## **🗺️ Overview**

The **AI Visibility** tool (left sidebar → **AI Visibility**) lets you monitor how large language models (LLMs) cite your brand and competitors across different locations. This article explains exactly how the country picker works, whether city-level targeting is available, and how location behavior differs across ChatGPT, Gemini, and Google AI Mode.



## **🌐 How the Country Picker Works**

The country selector in AI Visibility does **two things simultaneously**:

- **Filters the data** you see in your dashboard to show only results collected for that country.
- **Injects a simulated geolocation context** into the queries sent to supported LLMs, so the model responds as if the user is located in that country.

This means your results reflect both localized data storage *and* geographically influenced model responses — not just a filtered view of a single global dataset.



## **🏙️ City-Level Targeting**

City-level targeting is **available for select queries** in AI Visibility. Search Atlas has been actively expanding city-level coverage, and new cities are being added on an ongoing basis. If you do not see your target city listed, it may not yet be supported — but the city list continues to grow.

To check available cities, open QUEST AI Visibility, select your country, and look for the city dropdown beneath the country selector. If a city option is not present for your region, country-level targeting will apply by default.

**Note:** A known issue is being investigated where the country/location selection can reset from a non-US country back to the US when navigating from Site Explorer into AI Visibility. If you notice your location has changed unexpectedly, simply reselect your preferred country and city before running a query.



## **🤖 How Location Behavior Differs by LLM**

Each LLM handles geolocation context differently. Here is a breakdown of how each model responds to location signals set in AI Visibility:

- **ChatGPT (OpenAI):** Geolocation is simulated by including location context in the query prompt. ChatGPT does not have native real-time location access via the API, so Search Atlas injects country and city context directly into the request. Responses reflect the instructed location rather than a true IP-based location.
- **Gemini (Google):** Gemini accepts geolocation parameters that influence local relevance in its responses. Search Atlas passes the selected country (and city where available) as part of the request, and Gemini factors this into rankings and citations.
- **Google AI Mode (AI Overviews / AI Mode in Search):** Google AI Mode is the most location-sensitive of the three. It uses both the simulated location context *and* Google's own localization signals. Because of this, results in Google AI Mode may be harder to replicate manually in a browser unless your browser location, language, and search settings exactly match the simulation parameters. An active investigation is underway for cases where AI Visibility reports a website citation that cannot be replicated directly in Google AI Mode.

## **🗣️ Does Query Language Affect Which Model Version Responds?**

Yes — query language can influence model behavior in the following ways:

- **ChatGPT:** Queries written in a non-English language may route to language-optimized model configurations. The underlying model version remains the same, but response style, phrasing, and local knowledge depth can vary by language.
- **Gemini:** Gemini is multilingual and adjusts its response quality and sourcing based on the query language. Queries in languages with less training data may return different citation patterns than English queries for the same topic.
- **Google AI Mode:** Language strongly influences which localized index Google draws from. A query in French, for example, will pull from French-language web content even if the country is set to a predominantly English-speaking country. For the most accurate local simulation, align your query language with the language spoken in your target location.

**Best practice:** Write your tracked queries in the language your target audience actually uses when searching, and pair them with the matching country and city setting for the most realistic visibility data.



## **📍 Coming Soon: Per-Location Local AI Visibility Tracking**

Search Atlas is developing a dedicated **per-location local AI Visibility tracking** feature for Google Business Profile (GBP) locations. This will allow businesses with multiple physical locations to track AI-generated mentions and citations at a granular, location-specific level. This feature is currently in the planning stage and is not yet available.



## **✅ Quick Reference Summary**

- The country picker **both filters data and simulates geolocation** in LLM requests.
- City-level targeting is available for **select cities** and expanding over time.
- If your location resets to the US unexpectedly, reselect it manually — a fix is in progress.
- ChatGPT uses prompt-injected location; Gemini and Google AI Mode use native location parameters.
- Query language affects citation sources and model response quality across all three LLMs.
- Per-location GBP tracking is on the roadmap but not yet released.

If you need further assistance, open the chat widget in the bottom-right corner of the platform and type **human teammate** to be connected with a member of our team.