URLs

On the URLs page, you can identify specific pages that consistently influence AI responses and discover content formats that perform best in your industry:

• Source retrievals over time: Shows trends over time for the top 5 URLs, helping you track which specific pages gain or lose influence with AI models.
• URL movers graph: Lists URLs sorted by number of retrievals. You can toggle through different views:
  • Top: URLs with the highest number of retrievals in the selected timeframe
  • New: URLs retrieved for the first time in the selected timeframe
  • Trending: URLs showing the fastest growth in retrieval activity
  • Losing: URLs showing the largest decline in retrieval activity
• URL types chart: A category breakdown covering website page types, such as Article, Comparison, Listicle, Product Page, and others.

In the URL table below, you’ll find:

• URL: The exact webpage used as a source, with its page title.
• URL type: Automatic classification of the page type (Homepage, Article, Listicle, Comparison, Profile, and so on).
• URL tags: Tags you added to this URL.
• Domain type: The type of website the URL belongs to (Editorial, Corporate, user-generated content (UGC), Reference, and so on).
• Brand mentioned: Whether the brand you selected in the filter is mentioned on the page. Shown while a brand is selected.
• Mentions: Which tracked brands are mentioned on this URL.
• Retrievals: Total number of times this URL appeared as a source across all chats.
• Retrieved: Percentage of chats where this URL appeared as a source.
• Cited: Percentage of chats where this URL was cited inline in the answer.
• Citation position: Where in the answer this URL tends to be cited, as an average rank. #1 means it’s cited first.
• Citation rate: Average number of times the URL was cited per chat in which it was used as a source.
• Total citations: Total number of inline citations of this URL across your filters and time period.
• Citation share: This URL’s share of all inline citations across the URLs in your current view.
• Updated: The last time the content on this URL changed.

​

URL types

URL Type offers a more granular classification by identifying the specific type of content on a given page. This allows for a deeper analysis of the context in which your brand is mentioned.

​

Changing URL classification

Peec automatically classifies every URL based on its content, but we know your taxonomy may differ from ours. For example, a URL we label as “Product Page” might actually be a “Listicle”. With custom classification, you can now apply your own labels to every URL. You can override this at any time:

1. Find the URL in the URL table.
2. Click the classification label on that row (e.g., Listicle, Comparison, etc.).
3. A search-first picker opens, search existing types (including any custom ones you’ve created), or create a new one inline.
4. The override saves immediately and appears everywhere the source appears.

​

URL metrics

The URL table reports Retrievals, Retrieved, Cited, Citation position, and Citation rate. Each URL’s detail page shows Retrievals, Citation rate, and Citation position.

​

Retrievals

Retrievals is the total number of times a URL was used as a source, regardless of whether it was cited directly in the AI’s main answer. This metric indicates which URLs are most popular and most credible to AI models. Retrievals over your selected timespan are calculated as: Retrievals (URLs) = Total responses where URL was used as a source

​

Retrieved

Retrieved is the percentage of chats in which the URL appeared as a source. Where Retrievals gives you the count, Retrieved puts it against every chat in your selected time span and filters, so pages from different time spans or prompt sets compare directly. Retrieved is calculated over your selected time span as: Retrieved (URLs) = (Chats where the URL was used as a source / Total chats) × 100

​

Cited

Cited is the percentage of chats in which the URL was cited inline in the AI’s answer. It uses the same denominator as Retrieved. A chat can only cite a page it retrieved, so Cited is never higher than Retrieved. Cited is calculated over your selected time span as: Cited (URLs) = (Chats where the URL was cited / Total chats) × 100 For example:

• If a page was retrieved in 40 of 100 chats and cited in 10 of them, Retrieved is 40% and Cited is 10%.

​

Citation position

Citation position is where in the answer a URL tends to be cited. In each chat, the sources cited inline are ranked by where their first citation appears: the first source cited is #1, the next is #2, and so on. A URL’s citation position is the average of those ranks across the chats in your selected time span that cited it. Lower is better, and a move toward #1 shows as an improvement. Citation position is calculated over your selected time span as: Citation position (URLs) = Sum of the URL's rank in each chat that cited it / Chats that cited the URL For example:

• If a URL was cited first in two chats and third in another, its citation position is #1.7 ((1 + 1 + 3) ÷ 3).

​

Citation rate

Citation rate measures the average number of times a specific source is explicitly cited within AI responses over your selected time span. This metric indicates how often AI models find your content relevant enough to reference multiple times. Multiple citations in a single response indicate that AI models consider your content highly relevant and trustworthy for the topic at hand. The citation rate is calculated over your selected time span as: Citation Rate (URLs) = Total citations of that URL / Total responses where URL was used as a source

​

Citation share

Citation share measures the percentage of all inline citations attributed to a specific URL across AI responses over your selected time span. This metric helps you understand how much of the total citation activity a particular URL accounts for. A URL can be cited multiple times within a single response, and each citation contributes to its overall share. Citation share is calculated over your selected time span as: Citation Share (URLs) = (Total citations of that URL / Total citations across all URLs) × 100

​

Total citatons

Total citations measures the total number of times a specific URL is explicitly cited within AI responses over your selected time span. This metric helps you understand how frequently AI models reference your content when generating responses. A URL can be cited multiple times within a single response, and each citation counts toward the total. Total citations is calculated over your selected time span as: Total Citations (URLs) = Sum of all inline citations of that URL across the filtered responses

​

URLs detail page

When you open the URLs section, you see a list of pages that have been used as a source by the different AI models in their answers. Clicking any URL takes you to its Detail Page, providing further insights into how the models have used it. You can open this detail page for each URL by clicking it. Click “View page content” in the header to open a dialog showing the full text Peec retrieved from this URL. This is the same content that AI models read when they use this page as a source. Alternatively, you can click the URL to be redirected to the page where the content is located. The URL Detail Page will show you the following information:

1. URL and Title: The page title that is being pulled as a source, and the URL that you can visit.
2. URL Overview: Different metrics and information, such as retrievals, used as source, cited, citation rate, citation position, retrievals, number of prompts currently using the page, and first and last time the URL was seen, with their respective absolute change compared to the previous period you are filtering for.
3. Retrievals Chart: Line chart showing the number of times that URL appeared as a source in AI answers over the selected time span.
4. Retrievals by Model: Number of times this URL is retrieved in chats.
5. Prompts Table: Which prompts triggered this URL as a source, regardless of whether it was cited or not, and to which topic that prompt belongs.
6. Brands Mentioned: Which brands appear in the source content, and how frequently.
7. Chats: The actual AI answers where this URL was cited or used as a source.

You can use this information to:

• Track citation changes over time to know when to push more content on a topic, or pull back.
• Compare model usage across periods to decide where to optimize
• Study the prompts driving citations to reverse-engineer what questions are making this source relevant

​

Bookmarking URLs

Within Peec, you can bookmark URLs to build a personalized watchlist of sites you want to monitor closely, such as key competitors, your top owned assets, target publications for outreach, or any other URLs that matter to you. Bookmarking a URL is simple: Every source row has a bookmark icon in the left column. Clicking it adds the source to your watchlist under the Bookmarked tab.