Share this
by The Notified Team on Nov 5, 2026, 2:41:52 PM
What does AI say about your brand when you’re not in the room? A brand narrative audit can help communicators understand what AI knows, where that information comes from and where the story needs attention.
You know how your company describes itself. You know the capabilities you want customers to understand, the topics you want to be known for and the messages your communications team has carefully built.
Now ask ChatGPT what your company does. Try the same question in Claude or Gemini. Does the answer sound like the company you know today?
That’s the starting point for an AI brand narrative audit. It isn’t simply an exercise in seeing whether your brand appears in an AI-generated answer. The real value comes from understanding what AI associates with your company, which sources are informing that narrative and where gaps or outdated information may be influencing what audiences see.
That was a recurring theme in our recent webinar, How to Measure AI Visibility & Brand Influence: Key Metrics, Signals & Sources. The conversation with communications and AI visibility experts moved beyond simply asking, “Are we showing up?” and into the much more useful question: What can we learn from how we’re showing up?
1. Start With the Questions Your Audience Might Ask
You don’t need a sophisticated AI visibility platform to conduct an initial narrative audit. Start by asking a consistent set of questions across several AI engines.
- What does [Brand] do?
- What is [Brand] best known for?
- What topics is [Brand] considered an expert in?
- How does [Brand] compare with [Competitor A] and [Competitor B]?
- Are there outdated descriptions or legacy narratives about [Brand]?
These prompts come from the broader narrative-audit exercise in The AI-First PR Content Playbook, which includes questions covering brand positioning, expertise, competitive context, narrative gaps, media signals and more.
Run the same core questions in ChatGPT, Claude and Gemini. You’re looking for patterns rather than treating any single response as definitive.
Does one description keep appearing? Are you associated with the expertise you want to own? Does an outdated product name continue to surface? Is a competitor consistently recommended ahead of you for a category you consider important? Those patterns establish your baseline.
Get Your AEO Program On Track
Tips and tricks to help you optimize your content and improve AI visibility in less than 90 days.
2. Look Beyond Whether Your Brand Appears
Visibility matters, but presence alone doesn’t tell you much about the story being told. During the webinar, the speakers discussed citations, citation rates, brand mentions, share of voice or share of answer, brand traits and average rank. Which metrics matter most depends on what the organization is trying to achieve.
For a narrative audit, simplify that into four questions:
- Are we there? Does your brand appear when somebody asks a relevant category, expertise or recommendation question?
- Where are we? If the answer includes several companies, where does yours appear relative to competitors?
- What is AI saying about us? Note repeated descriptions, strengths, weaknesses and areas of expertise.
- Is it accurate? Check names, products, capabilities, leadership information, positioning and other material facts against what is true today.
That last question can lead to the most valuable part of the exercise.
3. Follow the Sources Behind the AI Answer
When an AI answer is wrong, incomplete or outdated, it’s tempting to focus on the model itself. Instead, investigate the information feeding it.
In the webinar, Marcel Goldstein described conducting this kind of analysis for brands that were surprised by outdated information appearing in AI answers.
Corporate websites accumulate history. Old product pages survive migrations. Press releases capture a point in time. Executive biographies become stale. A company can change its positioning without every previous description disappearing from the web.
Third-party sources complicate the picture further. The webinar highlighted review sites such as G2 and Trustpilot, social platforms including LinkedIn, Reddit and YouTube, trade media and owned content as potential influences. Marcel also cautioned against treating any source as universally important because the mix varies by company and category.
That’s why “Which channels does AI prefer?” isn’t a particularly useful audit question. A better one is: Which sources are shaping AI’s answers to the questions that matter to our brand?
Follow the citations where they’re available. Look for repeated sources. Then compare those sources with the narrative you’re seeing in the answer. You may find that the AI problem is actually a content-governance problem.
4. Check How Old AI’s Version of Your Story Is
This deserves particular attention because old information doesn’t necessarily disappear when new information is published.
Brittany Brady discussed an interesting example during the webinar: citation reporting for GlobeNewswire can surface AI activity around press releases published years earlier.
Her point wasn’t simply that old content can remain visible. If a five-year-old announcement is answering a question today, communicators should ask what question is causing that information to surface and whether the answer is still accurate.
New Notified research adds useful context. An analysis of more than 25,000 GlobeNewswire press releases and nearly one million AI citations found that more than 56% of citations from ChatGPT and Claude arrived after the first week. The research also found that both models mostly cited the version of a release they crawled first. The larger dataset and model-level findings are available in Written for People. Read by Machines.
For communicators, this makes historical content worth examining as part of an AI audit. Review older press releases, newsroom pages, product descriptions, executive biographies and corporate messaging that could still be informing how your company is understood.
5. Turn What You Find Into Action
You don’t need an elaborate scoring system for your first audit. Sort the findings into four buckets:
Accurate and current: The narrative reflects the company today. Note which sources support it and continue reinforcing the message.
Accurate but incomplete: AI understands part of your story but is missing an important capability, proof point or area of expertise. This is a content opportunity.
Outdated: Identify where the older information originates. Determine whether the source can be updated or whether newer authoritative content needs to clarify the current position.
Incorrect: Trace the claim before responding. If inaccurate information originates on an owned page, start there. If it comes from third-party coverage or another external source, consider what corrective or authoritative information you can realistically influence.
This is where AI visibility measurement becomes useful to a communications team. The audit gives you evidence to inform what you publish next instead of simply adding more content to an already crowded ecosystem.
How Often Should You Audit Your AI Brand Narrative?
AI’s understanding of a company changes as new information enters the ecosystem, so a narrative audit shouldn’t be a one-time exercise.
The AI-First PR Content Playbook recommends establishing a baseline, repeating core prompts every 90 days and conducting targeted audits following major events such as launches, mergers, crises or rebrands.
Keep the prompts consistent from one audit to the next. That makes it easier to see whether descriptions are changing, newer sources are beginning to appear and important narrative gaps are closing.
The goal isn’t to control every sentence an AI system generates. It’s to understand the information environment surrounding your brand well enough to make smarter communications decisions. Once you know the gap between the story you’re telling and the story AI is telling about you, you know where to start.
Put Your AI Brand Narrative to the Test
Ready to go deeper? Download The AI-First PR Content Playbook for the complete brand narrative audit, a 90-day AEO action plan and practical guidance for strengthening the signals AI systems use to understand your brand.
You can also catch the playback of Watch How to Measure AI Visibility & Brand Influence: Key Metrics, Signals & Sources for the expert conversation that inspired this exercise.
About Notified
Notified, Equiniti’s market engagement business, helps public relations and investor relations professionals earn attention, measure engagement and deliver results through its integrated platforms Content OS™ and IR Hub™. Click here to learn why more than 50% of the S&P 500 trust Notified’s solutions and services to drive demand and attract capital.
.png)
How to Run an AI Brand Narrative Audit
.png)
Do FAQs Improve AI Visibility? Our Data Says Not Really
.png)
5 Things IR Teams May Be Missing Before Their Next Earnings Call
