Semrush helps you find gaps in your AI visibility by identifying topics, sources, and narrative themes where competitors are mentioned in AI responses but your brand is not.
The process is similar to competitive SEO analysis. You find out when the opponents are winning and close the gap.
The Semrush AI Visibility Toolkit (available with Semrush One) helps you do just that.
Here’s how to get started.
1. Find Topics Where Your Brand Is Missing With a Competitive Research Report
The Competitor Research Report identifies topics and commands where your competitors appear in AI-generated responses but your product does not.
Start by introducing “Competitor Research” within Semrush’s AI visibility toolkit.
Enter the URL of your site and up to four competitors. If you’re not sure who to track, Semrush will suggest top competitors based on your domain.
Then, click on “Start a competitor analysis.”
You’ll see a benchmark overview that compares your product’s AI performance against your competitors across several AI platforms.
This includes:
- Appearance of AI: A benchmark score (0–100) that measures how clearly and reliably a product is reflected in relevant AI-generated responses.
- Audience: A measure of the total reach/disclosure of a product’s AI based on the volume of all instructions and topics where the product is mentioned.
- Meaning: The total number of unique details where your brand name appears in the text of the AI response.
Use the dropdown above to filter by specific AI area:
- ChatGPT
- Gemini
- Overview of Google AI
- Google AI mode
For topic specific posts, scroll down to the “Topics & Information” section.
You’ll see a list of themed collections that you can sort based on what each product says. You will also be able to check the volume of AI estimated for each topic.
Use this metric as a guide to prioritize which topics represent the greatest opportunities.
To check specific commands, click on any heading in the table.
use the “Lost” filter topics where AI platforms talk about competitors but not your brand.
Sort the column of one of the competitors from top to bottom to sort the table with the highest number of mentions.
This will give you articles where a certain competitor has a high number of mentions, while your product is completely absent.
Review this list to prioritize topics that are most relevant to your business, then expand to see specific commands that drive those posts.
To see more granularity, switch to “hurry up” look at the top of the table and use the same “Nothing” filter.
Here you will find specific questions and instructions that users give to the AI where competitors appear but your product does not.
If these warnings resonate with your business, then you have an AI visibility gap.
Plan to bridge the gap by creating and optimizing the content of this article specifically so that AI platforms can cite your site as a source and mention your product when prompted.
2. Find Citation Sources That Influence What Your Competitors Are Saying
An AI citation gap is when an AI model cites a particular source (such as a third-party review site, news outlet, or niche blog) to answers that mention the competitor, but not your brand.
For example, think of Monday and Asana as two competing project management softwares.
If ChatGPT recommends Asana but not Monday in a response that cites a niche tech blog to justify that choice, that blog will be Monday’s citation gap.
Identifying these gaps allows you to reverse engineer the sources of influence behind your competitors’ AI dictates of the most important commands.
Like networking and link building, you can organize outreach to showcase partnerships and talk about your brand to these relevant influencers.
To find quote spaces for your AI, stay at “Competitive Research” report and click “Sources” tab.
You will see a list of domains referring you and your competitors.
Claim “Lost” filter to see which sites are citing competitors but haven’t mentioned your product, and also filter by mention of a competitor.
Prioritize sources by looking for three signals:
- The number of the message it comes from
- Their traffic is organic
- The number of URLs cited on that site
Together, these tell you how powerful the source is in your category.
Click on any domain to expand it and see the exact pages your competitors are quoting.
Then group them by type to determine how to do it:
- Industry publications and niche blogs: Review specific competitor citation pages to understand the angle (collect, compare, best list)and suggest editors to be included in those pages.
- Review and compare sites: Find specific listings or comparison pages that refer to competitors and arrange access to have your product added directly to them.
- Social platforms like Reddit and Quora: Contribute directly to specific contexts that AI already draws from.
3. Spot Narrative and Emotional Gaps in Product Performance Reports
Narrative and emotional gaps occur when AI models associate competitors with beneficial features (such as reliability, security, or ease of use) more often than your product.
To find these gaps, use Product Performance reports.
To get started, open the “Product performance” report within the AI Visibility Toolkit.
Enter your domain, select a target location, and select a target language.
These include ChatGPT, SearchGPT, Google AI Mode, Perplexity, or Gemini.
|
Note: Product Performance Reports are tied to the specific domain, location, and language combination you select during setup, and each combination uses a single Product Performance analysis threshold. |
Identify the Key Business Drivers That Influence Your AI Narrative
The Key Business Drivers heat map within the Product Performance report shows which of these attributes (such as price, fulfillment speed, or product variety) AI talks more about each product in your category.
These are the decision-making factors that AI models use to differentiate you from competitors. When users ask for recommendations (such as “What is the best project management tool for small businesses?”), LLMs determine which brands are “for” specific attributes—price, speed of fulfillment, features, product variety, or value.
Notice the medal icon that marks each driver’s leader.
Scan for articles where the competitors are award-winning and your product is inferior. These are your narrative spaces.
Find emotional gaps with a question
Next, you can go deeper by analyzing your emotional gaps with specific questions.
The Narrative Drivers Report shows your brand sentiment (positive or neutral) from AI responses to a set of prompts and specific questions.
Scroll down to “Split by question “ to see your product’s position and sentiment score for specific consumer orders alongside each competitor.
At the top, use the following set filters to focus on niches—these two are especially important for competitive analysis:
- Answers | Non-brand: a style of question where you can see your and your competitors’ state of citation and general feeling
- Quotes | Non-brand: sources that AI cites for queries that are not brands in your category where your brand is not represented.
Of “Answers | Anonymous” view, you will see your product position, emotional points, and how both have changed over time; alongside the same data for every competitor.
You can also filter by mood swings, mood swings, and specific competitors to further narrow down the table.
For example, filtering to find information where Asana doesn’t appear but competitors rank #2-5 yields questions like:
- What is the best way to manage risk registers in a project tool?
- How do engineering teams manage cross-team projects without issue tracking?
- What are the effective ways to manage planning and quarterly resources?
These are your competitive positions.
In “Quotes | Non-brand” view, you can see the average experience of each product in all the instructions where a specific domain is cited.
For example, in this first row, Asana has 92% sensitivity and Monday has 67% sensitivity for all AI responses citing atlassian.com.
On Monday, that would be another quote gap to explore and try to improve its brand experience when mentioned on the Atlassian site.
Note: Unlike the Competitor Research report, which pulls from Semrush’s 239M+ database of information, Product Performance reports use a different approach. This creates up to 200 artificial intelligences designed to replicate real-world consumer testing conversations, giving you a controlled view of how AI perceives your brand when prospects compare options.
Improve Your AI Visibility Today with Semrush One
Many brands still don’t have a clear picture of where they stand in AI-generated responses, let alone where competitors have moved forward.
Measuring AI visibility is still a new field, and many teams are still figuring out where to start.
But don’t make it too difficult. Manage to find AI visibility gaps as an additional layer to your existing competitor analysis.
With Semrush One, you get a clear view of competitors and gaps in both AI Visibility and SEO. Use it to start finding vacancies today to take action and get ahead of the competition.
