Skip to main content

Evertune FAQs

Commonly asked questions and answers about our platform. If you can't find what you're looking for, try asking our AI agent in the bottom right corner of our platform, checking out our product specific guides, or asking your Account Manager.

Updated over 2 weeks ago

Evertune FAQs

Are you planning to add additional models into the platform?

Yes! We're continually adding new models and prioritize them based on consumer use volume. For any further questions, contact your Account Manager.

How can I change my score?

Evertune is capturing what answers the models naturally return on a range of prompts. AI learns by reading content online so improving how you are showing up is dependent on how your brand is appearing online. Developing content and communication strategies that are published on as many authoritative domains as possible following EEAT best practices is a good strategy.

How long does it take to influence an AI model on a consumer preference or category?

Expect AEO changes to move slowly. Most improvements take 3–6 months to show meaningful impact. Early signs, like better indexing or occasional inclusion in AI answers, may appear after 6–8 weeks, but large, consistent visibility gains typically require a full quarter or more. Month-to-month data will look relatively flat while models crawl, process, and adjust to new content and authority signals, so AEO should be measured on a quarterly horizon rather than monthly swings. You can think of the base model as the long term game, while influencing citations in search will show quicker improvements.

Can you generate insights for a specific audience?

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.

Will your attribution model estimate brand lift in zero‑click scenarios where no link is provided?

Yes. Evertune can detect and report brand mentions in AI outputs even when no URL or citation is included.

Aren’t LLMs changing constantly?

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.

How do you come up with the prompts?

Our prompt selection is based on market research and EverPanel's insights from real LLM conversations. We've built a proprietary panel of 25 million internet users in the U.S., demographically weighted to reflect real-world usage patterns. The prompts are built to mimic how a consumer would search for answers in and around the category. We cluster prompts into broader intents and topics rather than exact keywords because over 80% of AI prompts are unique and conversational - people ask the same thing in completely different ways. You can also supplement our data-driven approach with Custom Prompts - up to 250 prompts specific to your strategy for tracking specialized use cases unique to your business.

What is your data methdology? How do you prompt the models to get these insights?

We run 1M+ custom prompts per brand monthly and analyze every response, capturing patterns across a massive distribution of responses from all the major AI chatbots. How do we get to 1M+ prompts? 100+ prompts x 100 times x 10+ AI models x 10 categories = 1+M prompts per month! We do this through 3 methods:

  1. Direct API Access - Inside the AI's brain: Direct connections to Large Language Models (LLMs) measure how the base model perceives your brand—before any live search clouds the picture. Think of it as reading AI's first impressions at enterprise scale. These insights are available in Word Association, AI Brand Index, and Consumer Preferences.

  2. Consumer App Data Collection - What the model finds in real time: Data from actual consumer apps reveals how real-time search influences recommendations. Because what AI knows versus what AI finds can be two different stories. These insights are available in Word Association, AI Brand Index, Consumer Preferences, and Content Analytics.

  3. EverPanel - What real people actually search for: Our consumer panel of almost 25 million people shows what they're asking, the language they use and how often they're searching, across your brand, category and competitors. These insights are available in AI Usage and Prompt Volumes.

How do I copy a chart? I want to use a chart in a Powerpoint/Google Slide presentation.

To copy a chart, click the camera icon on the top right corner of the chart. Currently, we support copying bar charts and radar charts. Copying donut charts, word-cloud images and data tables is coming soon!

AI Models

What is the benefit of direct integration to the API of the API models? How does Evertune stand out against competitors' solutions?

Most competitors collect data from the consumer-facing app as their only methodology. So much like SEO tools of old, they scrape the consumer UI.

We do this too, but we also plug into the model maker API's as:

  1. Clean, Reliable Data Collection: We pay the model providers for access to connect directly through their API. This provides us scaled access ensuring clean, reliable, repeatable data collection. Most competitors do not have this and are in a cat-and-mouse game, hoping not to have their IP blocked from accessing the consumer app.

  2. Statistical Significance: Because we tap into the API, we can analyze 1M+ prompt responses per brand monthly, giving us statistically significant, reliable and robust metrics rather than small sample sizes.

  3. Base Model Insights: The APIs are the only way to isolate the base model’s knowledge. There isn't a way to force this in the consumer app. This allows us to understand what AI's preferences are at its core. This is especially important as the consumer app and all agent experiences are built on top of the base models. So these agents inherit brand preferences from the base model.

  4. Brand Building: API is a more stable measure of long-term AI brand building. It does not have access to live searches, so is less reactive to changes in search indexes. It shifts when there has been a true model update.

AI is unique in that we have a way to programmatically query the underlying model. It's equivalent of Google opening up their page rank algo and allowing people to query it directly vs. having to infer the results based on SERPs. The only way to access this is through API.

You can see the results in the UI:

  • ChatGPT - is the base model collected directly from API

  • ChatGPT Search - collected through the consumer app. The biggest difference is that search is present

  • Gemini - base model collected through API

  • Gemini Search - collected directly from the consumer app. The biggest difference is that search is present

  • Google AI Mode - Google's AI mode experience on Google.com

What happens if a brand is too small? E.g. a regional restaurant or soda brand that won't place against Coca-Cola in a more national survey.

If the user prompts about the specific niche the brand operates in it can cut through. But if the brand is too niche and has no digital footprint, then AI won't know about it and will bias to less niche brands to try and answer users' prompts. Models are optimized to try and be helpful so rather than say "i dont know" it will pay attention to other parts of the users prompt. E.g. if someone is asking for "local craft beer from Springfield, NY" it will likely put more attention on "Craft Beer" than "Springfield, NY."

Is there a correlation between running more CTV (Connected TV) ads and improving AI Brand Scores?

While there is no direct correlation, running more CTV ads drives greater brand awareness, which leads to more people writing about your brand on the internet. More online content about your brand can lead to better performance in AI evaluations.

Which LLMs and AI search engines do you monitor?

Currently: ChatGPT, ChatGPT Search, Gemini, Gemini Search, MetaAI, Claude, Perplexity, DeepSeek, Google AI Mode, Google AI Overviews, and CoPilot.

Reporting

What reports does Evertune provide?

Do you have a reporting API?

A reporting API is on our roadmap. Platform users can download an Excel or print to PDF through the user interface. The Excel file contains the raw data that populates the charts visible on the platform. Additionally, we can push files to cloud storage providers like GCP, AWS, Azure, Snowflake as well as Google Drive, and Dropbox folders.

How often do reports refresh?

By default the trackers are set to auto refresh every 30 days. If you would like to change that setting you can do so by navigating to one of the reports under a product category, then click on Settings located in the upper right corner of the screen. The Settings modal will appear. In the modal, you can change the specific day of the month you would like the data to refresh or toggle the automatic data refresh off.

How long will it take for reports to complete?

The speed of reports depends on how quickly the AI models return answers, as Evertune has to wait for each model to respond. Because Evertune is asking thousands of questions at a significant scale, this can take several hours, but it is usually much faster. We will send you an email when the report is complete. If a report has taken longer than 12 hours please contact [email protected] or contact your account manager.

Did this answer your question?