Why adapting assets for social wastes money

Cutting the hero film down for social is the most reasonable decision on the production schedule. The idea is approved, the footage is paid for, and a fifteen-second 9:16 version costs almost nothing against the shoot. It is also where the money goes. The viewer's test on a skip medium is relevance, relevance lives in the situations, people and environments in the frame, and those are local by nature, so a central asset adapted for markets is the same ad failing the same bar in every language, at auction prices, in every feed it reaches.

The cascade

The global toolkit was a real achievement. One shoot at the centre, then master films, key visuals and adaptation guidelines cascading outward through a portal where every market localises within the lines. It solved consistency and cost for a world where an ad was a fixed artefact and localisation meant fitting it to local specifications, and its quiet assumption was that an ad travels. Social expired the assumption and the cascade kept running, and the result sits in feeds all over the world, the same cutdown immaculately translated into forty languages, native to none of the forty markets, skipped fluently in all of them.

What adaptation cannot insert

Feed-native work earns attention through recognisability, real people in real environments, the texture the viewer's own world is made of, and that texture is local by definition. The kitchen that says home in Bangkok does not say it in Birmingham. The light, the streets, the gestures, the pace of speech, none of it cascades, so a centrally shot film carries the texture of exactly one place and every other market receives it as foreign content with subtitles, classified in two seconds and swiped. Category entry points break the same way. The after-work moment, the school run, the weather that triggers the purchase, are culturally shaped, and a master built on the centre's idea of the moment misses the versions that occur in each market. Adaptation cannot insert what was never shot. Jenni Romaniuk's distinction between the codes, which must be consistent, and the executions, which must vary, applies unchanged to geography.

The evidence travels, the asset does not

The finding that the same creative does not travel comes from vendors and platforms with different interests and all of it points the same way. Ipsos, which sells creative testing, reports from 3,122 ads that the same creative does not perform equally across formats. TikTok's own European creative analysis, as cited by Tom Roach, has native creative consistently outperforming repurposed and adapted ads, and TikTok Marketing Science puts TikTok-first ads at three seconds more watch time and 25% higher completion than repurposed ones. Snapchat's work with OMD and Amplified Intelligence found bespoke ads earning 1.4 times the attention of repurposed ones. In Facebook's 2016 Bose case a ten-second vertical cut optimised for mobile delivered 20% more completed views and ad recall six points higher than the original commercial. The waste never appears on the production invoice, it appears downstream, in media spent behind an asset the feed skips in every market, and the watched rate is where it shows.

Hold the codes, make the work

The central team's instincts are correct. A global brand must be one brand everywhere, a single shoot amortised across many media budgets is a real saving, and a compliance failure in one market is a global problem the moment it travels. None of that argues for central production, it argues for relocating where consistency lives. Consistency belongs at the code level, the distinctive assets, the colour world, the voice, the claims discipline, codified once and enforced on every asset before any media spends. Creation belongs at the market level, local creators producing native work inside that system, local versions of the category's moments, made by people who live in the culture rather than researching it. The brand system travels perfectly because it is information. The production does not have to travel at all. When to use brand ads, UGC, influencers and creators draws the line between creators as production talent and influencer work, and Where AI belongs in a creative system covers the versioning and localisation off real footage that a machine can take on.

The global toolkit

The counter-position is how most global brands operate and it was, for its era, the right answer. One idea, produced to a standard no local market could afford, protected from drift by guidelines and a portal, amortised across every media budget in the company. Procurement can price it and the global brand team can sleep. Where it stops is at the feed's relevance bar, judged locally, in two seconds, by people who know what their own lives look like and can tell when an ad was shot somewhere else. Hold the codes globally. Make the work locally. In eighty markets, working means being recognisably from here, eighty times. Media creates the opportunity to see your ads. Creative determines whether they work.

What to do with this

Sort last quarter's ads into two piles, those cut down from a master and those built for the feed, and compare their stop and watched rates on one platform with one threshold. The gap is the cost of adaptation in the brand's own numbers, and multiplied by the media behind the cutdowns it is the size of the opportunity. Write the brand system down as a scoring standard and put every asset through it before media attaches. Brief the next round around the category's moments as they occur in each market, with people from that market in frame, and keep the hero film for the job it was shot for. What the feed is testing when it decides to stay or leave is the subject of What high quality creative means on paid social.

Questions

Isn't a 9:16 cutdown of the TVC good enough for social?

It fits the frame, which is the only thing adaptation can change. The things the feed rewards, a recognisable person, a familiar situation, an opening built for a viewer who can leave, were decided at the shoot, and a cutdown cannot insert what was never shot. Evidence from Ipsos, TikTok, Snapchat and Facebook's own cases has work built for the feed outperforming work repurposed for it on watch time, completion and recall, so the cutdown is cheap on the production line and expensive in the media line.

How do we keep a global brand consistent if every market makes its own ads?

By moving consistency from the asset to the code. Distinctive assets, colour, voice, claims discipline and the brand's category entry points are information, they travel perfectly and can be written down as a standard every asset is scored against before it spends. Under a cascade the centre found out what markets did with the master when the reports came back. Under a brand system it sees every asset, every score and every failure, market by market, before any of it runs, which is more control rather than less.

Doesn't local production cost more than one central shoot?

Per shoot, yes. Per watched impression, usually not, because the cascade's costs hide downstream, in adaptation layers, regional approval cycles and media spent behind creative the feed skips. Feed-native production by local creators is modest relative to a hero shoot, and the comparison that matters is all-in cost per watched impression against the brand's own adapted work on the same media, which any account can run from its own Ads Manager data.

Sources

• Ipsos, Rachel Rodgers, "How to stop wasting money on 'efficient' digital advertising", 11 February 2022. The same creative does not perform equally across formats, 3,122 ads. Ipsos sells creative testing.
• TikTok Marketing Science, TikTok-first ads against repurposed ads (watch time, completion), and TikTok Marketing Science EU proprietary creative analysis (UK, France, Germany, Spain, Italy), both as cited by Tom Roach, "The 3rd age of effectiveness", circa 2023. Platform funded, TikTok sells the media, primary not located, quoted as cited.
• OMD, Amplified Intelligence and Snapchat, bespoke Snap ads at 1.4 times the attention of repurposed ads, as cited by Tom Roach. Platform funded, Amplified sells attention measurement.
• Facebook for Business, Bose case study, 2016. Completed views and ad recall for a mobile-optimised vertical cut against the original commercial. Platform case study, Facebook sells the media.
• Jenni Romaniuk, Building Distinctive Brand Assets, 2018. Consistency of codes, variety of executions. Ehrenberg-Bass Institute, sponsored by large advertisers.