Currently: ChatGPT, ChatGPT Search, Gemini, Gemini Search, Google AI Mode, Google AI Overviews, MetaAI, Claude, Perplexity, DeepSeek. We regularly add new models.
How AI models describe or mention a brand without the brand’s name in the prompt. For example, if a user prompts an AI model “What is the best luxury SUV?” and the AI model responds with “BMW”. The brand was not mentioned in the prompt, but the AI model mentioned the brand. Evertune’s AI Brand Index and Consumer Preferences reports analyze unaided awareness.
How AI models describe a brand when the brand’s name is included in the prompt. Evertune’s Word Association report analyzes aided awareness by asking the AI models to describe specific brands within a product category. For example, “how would you describe the car brand BMW?”
Different versions of the same model produce different results because newer versions are usually bigger, trained on more recent information, and built with improved techniques. Even though they might seem similar, each version has learned different patterns from different data, so they naturally respond differently to the same prompts.
Absolutely. Evertune can generate an AI Brand Index report for any product category. This allows you to see AI preferences for different products, audiences and price points. For example you could see what brands AI recommends for "SUVs" or get more specific with "luxury SUVs" or even call out product features and price points like "3 row luxury SUVs under $100k".
While citations provide useful guidance, they should be treated as indicators rather than definitive proof. LLM citations aren't always accurate and may reflect the model's built-in preferences, so they can sometimes reinforce incorrect conclusions rather than validate them. Additionally, due to AI models probabilistic nature, citations can vary across responses. Therefore at Evertune we advise clients to focus on appearing in a diverse set of authoritative and AI friendly content as much as possible and using our AI Education Score to help prioritize where to focus.
Over 80% of AI prompts are unique—meaning they’ve never been seen before. That’s because people tend to write prompts the same way they’d talk to another person:
For example, two people might ask the same thing in completely different ways:
Same intent, different words. That’s why Evertune’s Prompt Volumes doesn’t just scan for granular keywords—we cluster prompts into broader intents and topics, giving you a more complete and accurate measure of user activity.
Yes, just enter a specific audience—like "soccer moms in Texas"—in the product category field, and Evertune will generate insights by asking the models for results tailored to that group.
Yes. Evertune can detect and report brand mentions in AI outputs even when no URL or citation is included.
They are, which is why Evertune’s platform tracks and updates reports regularly. Our methodology adapts in real time as new model versions are released or retrained. We also allow you to compare how your brand shows up on different model versions (e.g. Chat GPT 4 vs. Chat GPT 5).
AI models can sometimes distinguish sponsored from organic content, but it depends on how clearly it's marked. If labeled with tags like "sponsored" or "advertisement," such content can be filtered out during dataset curation. However, when content is not clearly labeled - especially if it takes the form of high-quality guides, reports, or whitepapers - it may still influence the model, particularly if it comes from a recognized brand or authoritative source.
Today, most consumer-facing AI is text-first and typically learns from text derivatives such as video transcripts, captions, or OCR-extracted text from images, rather than directly from the video or images. Sometimes you may see YouTube URLs listed in our Sources report. This means that the AI model is reading the title and description of the video, rather than watching the actual video.
As AI models evolve, it’s important to remember that access to training data depends on the AI model’s parent company. For example, Google can use YouTube video data to train its AI models, Meta can use Instagram data to train Llama, and OpenAI has licensing deals with platforms like Reddit. Unless there's a direct agreement, most social media, video content, or in-app content is not included in training sets.
In traditional search, users typically enter in keywords + "near me" without typing in their location as their browser typically has location settings enabled. When using AI models, users tend to include their location in the prompt. If you want to track a product category that differs greatly across regions, we recommend creating unique Evertune trackers for each key location + product category you want to track.
Yes, personalization is evolving. For example, in April 2025, OpenAI expanded ChatGPT's memory capabilities, allowing it to retain facts about a user across sessions (such as name, preferences, or prior context). This enables more personalized answers. Other assistants are developing similar features. However, unless explicitly shared or remembered, models don't automatically bias results by age, location, or preferences. If your brand is hyper local or age specific, we recommend creating multiple trackers in our platform to compare across personas (i.e. clothing for women vs. clothing for girls) or locations (i.e. northeast credit unions vs. southeast credit unions).
