Real-world application of ATL: TV, print, OOH & radio

Oatly's problem, as it moved from coffee shops into British and American supermarkets, was not awareness among the people already ordering an oat flat white. It was awareness among the other ninety-something percent of shoppers who walk past a chiller cabinet and reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → for the carton they recognise. A barista-led distribution strategy had given the brand roughly 32,000 coffee shops and 60,000 retail doors across more than 20 markets by the time it filed to list in 2021. Retail listings are a use-it-or-lose-it asset: if the carton does not move, the buyer reallocates the shelf. That is the brief that put Oatly on posters and on broadcast.

Everything below assumes the mass-reach mechanics the foundations lesson sets out, and the share-of-voice arithmetic the frameworks lesson works through. This is one push followed from brief through site selection to what showed up on the tracker, including the part that went wrong afterwards.

The brief: a carton that has to be recognised at two metres

Oatly's creative has been made in-house since Toni Petersson took over as CEO and brought John Schoolcraft in as creative director, and the house style is deliberately anti-advertising: long copy, no lifestyle photography, no product-in-hand shots, the carton itself doing the work. "It's like milk but made for humans" got the brand sued by Sweden's dairy lobby in 2014, lost the case, and got banned from using the wording there. Oatly published the court documents and kept the fight in public view, which is the cheapest reach it ever bought.

That house style dictates the media. Copy that takes eight seconds to read cannot go on a 48-sheet next to a dual carriageway where dwell time is under two seconds. It needs environments where the audience is standing still: platform panels, escalator runs, bus shelters at a red light, six-sheets inside a station concourse. The brief is therefore not "buy OOH", it is "buy dwell time in postcodes where the product is already stocked".

Site selection: working backwards from the stockist list

The discipline that separates a working OOH plan from wallpaper is the distribution overlay. Before you look at a single panel, you take the stockist file, plot it, and refuse to buy anything outside a walkable or drivable catchment of a shop that has the product. For a brand at Oatly's stage this cuts the buyable universe by more than half, which is the point: half the budget in half the country beats a national smear at frequency one.

Then you build against the audience data. In the UK that means Route, the industry measurement that underpins OOH trading, and it means arguing with the panel owner. JCDecaux, Europe's largest out-of-home operator, sells this inventory, so treat its planning decks as a sales tool: the "reach" number on the page is the panel's total audience, not your target's. Ask for the target-specific figures, ask for eyes-on rather than passage counts, and ask what proportion of the digital screens' loop you actually own. A six-slot loop means you are visible one sixth of the time, and a site that looks dominant in a photograph can be delivering a fraction of the impacts quoted.

Two practical negotiating points that rarely make it into the brief. Fill inventory exists: unsold digital slots in the last fortnight of a month go cheap, and a challenger with pre-approved creative can pick them up. And packs are sold as bundles for the seller's convenience, not yours. Getting a dozen weak sites stripped out of a pack in exchange for accepting a longer commitment is a normal trade.

The broadcast leg and what it costs to be unmissable

Posters build recognition. Broadcast builds fame, and it prices accordingly. Oatly bought a 30-second slot in Super Bowl LV in February 2021, list price around 5.5 million dollars, and ran a spot originally made in Sweden in 2014 in which the CEO sings badly in an oat field. It was widely voted among the worst ads of the night. Oatly then sold t-shirts reading "I totally hated that Oatly commercial". The stunt worked because the brand had a position clear enough to survive being mocked, and because the campaign was engineered to make the mockery repeat the brand name.

Most brands do not have a Super Bowl budget or that tolerance for risk. The realistic comparison for a UK push is ITV, which sells the largest single block of commercial impacts in the market and therefore has its own effectiveness research to promote. A national burst with enough weight to register starts in the low hundreds of thousands of pounds; below that, regional TV or radio drive-time on a two-city plan will do more for you than a thin national schedule. Radio deserves more attention than most CMOs give it: it tolerates high frequency without wearing out, and it reaches people in the car on the way to the shop that stocks you.

What moved on the tracker, and how fast

Set expectations before the flight, because unaided answers move slowly. Spontaneous awareness, the unprompted "which brands come to mind" question, typically shifts by single-digit points off the back of one burst, sometimes one or two. Prompted awareness moves faster and flatters the campaign. If your board sees only the prompted number, they will conclude the campaign was twice as good as it was, and they will expect the same again from a smaller budget.

What the Oatly case shows is the compounding, not the single burst. Revenue went from roughly 204 million dollars in 2019 to about 421 million in 2020 and around 643 million in 2021, across a period of continuous poster and broadcast presence plus a very high earned-to-paid ratio. Read the direction of travel, not the individual flight.

Three measurement habits worth adopting: run the tracker in a control market with no ATL weight, so you are comparing against something; separate spontaneous awareness from spontaneous consideration, because a brand can become famous and still not get bought; and keep the question wording frozen for years, since changing it destroys your own back series.

CMO action items

• Put the stockist file in front of the media planner before the brief is written, and refuse any panel outside a catchment where the product is on shelf.
• For every digital OOH line on the plan, ask for slot share, loop length and target-specific eyes-on impacts. Reprice anything where you own less than a fifth of the loop.
• Fix your spontaneous awareness question and your control market now, this quarter, so the next campaign has a baseline that predates it.
• Have legal read the broadcast claim, not just the creative. Substantiation, not tone, is what gets ads pulled.

Where this goes wrong

Demand arriving before supply. Oatly's fame outran its factories: through 2021 and 2022 the company was reporting capacity and execution constraints in the US while its advertising kept pumping. An out-of-stock carton converts a hard-won first purchase into a trial of a competitor, and the shelf space goes with it. The IPO valuation of roughly ten billion dollars in May 2021 lost the large majority of its value inside two years. Reach is the easy part to buy and the expensive part to waste.

Building the category instead of the brand. Generic "plant milk is better" messaging hands the growth to whoever owns the most shelf, usually the incumbent or the supermarket's own label. Oatly avoided this because the carton, the typeface and the voice are unmistakably its own, so a shopper who half-remembers an ad reaches for the right pack. Without that brand linkage, a challenger's ATL spend is a subsidy to the market leader.

Claims that cannot be substantiated at broadcast scale. In January 2022 the UK's Advertising Standards Authority upheld complaints against several Oatly ads, including a TV spot, finding that comparative claims about the climate impact of dairy and meat were misleading as presented. Above the line, a claim is read by regulators, competitors and journalists at the same moment as by consumers, and a ruling becomes the story.

Measuring it like a click. Do not put a QR code on a poster and judge the campaign by scans. Judge it on unaided recall, brand-linked recognition and market-level sales, ideally with a tracker from a specialist such as Kantar or Ipsos. If you measure a poster like a paid social ad, you will conclude posters do not work, and you will be wrong.