How to Set Up Browse AI Alerts When Webpage Data Changes
Create a Browse AI monitor, choose what counts as a change, and route alerts to email or a webhook. Follow the setup steps and fix common alert problems.
To get Browse AI alerts when webpage data changes, create or choose a robot that captures the exact content you care about, approve and save it, then add a monitor, choose its schedule and monitoring type, and enable email notifications or configure a webhook. A monitor only detects changes in the data its robot captures, and checks run on the schedule you choose.
1. Prepare a robot that captures the right data
A monitor compares what a robot captures over time. Before setting up alerts, make sure the robot extracts the specific fields or page content you need. If a price, availability label, or paragraph is missing from the robot’s capture, the monitor cannot reliably alert you when that item changes.
For a new robot, start with the page’s Origin URL and review the extracted preview before saving:
- Use Table Studio when the information is already visible on the page and can be captured as a table.
- Use Robot Studio when the page requires actions such as clicking, typing, entering a form value, or logging in before the information appears.
After training the robot, review its preview and save or approve it. Browse AI’s approval guide says monitoring is available after approval, and its first-robot guidance says scheduled runs do not begin until the robot is saved. See Browse AI Help Center for the current robot-building and approval instructions.
2. Add a monitor
- Open the robot in your Browse AI dashboard.
- Select Monitor.
- Choose Add monitor or Create new monitor. Browse AI also offers monitoring while creating a robot.
- Configure the monitor settings, then save to activate it.
The Monitor tab is also where you can edit an existing monitor. The exact wording or availability of controls may vary by account or interface.
3. Choose what counts as a change
| Monitor type | Best for | What it detects |
|---|---|---|
| Text monitoring | Prices, availability, lists, or page copy | Changes in captured text, lists, or HTML |
| Screenshot monitoring | Layout, styling, or other visual differences | Visual changes, with a sensitivity setting |
| Text monitoring with input parameters | Repeated searches or filtered views | Captured text changes for different input values, such as search terms or form filters |
Choose text monitoring if the values themselves matter. Choose screenshot monitoring if a visual difference matters even when extracted text stays the same. For screenshot monitoring, the documented sensitivity choices are small (1%), medium (5%), and large (10%). A larger threshold is less sensitive to minor visual changes. These are configuration thresholds, not guarantees that every account or monitor exposes every option.
For monitors with input parameters, set the custom values the robot should use for each run. For example, one robot may check a search-results page for several terms rather than requiring a separate robot for every term.
4. Set the schedule
Browse AI’s guide lists hourly, daily, weekly, and minute frequency choices, plus day-of-week options and a specific time or Any time. Availability can depend on the account or interface. Set the cadence based on how often the information changes and how quickly you need to know: a schedule determines when a change can be detected, so it is not an instant alert system.
Check the Origin URL and any input parameters before saving. A wrong URL, filter, or form value can make the robot monitor the wrong page or dataset while still appearing to run successfully.
5. Turn on email alerts
- Open the robot and select Monitor.
- Add a monitor or edit the existing one.
- Enable Notify me by email if there is a change in a captured text.
- Select Save.
Browse AI says the notification email shows what changed. Its monitoring guide also describes changed content as highlighted in results: green for new content, yellow for modified content, and red for removed content. Results are retained in your Browse AI account for historical tracking.
6. Route changes to a webhook
Use a webhook when a change should trigger an external workflow instead of, or alongside, an inbox notification. Browse AI documents the taskCapturedDataChanged event for monitoring changes.
- Open the approved robot and select Integrate.
- Find Webhooks and add the destination URL provided by your receiving service.
- Select the
taskCapturedDataChangedevent. - Save the integration.
The destination must provide a webhook URL. Browse AI’s guide gives Slack, Zapier, and Make as examples of services that can receive webhooks. Treat a webhook URL like a secret: avoid putting it in public code, a public issue, or a page that others can access. See Browse AI’s webhook setup guide for its current dashboard steps.
7. Verify that the monitor is useful
- Confirm the robot preview contains every field you expect to watch.
- Check that the Origin URL and any text-monitoring input values point to the intended page and data.
- Confirm the monitor is saved and its schedule reflects the detection cadence you need.
- Verify the email notification option is enabled, or confirm the webhook event and destination are saved.
- When a change is detected, inspect the recorded result and highlighted changes to distinguish new, modified, and removed content.
Choosing text or screenshot monitoring, email or webhook
Text and screenshot monitoring answer different questions: text monitoring asks whether captured content changed, while screenshot monitoring asks whether the rendered appearance changed. Email is the simplest destination for a person who wants to review a change. A webhook is more suitable when another service should process the event or start an automation. More frequent schedules can surface changes sooner, but they also mean more frequent checks and potentially more alerts to review.
