Many websites hide content behind clickable elements to keep pages clean and load faster. Browse AI can extract this hidden content by recording your clicks and learning to reveal the data during extraction.
Types of click-to-reveal content
Expand/collapse sections: "Read more" links that reveal full text.
Hidden details: Product specifications that appear when clicked.
Dynamic panels: Content that switches in the same space when different options are clicked.
Tooltips and info boxes: Information that appears near the clicked element.
Hidden form fields: Additional inputs that appear based on previous selections.
Load on demand: Content that only loads from the server when requested.
Basic click-to-reveal extraction
Training your robot to click and extract
Train your robot to click and capture the dynamic content. The robot learns to click before extracting.
Start training your robot in Robot Studio
Click the element that reveals hidden content (button, link, icon)
Wait for content to appear (the robot records this pause)
Use Capture Text or Capture List on the revealed content
Example: "Read more" expansion
Scenario: Product descriptions with "Read more" links
Navigate to the product page
Find the truncated description with "Read more" link
Click "Read more"
Wait for full text to expand
Click Capture Text and select the full description
Label it appropriately
The robot will always click "Read more" before extracting
Multiple click-to-reveal sections
Sequential clicking
When a page has multiple expandable sections:
Click to expand the first section
Capture the revealed content
Click to expand the second section
Capture that content
Continue for all sections
The robot will click each in sequence
Example: FAQ pages with collapsible answers
Click the first question to reveal its answer
Capture the answer text
Click the second question
Capture its answer
Repeat for all FAQs you need
Label each clearly (e.g., "FAQ_1_Answer", "FAQ_2_Answer")
Note: If sections auto-collapse when opening another, that's fine - the robot will handle them one at a time.
Dynamic content panels
Content that replaces other content
Some sites show different content in the same space based on clicks:
Click the first option (e.g., "Description" tab)
Capture the content
Click the second option (e.g., "Specifications" tab)
Capture that content
Repeat until all content is captured.
Example: Product information tabs
Click "Overview" tab → Capture overview text
Click "Features" tab → Capture features list
Click "Reviews" tab → Capture review data
Each click replaces previous content
Robot learns to click each tab and extract
Show/hide toggles
Binary toggles
For simple show/hide switches:
Click the toggle or switch to "show" position
Capture the revealed content
If needed, click again to hide
Robot learns the toggle state needed
Example: "Show technical details" toggle
Find the "Show technical details" toggle or checkbox
Click to enable it
Technical specifications appear below
Capture the specifications
Robot will always enable before extracting
Extracting from multiple click combinations
Dependent clicking
When one click reveals options for another:
Click the first element to reveal options
Click a revealed option to show more content
Capture the final revealed content
Robot learns the complete click sequence
Example: Size charts
Click "Size Guide" link
Size chart panel appears
Click "International Sizes" within the panel
Conversion table appears
Capture the conversion data
Robot performs both clicks in sequence
Advanced tips
Efficiently handling many expandables
For pages with dozens of expandable sections:
Identify if there's a "Expand All" option first
If not, consider if you really need all sections
Train with most important sections
Use monitoring to detect changes
When clicking changes the URL
Some clicks navigate to new pages which is training your robot to navigate. While in come cases this might work, you should likely consider:
Training a second robot for that new page.