The LLM Shift: Why Branding Is Becoming a Competitive Advantage Again

For the past fifteen years, digital marketing has given advertisers the perfect playground for performance-driven tactics: trackable, measurable, and endlessly optimizable in real time. Brand-focused digital channels, by contrast, have remained relatively underfunded. According to the 35th edition of the SRI Observatoire de l’e-pub ("e-advertising observatory"), Search still accounts for 40% of advertisers’ media investments in 2026. And that figure does not even include other performance levers such as affiliate marketing or retargeting.

The underinvestment in branding is not due to a lack of interest. Rather, branding effects are broader, slower to materialize, and poorly suited to the quarterly reporting cycles that shape many marketing organizations.

But 2026 marks a turning point. As users increasingly shift toward LLMs (ChatGPT, Gemini, Perplexity, Claude, and others), the logic behind purchase decisions is being fundamentally reshaped. According to a SEMrush study conducted in December 2025 among 1,030 U.S. shoppers, 85% of consumers now use AI tools weekly for purchase research. This behavioral shift is changing the media equation for advertisers.

At fifty-five, we believe branding is about to play a much larger role, especially since very few advertisers are truly prepared for this transition.

Branding: Digital Marketing’s Neglected Discipline

Across most of the digital media mixes we manage at fifty-five (luxury brands excluded), performance marketing still dominates. The reasons are well known and reinforce one another:

• A structural measurement bias: performance channels are easier and faster to measure, especially through last-click attribution. Branding, meanwhile, works through long-term effects such as aided awareness, consideration, and brand preference, which connect less directly to short-term revenue.
• Short-term budget trade-offs: for marketing departments operating under quarterly objectives, it is far easier to justify one euro spent on paid search than one euro invested in a brand campaign.
• Measurement tools that remain imperfect: even Marketing Mix Modeling (MMM), despite major advances in sophistication over recent years, still struggles to fully capture branding effects. The challenge lies in reconciling two very different time horizons: short-term performance drivers and long-term brand-building efforts. Methodologically, this remains an open field.

The result is clear: many advertisers underinvest in branding today not out of conviction, but by default. And this is precisely the bias that the rise of LLMs will force the industry to reconsider.

LLMs Are Reshaping the Decision Journey… and Media Budgets

Traditional search, meaning all channels built around user queries processed by conventional search engines (brand and non-brand SEA, Shopping, SEO, and even much of Performance Max through Shopping feeds), still absorbs the largest share of digital media budgets for many advertisers.

But this dominance is beginning to erode.

Back in 2024, Gartner predicted that search engine query volume would decline by roughly 25% within two years due to AI chatbots and virtual agents. McKinsey goes even further: by 2028, AI-powered search could influence $750 billion in U.S. revenue, while unprepared brands risk losing between 20% and 50% of their traditional search traffic.

Recent signals confirm the trajectory. In early 2026, ChatGPT already accounts for roughly 20% of global search traffic. Meanwhile, Google’s AI-generated summaries (“AI Overviews”) are dramatically compressing click-through rates. Ahrefs’ study of 300,000 keywords found that pages ranking first lose 58% of their CTR when an AI-generated answer is displayed.

In practice, more and more users ask an LLM a question, receive a synthesized answer, and make a decision without ever visiting a traditional Google results page. The journey is becoming shorter.

This leads to a simple strategic question for advertisers: if “traditional” Google search loses volume, where does the budget go?

Advertising products for LLMs are still in their infancy. Google is integrating Gemini into its ad ecosystem, while OpenAI is exploring monetization paths for ChatGPT. But user behavior is evolving much faster than advertising supply. Many major players do not even have a clear short-term advertising roadmap yet.

The budgets freed up from search will therefore need to be reallocated elsewhere. And in our view, that “elsewhere” must prioritize branding.

In an LLM Environment, There Is No Second Page Results

To understand why branding is becoming mechanically more important, it helps to look at how LLMs generate answers.

On Google, search results pages display around ten links, and often more. Users can scroll, compare, and click across multiple results. Even brands that are not ranked first still have opportunities to capture attention.

With an LLM, the logic is radically different.

When a user asks, “What is the best solution for X?” or “Recommend me a brand for Y,” the answer typically synthesizes three to six options. End of story. No page two. No infinite scrolling. Either your brand appears in the response, or it effectively does not exist for that query.

