same organisation.

Logged W32 on a single commercially interested source and never corroborated since. PepsiCo's move into the fresh perimeter is adjacent — processed to fresh, not food to beauty — and does not count. Six weeks is the point where the entry gets closed the way dupes was. W38 closes it unless corroborated.

Cross-Border Commerce Europe's 2026 ranking puts the top 500 online fashion, beauty and luxury retailers at €120bn in 2025, €79bn of it cross-border, with beauty outperforming on repeat purchase, engagement and lower return rates. Plausible, directionally consistent with everything else on this board — and reported once, with no series behind it and no Nordic breakdown. Emerging, not substantiated.