Measuring brand experience in LLMs: the Amplif-AI approach

An infrastructure that combines human listening and statistical measurement to understand brands in generative AI

Generative artificial intelligence is reshaping the way people discover, evaluate, and choose products and services. Increasingly, linguistic models are becoming the first point of contact for clarifying doubts, comparing alternatives, and validating purchasing decisions. In this new scenario, the responses provided by LLMs shape brand perceptions, influence preferences, and redefine the relationship between customers and brands.

It is in this context that we developed Amplif-AI, a proprietary infrastructure designed to empirically and statistically significantly measure how brands are represented by generative models. We launched the service experimentally in 2025 and have now evolved it into a mature solution, already adopted by leading brands in the banking, insurance, energy, and consumer goods sectors.

But Amplif-AI is more than just a monitoring tool. We conceived it as an evolution of our method: starting from people’s real experiences to understand how the brand experience is changing and how customers use LLMs in their decision-making process.

Generative AI as a new experience space

Generative AI isn’t just an emerging technology. It’s a new environment where trust, expectations, and decisions are built. People don’t just seek information: they interact with systems, ask complex questions, and seek comparisons and validation.

We have observed that this shift has profound implications for brands:

• LLM responses become an integral part of the decision-making process.
• Brand representation in generative outputs influences perception and trust.
• Selection criteria evolve based on the information received.
• New information expectations emerge that traditional channels don’t always meet.

In this space, visibility isn’t enough. What matters is how a brand is described, what attributes emerge, what associations are created, and how it positions itself compared to competitors. Measuring the brand experience in LLM programs therefore becomes a strategic priority for those who want to understand and lead this change.

Observing real behaviors to understand change

Every Amplif-AI project begins with research dedicated to the brand’s specific target audience. We don’t offer general analyses or abstract simulations. We choose to observe the real behaviors of customers and prospects:

• What questions do they ask LLMs during the decision-making process?
• At what points in the decision-making process do they question them?
• What selection criteria are influenced by the answers received?
• What new information expectations emerge?

This phase allows us to understand how AI intermediation is changing the customer journey and to highlight the points where brand representation can impact perception, trust, and the final decision.

We built everything by listening to people: not as a theoretical starting point, but as an empirical foundation upon which to build each subsequent phase of analysis. Visibility in LLMs is not the initial goal, but the natural consequence of a deep understanding of target audience behaviors.

The right questions emerge from research

Qualitative research provides us with concrete insights that guide the entire measurement infrastructure:

• Which topics and attributes are relevant to the target audience?
• Which competitors are considered in the comparison?
• What information gaps exist between what people search for and what they find?
• How is the language used in interactions with generative artificial intelligence changing?

These elements become the basis for building a broad, representative, and market-accurate inquiry system.

Proprietary Infrastructure: Data Amplification and Analysis

Starting from real insights, we’ve built a proprietary amplification infrastructure that submits thousands of queries to leading linguistic models. The goal is to create a broad, robust, and replicable database that allows us to measure brand representation with statistical rigor.

We don’t just verify the presence of a brand in the results. We analyze:

• How is the brand described and what attributes emerge?
• What associations and values ​​are linked to the brand?
• How are competitors positioned and what elements are recurring or absent?
• What territorial or segment dimensions influence the representation?

We’ve limited our analysis to the Italian market and can now delve deeper into specific regional dimensions, providing a reading that reflects the real competitive context.

From data to strategic insights

We organize the collected data into customized dashboards and translate them into strategic insights useful for guiding operational and communication decisions:

• Review of information assets and digital content
• Definition of intervention priorities based on their impact on perception
• Building an ongoing optimization and monitoring process
• Identification of positioning opportunities compared to competitors

We hosted the infrastructure on GDPR-compliant systems, ensuring full control and data governance. This is crucial for brands operating in regulated sectors or handling sensitive information.

Continuous measurement and time comparison

Amplif-AI isn’t a one-time analysis. We’ve designed the infrastructure to be activated over time, allowing you to:

• Monitor the evolution of brand representation
• Evaluate the effectiveness of actions undertaken
• Intercept changes in the competitive environment
• Adapt strategies based on new people’s behaviors

This temporal dimension transforms the measurement of brand experience from a static snapshot to a dynamic process of learning and continuous improvement.

A tailor-made project for every brand and target

In a market where many solutions promise to monitor LLMs, we wanted to distinguish Amplif-AI through our approach, depth, and method. We don’t offer one-off tests or generic dashboards, but rather a tailored project built on the specifics of each brand and the real-world behaviors of its target audience.

Our goal is not simply to increase visibility in generative systems, but to understand how generative AI is redefining the overall brand experience and provide organizations with concrete tools to manage it strategically and sustainably.

Why Amplif-AI is not just a monitoring tool

The difference lies in the method we have built:

1 – We start with people, with their experiences and real needs
2 – We connect qualitative research and quantitative measurement into a single, coherent flow
3 – We provide strategic insights, not just data
4 – We adapt to the specific context of each brand and industry
5 – We ensure governance and compliance of the data collected

We designed Amplif-AI for those who want to truly understand what’s happening, not just observe it from afar. For those who know that technologies evolve, but that the truth always lies in people’s real experiences.

“Generative AI is becoming a space for building trust,” explains Edoardo Ferrini, Head of SEO & GEO at TSW. “With Amplif-AI, we’ve connected human research and statistical measurement on LLMs to offer organizations a concrete understanding of what’s happening and the tools to intervene effectively.”

“Technologies evolve and redefine brand interaction spaces, but the truth always lies in people’s real experiences: customers and prospects remain and will remain the starting point for every one of our projects,” adds Christian Carniato, CEO of TSW.