Visibility becomes binary, and that shift fundamentally changes the role of branding from building preference at the end of the funnel to simply existing within the LLM’s answer universe itself.

Why Brand Activations Are Becoming Strategic Priorities

Two main dynamics are reinforcing each other.

1. Brand Content Feeds LLM Knowledge Systems

LLMs rely on massive corpora to construct answers: articles, videos, reviews, discussions, and structured content. The more a brand is present, referenced, and discussed within those ecosystems (especially through content that generates engagement) the more likely it is to appear in synthesized responses.

And the content that drives engagement (comments, shares, editorial coverage) is rarely content designed purely for performance marketing. Performance content is often highly transactional and product-centric.

The content that truly resonates is brand content: a well-produced and sponsored YouTube video, a campaign that sparks conversation, a bold brand statement, a branded documentary.

In 2026, YouTube has become the most-cited platform in LLM responses, overtaking Reddit, which had dominated through 2025. Several converging studies (Profound, Goodie AI, Bluefish) confirm this shift: YouTube now appears as a source in 16% of LLM responses, compared with 10% for Reddit.

Choosing branding therefore means indirectly feeding LLMs with richer associations about your brand — and statistically increasing your chances of being recommended.

2. Awareness and Consideration Become Critical Selection Drivers

When an LLM presents three to six options, what determines which one a user clicks?

Brand recognition.

This is exactly what advertisers already measure in brand-tracking studies: aided awareness, consideration, and top-of-mind recall. In an LLM-driven ecosystem, these metrics become doubly strategic. They influence both:

• the likelihood of being included in the AI-generated shortlist,
• and the likelihood of being selected once included.

This creates a virtuous (or vicious) cycle, depending on how prepared a brand is. LLMs tend to recommend brands associated with strong, positive signals, which naturally reinforces established market leaders and makes it harder for challengers to break through.

3. One Final Advantage: LLM Traffic Converts Better

The final argument may be the most compelling: traffic generated by LLMs converts significantly better than traditional search traffic.

According to SEMrush, visitors coming from LLMs convert 4.4 times better than visitors from organic search. Microsoft Clarity found even larger gaps:

• Copilot traffic converts up to 15 times better than traditional search traffic,
• Perplexity converts 7 times better,
• Gemini converts 3 times better.

Volumes remain modest today (on average, under 2% of referral traffic) but the quality of that traffic is already far superior, and the growth trajectory is unmistakable.

fifty-five’s Branding Recommendations

We see three major areas where brands should evolve their approach.

1. Rebalance the Media Mix

Brands need to deliberately rebalance budgets in favor of branding and accept that ROI will not always appear immediately in short-term dashboards.

This also requires equipping marketing teams with the tools and frameworks needed to defend these decisions internally, especially against finance departments accustomed to evaluating investments primarily through quarterly ROAS.

2. Take Branding Measurement to the Next Level

This is where much of the future credibility of branding investments will be decided. As long as branding effects remain poorly measured, budget allocation will continue to favor performance marketing by default.

In our view, three priorities can meaningfully improve measurement:

• Better integrate long-term effects into MMMs. Most current MMMs remain heavily calibrated toward short-term outcomes, structurally disadvantaging brand channels. Models must evolve to explicitly separate and quantify short-term conversion effects from long-term brand equity effects.
• Move beyond revenue-only measurement. MMMs increasingly need to incorporate broader indicators capable of capturing branding impact more accurately — including aided awareness, consideration, and purchase intent metrics collected through third-party providers such as YouGov.
• Measure synergies between branding and performance. Beyond evaluating each channel independently, advertisers must quantify how branding amplifies performance: how a YouTube brand campaign boosts Search efficiency, how Display investments improve conversion rates, or how branding strengthens revenue generated through GEO (Generative Engine Optimization). This is already a core component of the MMM work we conduct for clients at fifty-five.

3. Treat Brand Content as an LLM Asset

Every brand message is no longer just a potential impression or click. It is also a building block that either strengthens or weakens a brand’s presence inside AI-generated answers.

This requires rethinking formats, channels, and campaign KPIs around a new question:

Is this content visible, engaging, and widely referenced enough to become part of the corpora that feed LLMs?

The opportunity window is still open but it is closing quickly.

Brands that invest now in their visibility within LLM ecosystems, and develop the measurement capabilities to track it, will build an advantage that latecomers may struggle to overcome. Few media trends today make timing matter as much as investment size.