One campaign, 36 languages: How we stay on brand across Black Friday using Adobe Brand Intelligence.
11-19-2026
As for many companies, Black Friday is one of Adobe's biggest campaigns of the year, and every year my international marketing team has the same job: take one big idea and make it feel local to shoppers in Munich, Seoul, Dubai, and dozens of other markets. Doing that means producing creative assets at a huge scale, in 36 languages, without losing what makes Adobe look and sound like Adobe.
This year, that job got even bigger. The campaign’s total paid media deliverables add up to more than 57,000 assets, nearly double the 31,000 we produced in 2025 and more than three times the 16,000 produced in 2024.
That growth largely came from rolling out creator and community platforms – YouTube, Reddit, and Pinterest – to more regions, as well as extending last year’s U.S. Firefly offer globally.
Adobe’s Holiday Spending report shows shoppers starting earlier, with online spending in October expected to rise 8% year on year as retailers’ pre-season sales events reshape the season. Black Friday is forecast to grow faster than Cyber Monday, as earlier discounts pull shoppers in. And social media and affiliate partners, including influencers, are expected to see the biggest gains in share of revenue, up 17% and 12% respectively, compared with 4% for paid search. That forecast covers the U.S., but it points to something we feel across markets. In the UK, Adobe forecasts a record £28.52 billion in online spend across November and December, and traffic to UK retailers from generative AI tools is projected to surge 214% year on year over the festive period.
For marketers, the takeaway is clear: influencers, affiliates, and social are key drivers of holiday success, for Adobe as much as for any other brand. They are also the channels where we produce the most creative, and they need a constant supply of fresh assets built for each platform. The real difficulty, though, is that with compounding asset volume comes compounding challenges with maintaining brand consistency. And with shoppers’ attention split across a myriad of channels, and a growing “sea of sameness” deriving from AI-based content generation, there is more pressure than ever for brands to stand out.
To meet that demand, we used Adobe Brand Intelligence, the brand system we introduced at Adobe Summit in April, alongside Firefly Creative Production for video. This is how it went.
What 57,000 assets looks like.
Behind that number is a matrix that no team could build by hand in a reasonable time:
- Reach: 6 regions, 74 markets, and 36 languages, including a dedicated White Friday campaign in the Middle East and North Africa.
- Offers: Creative Cloud, Creative Cloud Teams, the Student and Teacher Edition, Firefly, and Acrobat.
- Formats: roughly 38,000 static assets and 19,300 videos.
- Channels: Affiliate, Display, and social on Meta, Snapchat, TikTok, YouTube, Reddit, and Pinterest, plus connected TV.
EMEA accounts for the biggest share, with nearly 75% of the total assets. APAC and Japan together account for just under a quarter, and North America and LATAM make up the rest.
The expansion of YouTube, Reddit, and Pinterest to more regions brings us incremental reach across creative communities. Together, these three channels account for more than a quarter of the global assets. Platform specifications also change on the platforms' schedule, so staying current is its own job regardless of whether it’s a new or old channel.
Why staying on brand gets harder the further you travel.
Localization at Adobe doesn’t run on a single-solution model and what I mean by this is that there is no master English layout that every market adapts itself to. Each market brings its own inputs. Some are typographic such as German and Finnish running long, so the layout has to give somewhere; Korean needs its own line spacing; Arabic and other right-to-left languages require rebuilding a layout while respecting the design system. Others come from regional teams who know what lands culturally, which tone reads as credible, and where a literal rendering of English would fall flat. The result is 74 markets worth of input shaping the campaign’s success.
Our AI innovations had already helped eliminate much of the repetitive production work over the past two years. What they hadn’t addressed was the challenge of managing workflows across multiple different systems. Localized copy arrived from our global localization team in one system. Artwork was built in another. Delivery was tracked somewhere else, and review happened somewhere else again. Every handoff was a chance for a version to go stale or a comment to get lost, and the only way to catch a mistake was for a person to spot it. And every creative is only half the job: those assets then have to reach each channel, in the right spec, at the right moment.
How Adobe Brand Intelligence changed the way we work.
Brand Intelligence learns what a brand stands for from its guidelines, its approved assets, and — importantly — the decisions people make in review: the comments, the design changes, the approvals. It then applies that knowledge at scale through three skills:
- Instruct to Assemble builds the assets. It takes approved masters and the campaign brief and assembles the finished variants.
- Validate checks each asset against brand guidelines before it goes anywhere.
- Simulate lets synthetic audiences react to creative before full production, so teams can test ideas before committing resources.
How did that look in practice? Our Creative team supplied master templates for every size across each offer, including bespoke layouts for India and right-to-left languages, along with the components that fill them: headlines, body copy, calls to action, logo placements, and the campaign's artwork, built around an expressive take on the Adobe “A” logo and created by 20 artists we commissioned from our global creative community. My team specified every variant the campaign needed. Brand Intelligence assembled the static assets, already localized and formatted, and checked them without a designer having to rebuild each version by hand, while Firefly Creative Production scaled the video.
The regional checks needed real work, and that work moved to the front of the process. Our Creative teams wrote regional creative guidelines and best practices so Brand Intelligence could check each asset against its own market's rules, rather than applying one global standard everywhere. Instead of troubleshooting after assets were built, we put the rules in place first, which leaves our people to make the final call on whether each asset feels right.
A more ambitious campaign.
The first payoff was creative ambition. Under the old process, the realistic plan was one piece of art stretched across hundreds of executions, because every extra variation multiplied the manual work required to apply it to each and every asset. With automated assembly, we built four ranked versions of the artwork for each audience and placed them across the full campaign. If one doesn’t land with an audience, we swap that single component in the master template rather than editing thousands of individual files.
The second payoff was flexibility on offers. Promotions are the last thing to lock and the most likely to move, and Black Friday runs in phases with different pricing in each. A late change used to mean opening every layout, pasting in the new headline, and hoping nothing was missed before export, days before launch. Now the campaign can respond to pricing decisions throughout its run instead of freezing creative to protect the production schedule.
Where people still matter most.
Transcreation, the adaptation of tone and message so that cultural nuance survives translation, is human work, and it stays that way. In the Middle East and North Africa, the retail moment is not Black Friday at all. It’s White Friday.
Even with AI doing the heavy lifting, human experience, taste, and judgement matter more than ever. A system can apply a rule across thousands of layouts. It can’t tell you when a rule should bend, or whether a literal translation will land flat.
What has changed is how much time the team has to spend on those calls. The hours that used to go into applying copy across thousands of layouts now go into deciding which markets need a different tone and which need different artwork. With four ranked variations per audience, the team can review which artwork suits which region rather than defaulting to one version everywhere.
What I have learned as customer zero.
Being customer zero means we run new technology against the real complexity of our own business before we ask anyone else to. Three lessons stand out.
- Brand judgement has to be written down. The regional guidelines we built for this campaign are now an asset in their own right. They hold knowledge that used to sit in the heads of a few long-serving colleagues, and they move the work to the start of the process instead of the end.
- Scale and standards are not a trade-off. Checking work as it’s created, rather than after it’s built, is what lets us take on more without accepting more risk.
- The point is expanding what people can do. More of my team can now hold the line on brand quality, because the system carries out the routine checks.
A year from now, our Black Friday campaign will be bigger still. Yet for all that’s new here, the core task remains the same: we take a single idea and make it land in every market, without losing what makes Adobe look and sound like Adobe.
Learn more about how Adobe Brand Intelligence helps you scale on-brand content with precision.
Recommended for you
https://business.adobe.com/fragments/resources/cards/thank-you-collections/brand-intelligence