Most legacy companies are not being beaten because they forgot how to market. Many still have good products, trusted brands, strong distribution, serious teams and large customer bases.
The practical decision for a Luxembourg SME or mid-market company this quarter is different: stop treating AI content as a cheaper production trick, and start treating it as a way to increase the speed of market learning.
That is the real threat.
A legacy company may plan one polished campaign, approve it through several layers, produce a small number of assets and launch it across broad audiences. An AI-native challenger can test dozens of hooks, visuals, formats and audience angles in the same time. It does not need to win the whole market immediately. It only needs to learn faster in small segments that the larger company is too slow to notice.
The campaign is becoming a learning system
The old campaign model was built around scarcity. Production was expensive. Video required locations, talent, crew, editing and time. A shoot produced a limited number of assets, so every decision had to carry more weight.
The new model is not simply "generate more content." More content by itself can easily become noise. The better version is a creative learning system:
- Identify one narrow audience.
- Generate several message and visual directions.
- Launch small tests.
- Read the signals.
- Create the next round from what the market actually did.
The advantage is not the number of images or videos. The advantage is the number of feedback loops.
The evidence is already visible
The advertising market is moving in that direction. IAB reported in July 2025 that half of advertisers were already using generative AI to build video ads. It also said 86% of buyers were using or planning to use GenAI for video ad creative, and buyers expected GenAI creative to reach 40% of all ads by 2026.
The same IAB report is especially relevant for smaller companies. It says small and mid-tier brands are adopting GenAI faster than the largest brands, using it to create digital video ads quickly, affordably and at scale. Advertisers are not only using AI to make generic assets. IAB says they are using it to create versions for different audiences, visual style changes and contextual relevance.
That is the strategic shift. A company can now create marketing material for smaller audience cells: a specific age group, lifestyle, country, problem, aspiration, platform or moment.
Adobe's marketing research points to the pressure behind this. In October 2025, Adobe wrote that content demand had doubled for 96% of marketers, with nearly two-thirds reporting a fivefold increase. It also said 76% of marketers reported shorter timelines and 48% of creative teams were struggling to keep up.
This is why AI-native teams feel different. The production bottleneck has moved.
One enterprise example shows the operating model
Adobe said in June 2025 that Lumen reduced the time to launch B2B marketing campaigns from 25 days to 9 days using Adobe GenStudio for Performance Marketing. Adobe also said the time to create four ad variations for Meta properties across two target personas was reduced by 65%.
The interesting part is the operating model. Brand foundations stay controlled, but marketing can activate and optimize variations without rebuilding the process every time.
That is the lesson for legacy companies. The question is not "which AI image tool should we buy?" The better question is "what approval, measurement and brand system would allow us to test more without losing control?"
Personalization is not a nice extra anymore
McKinsey says personalization can reduce customer acquisition costs by as much as 50%, lift revenues by 5% to 15%, and increase marketing ROI by 10% to 30%. McKinsey also says 71% of consumers expect personalized interactions, and 76% get frustrated when that does not happen.
This does not require creepy targeting. The useful interpretation is simpler: people respond when a message feels specific to their situation.
AI lowers the cost of making that specificity operational. A brand can test whether a product should be framed around performance, convenience, status, health, price, confidence, family, local identity or proof. It can test which visual language feels credible to one audience and irrelevant to another.
Legacy companies often have the data. What they lack is the creative and operating speed to turn that data into enough useful experiments.
CPG is the warning sign
Bain's 2025 insurgent brands research found that insurgent brands accounted for less than 2% of market share in the categories where they operate, but nearly 39% of incremental category growth in 2024. In food, Bain said these brands were responsible for more than 27% of growth despite less than 1% market share. In nonalcoholic beverages, they took more than 32% of growth while accounting for less than 3% share. In personal care, they held 3% of share while taking 45% of growth.
That is the slow erosion problem. A large brand may not lose the whole category. It may simply miss a disproportionate share of the new growth.
McKinsey's January 2026 CPG analysis makes the same point from another angle. In vitamins, minerals and supplements, an $18 billion category that grew by about $4 billion over five years, McKinsey says roughly half of new growth came from disruptor brands. McKinsey also says disruptors are using bold and culturally relevant messaging, digital fluency, social channels and faster experimentation to stay close to changing consumer needs.
NIQ and Kearney add the AI layer. Their March 2026 analysis says established niche brands increased U.S. market share by 1.5 percentage points from 2022 to 2025, while large and mid-size national brands declined by 2.1 points. They also report that 74% of shoppers use AI for some form of product discovery, 54% use AI for research and 20% use AI directly for shopping.
So this is no longer only about what people see in a social feed. It is also about what products AI systems can understand, compare and recommend.
Why big companies may miss it
If one competitor takes 10% of your market, everyone notices. If 40 smaller brands, creators, niche retailers and private labels each take a little attention, a little demand and a little category growth, the dashboard may show something vaguer: slower growth, weaker campaign efficiency, higher acquisition costs or lower share of new buyers.
That is why the threat is not size. It is speed.
Large companies still have real advantages: trust, distribution, compliance, retail relationships, brand assets, customer data and budget. But those advantages do not automatically protect a company from faster learning cycles.
A large customer base can keep revenue stable while new demand moves elsewhere. A large media budget can maintain awareness while challenger brands win smaller, more specific buying moments.
My opinion
My opinion: the next marketing advantage is not content volume. It is creative velocity with judgment.
The companies that win will not be the ones producing the most AI assets. They will be the ones that can turn customer insight into testable creative, publish it safely, measure it honestly and adapt quickly.
That requires more than tools. It requires a different operating posture.
What to do this quarter
- Measure creative velocity. Count how many meaningful creative tests your company runs per week, not just how many campaigns it launches per quarter.
- Build audience cells. Take one product or service and break the audience into five specific segments with different needs, objections and triggers. Do not start with 50. Start with five that sales, support or customer data can defend.
- Create an AI-assisted test workflow. Define who writes the brief, who generates variations, who checks brand and legal risk, who launches the test, and who reads the results.
- Protect the brand without freezing the team. Keep rules for claims, tone, regulated language, privacy and disclosure. But do not force every low-risk test through the same process as a national campaign.
- Watch the edges of the category. Track not only your obvious competitors, but smaller brands, creator-led offers, Amazon-native products, TikTok-native formats, retailer search results and AI shopping answers.
The risk is not that one AI-native company suddenly replaces a legacy brand.
The risk is that hundreds of smaller brands can now test thousands of sharper messages faster than the legacy brand can approve one campaign.