How to Use Browse AI to Track Product Price Changes
Train a Browse AI robot to capture product prices, schedule checks, and deliver change alerts. Learn how to validate the data and troubleshoot common issues.
To track product price changes with Browse AI, train a robot to capture the product identity and displayed price, test the extraction, then add a monitor and choose its schedule, change sensitivity, and delivery options. Check that each run captures the intended product and price before relying on alerts.
1. Choose what to monitor
Start with the page that matches your question:
- One product page: monitor a specific item or variant and its displayed price.
- A search or category page: monitor a set of listings when you need to follow several products or competitors.
Decide which fields you need before training. A useful minimum is the product name or other identity, the price, and, where relevant, the selected size, color, or variant. Consider availability as well if a price is meaningful only when the item can be purchased. Browse AI describes extracting structured product data and monitoring individual product pages, listings, and search results. The available capabilities can depend on the account and plan.
2. Train and test a Browse AI robot
- Open Browse AI and create a robot for the product page or listing page you selected.
- Follow the robot setup flow to show it the fields to capture, such as product identity and price.
- Run the robot once and inspect its output. Compare the captured values with the page, including the product variant and currency.
- Correct the selected fields or extraction pattern if the robot picked a sale price, a unit price, a price range, or a different product than intended.
Do not rely on a monitor until a manual run captures the right data. A page can display several prices at once: list price, discounted price, price per unit, subscription price, or prices for different variants. Record enough identity information to tell which value belongs to the product you mean to track.
3. Add a monitor and set its schedule
Browse AI documents two ways to start monitoring: select monitoring during robot creation, or open the Monitor tab for an eligible robot and add a monitor. The robot must capture the desired fields before scheduled checks can provide useful results.
- Choose the option to monitor while creating the robot, or open the robot’s Monitor tab and add a monitor.
- Select a check schedule and the change sensitivity offered in your account.
- Confirm the monitor’s timezone. Its configured timezone controls the schedule; it may differ from your browser’s timezone.
- Save the monitor, then review its first scheduled results and notifications.
Browse AI’s price-monitoring page describes schedules from every minute to monthly. Treat that as a vendor-described range, not a guarantee that every interval is available for every account: check the current choices in the dashboard and your plan. Choose a cadence that matches how quickly you need to react. More frequent checks can produce timelier observations, but do not make an incorrectly selected field more accurate.
4. Choose how to receive changes
Browse AI documents email notifications and data delivery through webhooks or integrations. Its e-commerce material also describes sending structured data to spreadsheets and other connected tools. The destinations available to you can depend on your current account and plan, so verify them in the dashboard.
For a simple watchlist, email may be enough. For a downstream price history or alerting workflow, send structured results to a destination your team already uses, if it is available. Include product identity with price data so that a change can be attributed to the right item.
5. Validate results and maintain the robot
- Compare robot output with the live page after setup and after meaningful page changes.
- Check the selected variant, currency, discount state, and whether the captured value is a unit price or full price.
- Review monitor runs and notifications rather than assuming every scheduled run reflects the intended field.
- After a site redesign, product-page change, or variant change, run the robot again and update its extraction if needed.
- If location or account context affects the displayed price, check whether the robot and account support the context you need and validate the result from that context.
Browse AI recommends testing robots and reviewing monitors. Keeping the extraction aligned with the page is practical maintenance: a robot that still runs can nevertheless select a different field after a layout change.
Configuration choices at a glance
| Choice | What to decide | Why it matters |
|---|---|---|
| Page type | Individual product, search results, or category/listing page | Determines whether you follow one item or a changing set of listings. |
| Captured fields | Product identity, price, variant, currency, and optionally availability | Helps distinguish the intended price from nearby values and other products. |
| Schedule | How often to check, using the intervals currently offered | Sets how quickly a change may be observed. |
| Change sensitivity | The monitor’s available threshold or sensitivity setting | Controls what differences are treated as changes; confirm the effect in the current interface. |
| Timezone | The timezone configured for the monitor | Defines when scheduled checks run. |
| Delivery | Email, webhook, integration, or another option shown in the account | Determines where change information reaches your workflow. |
| Context | Relevant location, account, or product variant | Some pages may show different prices or availability in different contexts. |
Common problems and fixes
| Problem | Likely cause | What to do |
|---|---|---|
| The captured price is wrong | The robot selected a nearby field, such as a list price, unit price, or another variant. | Run the robot manually, inspect the page alongside its output, and retrain it on the intended price and product identity. |
| The monitor reports changes that are not meaningful | The page contains dynamic content or the selected field includes changing text beyond the price. | Review the extracted fields and sensitivity setting. Capture a more specific price field and enough product context to validate it. |
| A product disappears or the output is empty | The listing changed, the product is unavailable, or the page layout no longer matches the robot’s extraction. | Open the page, run the robot again, and update the extraction or monitored page as appropriate. |
| Checks run at an unexpected time | The monitor uses its configured timezone rather than the browser’s timezone. | Review the monitor timezone and schedule in the account settings. |
| No alert arrives | The notification destination may not be configured, the change may not meet the sensitivity rule, or the selected delivery option may not be enabled for the account. | Inspect monitor results, sensitivity, and notification or integration settings; confirm current plan availability. |
| A scheduled monitor does not behave as expected after a redesign | The page structure or product fields changed. | Test the robot against the current page and revise the extraction before trusting later runs. |
Performance, reliability, and cost considerations
A shorter interval can help you notice a price movement sooner, while a longer interval may be sufficient for products whose prices change infrequently. Browse AI’s published material describes a range of monitoring schedules, but the actual choices and account limits should be checked in the product. Pick a cadence based on how quickly the information needs to reach you, and make sure the captured field is correct first.
Scheduled extraction is only useful when the page remains accessible and the robot continues to identify the same product and field. Review runs and test again after page or variant changes. Do not treat a notification as proof of a market-wide price: it represents the data captured from the page and context you configured.
The research material does not establish a price for Browse AI or which monitoring and delivery options are included in each plan. Check the current account pricing and feature availability before designing a workflow around a particular schedule, integration, or volume.
Or skip the browser setup
If you need a clean screenshot of a product page alongside your extracted price, ScreenshotNeo is a website screenshot API and MCP server. A screenshot can help you keep visual context for a page, but it does not replace Browse AI’s structured extraction and scheduled price monitoring.
For an on-demand capture, make one GET request:
curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webpSee the ScreenshotNeo API documentation for request options. ScreenshotNeo removes cookie banners, popups, and chat widgets before the shot; bot checks, blank pages, and failed loads are never billed; an MCP server lets AI agents take screenshots; and 1,000 screenshots a month are free with no card, with paid plans starting at $5 for 3,000. Sign up for free.
FAQ
Can I monitor a product listing instead of one product page?
Browse AI describes monitoring listings and search results as well as individual product pages. Choose the page type that matches whether you want to follow one item or a changing set of results, then validate the captured product identities.
Does the monitor use my local browser timezone?
Not necessarily. Browse AI’s documentation says the monitor’s configured timezone governs its schedule.
Can I send results to a spreadsheet or webhook?
Browse AI documents webhook and integration delivery, and its e-commerce material describes spreadsheets and other connected tools. Check your current account for the specific destinations available.
Should I monitor price alone?
Usually capture enough product identity to distinguish variants and listings. A price without its product or variant context can be misleading when the page contains multiple options.