Troubleshooting
| Problem | Likely cause | What to check |
|---|---|---|
| No email arrives | Email notification is off, the address is incorrect, or the message was filtered | Confirm the monitor’s email option is enabled and saved, verify the account email address, and check spam. |
| An obvious page change is not detected | The robot does not capture the changed field, the page structure changed, or the schedule has not run since the change | Review the robot’s captured fields and preview, check the page structure, and confirm the schedule. |
| Alerts arrive later than expected | The monitor checks on its configured schedule | Review the available frequency and choose a cadence appropriate to the information’s change rate and your needs. |
| The monitor tracks the wrong results | The Origin URL or custom input parameters do not match the intended page or filter | Check the URL, search terms, filters, and form inputs saved in the monitor. |
| A webhook workflow does not run | The endpoint URL, selected event, or receiving service setup may be incorrect | Confirm the URL in Integrate > Webhooks, select taskCapturedDataChanged, save, and verify that the receiving service is configured to accept the event. |
| The monitor cannot be added | The robot may not be approved or saved | Complete the robot’s approval/save step, then return to its Monitor tab. |
Browse AI’s monitoring and notification guides specifically recommend checking the captured fields, site structure, schedule, spam folder, account email address, and notification setting when alerts do not behave as expected.
Performance, reliability, and cost considerations
Monitoring reliability starts with capture quality. Keep the robot focused on stable, meaningful fields, and revisit it if the target site changes its layout or interaction flow. If the page needs a login or form interaction, configure the robot for that path before relying on scheduled checks. Choose a schedule that fits the data’s change rate; no schedule can detect a change before the next check.
Consider alert volume as well as detection speed. A sensitive visual threshold and frequent checks may create more changes to review. For a value that changes frequently, use text monitoring and the fields that matter; for appearance changes, choose screenshot monitoring and select the sensitivity that suits the level of visual noise you can tolerate.
The supplied Browse AI research does not establish current plan prices, run limits, or webhook delivery guarantees. Check Browse AI’s current account and product pages for plan-specific limits before designing a high-volume workflow.
Or skip the browser setup
If your goal is a clean screenshot of a page rather than a scheduled change alert, ScreenshotNeo provides a website screenshot API and MCP server. One GET request returns a PNG, JPEG, WebP, or PDF. It can remove cookie banners, newsletter popups, and chat widgets before capture, and it bills only clean shots: bot checks or CAPTCHAs, blank pages, timeouts, failed loads, and cache hits cost nothing. Each response identifies the page verdict and billing status in headers. Its MCP server provides take_screenshot, get_page_info, and capture_pdf for Claude, Cursor, and other MCP clients.
For example, save a screenshot as WebP with cURL:
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webp
Python:
import requests
r = requests.get(
"https://api.screenshotneo.com/v1/shot",
params={"access_key": "YOUR_API_KEY", "url": "https://stripe.com"},
timeout=90,
)
open("shot.webp", "wb").write(r.content)
Node.js:
const q = new URLSearchParams({ access_key: 'YOUR_API_KEY', url: 'https://stripe.com' });
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);
if (!res.ok) throw new Error(`Screenshot request failed: ${res.status}`);
await Bun.write('shot.webp', res);
See the ScreenshotNeo API documentation for request options and authentication. ScreenshotNeo includes 1,000 screenshots a month free with no card; paid plans start at $5 for 3,000 shots. Yearly billing gives two months free, and every feature is included on every plan.
Sign up for ScreenshotNeo and get 1,000 free screenshots a month, with no card required.
FAQ
Does Browse AI watch the whole page automatically?
The monitor compares the content the robot was trained to capture. Configure the robot to include the fields or visual area you want to track.
Can one robot monitor different searches or filters?
Text monitoring supports custom input parameters in documented configurations, so a monitor can use values such as different search terms or filters.
Can I use email and a webhook together?
The guides document email notification and webhook integration as available routes. Configure each where its setting is available and confirm both are saved.
Will I be alerted at the exact moment a page changes?
No. A monitor checks according to its schedule, so detection occurs on a scheduled run after the change.
Sources
- Browse AI Help Center, How to set up monitoring for website changes, published December 5, 2025.
- Browse AI Help Center, How to set up monitoring notifications, published February 21, 2025.
- Browse AI Help Center, How to approve a robot, published September 1, 2026.
- Browse AI Help Center, Building your first robot, updated as of October 3, 2026.
- Browse AI Help Center, Webhooks: Set up guide, published March 13, 2026.


