HubSpot AEO Grader (Now AI Search Grader): What It Scores and How to Use It

HubSpot's AEO Grader, now named AI Search Grader, is a free tool that checks what ChatGPT, Perplexity and Gemini say about your brand and turns it into a score out of 100. You enter your company name, location, product or service and industry, and in under two minutes you get a composite score, a breakdown across five dimensions and a written interpretation of the results.

Checked against HubSpot's AI Search Grader page, the HubSpot AEO product page and the HubSpot Knowledge Base on October 20, 2026.

The grader is a snapshot. It tells you how answer engines describe you today. If you want to track that over time against real buyer prompts, HubSpot sells a separate product, HubSpot AEO, which we compare below.

What is the HubSpot AEO Grader now called?

HubSpot's page is titled AI Search Grader. The old address, hubspot.com/aeo-grader, now redirects to hubspot.com/ai-search-grader, so both names lead to the same tool. Many articles still call it the AEO Grader, and HubSpot's own page description still uses the phrase "Answer Engine Optimization Grader."

AEO stands for answer engine optimization. HubSpot defines it as the practice of improving how your brand appears in AI answers on platforms like ChatGPT, Gemini and Perplexity. Older guides expand the acronym as "AI Engine Optimization," but that is not how HubSpot uses it.

What the AI Search Grader measures

HubSpot describes the grader as a free, one-time check of what ChatGPT, Perplexity and Gemini say about you "based on their training data." It sends your brand information to each model and asks how the model characterizes your brand.

How the score out of 100 is calculated

HubSpot publishes the weighting. Sentiment carries the most weight because, in HubSpot's words, it measures how AI characterizes your brand, not just whether it recognizes you.

Two details are easy to miss:

• Contextual accuracy and source credibility are not separate scores. Source credibility sits inside Sentiment, and source quality sits inside Presence Quality.
• A low Market Competition score usually means answer engines have weak or inconsistent signals about where your brand stands, according to HubSpot.

How to run the grader, step by step

1. Go to hubspot.com/ai-search-grader.
2. Enter your Company Name, Location, Product or Services (a few words, the way customers would describe it) and Industry.
3. Start the analysis. The tool queries GPT-5.4 mini, Perplexity and Gemini at the same time.
4. Complete the short form to see the full report.
5. Read the composite score, the five-part breakdown and the written interpretation.
6. Repeat with two or three competitors using the same location, product wording and industry, so the comparison is fair.

Write down exactly what you entered. Small wording changes in the product or industry field change the context the models answer in.

How to read your results

Treat the first run as a baseline, not a verdict. A few ways to use each part:

• Sentiment (40): Read the written interpretation for the phrases the engines use about you. If they repeat an old complaint or an outdated description, that points to the public sources you need to update or answer.
• Presence Quality (20): A low score here suggests the engines know little detail about you. Clear product pages, documentation and third-party coverage give them more to work with.
• Brand Recognition (20): Newer and smaller brands often score low simply because they appear less in training data. Compare against similar-sized competitors rather than category leaders.
• Share of Voice (10) and Market Competition (10): Check whether the label (Leader, Challenger, Niche Player) matches how you position yourself. A mismatch is a messaging gap, not only a visibility gap.

Keep the grader's limits in mind:

• It is one point in time. HubSpot calls it a one-time snapshot of how engines represent you right now.
• It reflects training data. HubSpot says the grader asks each model how it characterizes your brand based on its training data, so recent changes to your site may not show up.
• It covers three engines. The grader lists GPT-5.4 mini, Perplexity and Gemini. Claude and Google AI Overviews are not on the list.
• It does not track your buyers' prompts. It asks how each model characterizes your brand. It does not show which buyer questions you appear in.

AI Search Grader vs HubSpot AEO

HubSpot sums up the difference this way: the grader answers how AI represents your brand right now, and HubSpot AEO answers how that is changing and what to do about it. HubSpot lists AEO as a beta product.

HubSpot's product page says HubSpot AEO recommendations tell you when to create a new page, fix a technical gap, publish a video or reach out to third-party sites. Its citation analysis shows which domains, pages and content types answer engines cite for you and your competitors.

HubSpot AEO limits by plan

HubSpot's Knowledge Base lists these limits (last updated October 2, 2026):

The standalone and Marketing Hub versions both include brand visibility, prompt tracking, competitor analysis, citation analysis and recommendations. The Marketing Hub Professional and Enterprise version adds blog post generation and CRM-based prompt suggestions. HubSpot's product page says you can buy more prompts at any time.

To find it in a Marketing Hub Professional or Enterprise account, click More, then go to Marketing > AEO (if More does not appear, go to Marketing > AEO directly), and click Get started. Setup asks for your brand name, domain, brand name variations (comma separated and case sensitive) and business type, then competitors, products, ideal customer profiles and prompts. You need Marketing Access permissions. The Dashboard tab shows brand visibility, sentiment, the competitor landscape and citation analysis. HubSpot advises reviewing results across several days or weeks, because responses change over time.

Why this matters for search traffic

Fewer searches end in a click. SparkToro's June 8, 2026 study, based on Similarweb clickstream data, found that 68.01% of Google searches in the first four months of 2026 ended without a click. HubSpot's grader page calls this the zero-click funnel, where an answer engine may be the first and sometimes only touchpoint between a brand and a buyer. A grader score gives you a starting read on that touchpoint.

A worked example (hypothetical)

Imagine a regional accounting software company runs the grader for the first time:

1. It scores well on Brand Recognition but lower on Sentiment, and the written interpretation repeats a description of a product feature the company retired.
2. Market Competition labels it a Niche Player, though it sells to mid-sized firms in several countries.
3. The team updates its product pages and help docs, asks recent customers for reviews, and pitches a comparison article to an industry publication.
4. It reruns the grader each quarter with the exact same inputs, and grades two competitors the same way.
5. Once it wants to see which buyer questions mention it, it starts the HubSpot AEO trial and tracks 25 prompts.

The grader showed what to fix. The tracking tool shows whether the fixes are working.

Turning a grader score into a plan

INSIDEA is an Elite HubSpot Partner and the growth operating system for modern businesses. If you want help turning your AI Search Grader results into a content and visibility plan, our growth marketing team can work through them with you.