How to Automate Competitor Price Monitoring with n8n and AI

Watching competitor prices by hand is a job nobody schedules and everybody regrets skipping. The n8n tutorials that promise to fix it mostly do two things: bolt on a paid scraping vendor, then alert you on every single price change. The vendor is an avoidable cost, and the every-change alert trains you to ignore the channel within a week. A useful n8n competitor price monitoring workflow does the opposite. It alerts only when a change matters, and it tells you what to do about it.

Three things separate a workflow you keep from one you mute. A significant-change threshold so trivial fluctuations stay silent. An AI layer that recommends a response instead of just reporting a number. And a daily price-history log so a single alert has trend context behind it.

This is the build: fetch prices daily, compare against the last value, alert on real moves, add an AI recommendation, and log everything.

What you can automate in price tracking

Once n8n checks prices on a schedule, the monitoring becomes a system instead of a chore:

• Fetch a list of competitor product pages daily
• Extract the current price from each page
• Compare today's price against yesterday's stored value
• Alert only when the change crosses a threshold
• Attach an AI-recommended response to every alert
• Log every daily price to a sheet for trend charts
• Flag a product that's been undercutting you for several days running

The whole point is selectivity. A workflow that pings you on a $0.10 fluctuation is worse than no workflow, because you'll stop reading it.

The price-monitoring pipeline

Schedule Trigger → Fetch pages → Extract price → Compare to last
       → Threshold gate → AI response → Alert → Log to Sheet

Run daily. Pull each page. Read the price. Compare it to the last logged value. Only meaningful moves continue to the AI and the alert. Everything gets logged.

1. Trigger and fetch

A daily Schedule trigger is the right cadence for most pricing. Hourly is overkill and raises your odds of getting rate-limited or blocked by the target site. One honest caveat about scheduling: a Schedule trigger that fires while the previous run is still in flight will drop the new run, so if your fetch step is slow across a long product list, give the daily run enough headroom or split the list.

Fetch each page with an HTTP Request node. You don't need a paid scraper for most retail and DTC pages; a plain request with a sensible user-agent header gets the HTML. Reserve a scraping vendor for the few sites that hard-block automated requests.

2. Extract the price

Pull the price out of the HTML. An HTML Extract node with the price element's CSS selector works for static pages. For pages that render price in JSON-LD, parse the <script type="application/ld+json"> block instead, which is more stable than a CSS selector that breaks on every redesign.

3. Compare against the last value

Read the product's last logged price from your Google Sheet. Compute the percentage change. This comparison is what makes the threshold possible, so the stored history isn't a nice-to-have, it's the mechanism.

4. The threshold gate (kills the noise)

A Filter node checks whether the percentage change exceeds your threshold, say 3%. Below it, the price still gets logged but no alert fires. Above it, the item continues to the AI and the alert.

5. Add the AI recommendation

Here's the layer the scrape-and-alert tutorials skip. Feed the old price, the new price, and your own price into an OpenAI node and ask for a recommended response: hold, match, or undercut, with a one-line rationale. The alert now reads "Competitor X dropped 12% to $42; you're at $49; recommend hold, their drop looks like clearance not a permanent cut" instead of a bare number. The Competitor Price Intelligence template generates exactly this response recommendation so you act before revenue shifts.

6. Alert and log

Send the alert to Slack or email with the change and the recommendation. Then append today's price for every product, including the ones below threshold, to the history sheet. The full log is what turns a single alert into a trend you can read.

Implementation patterns worth copying

Pattern: log everything, alert selectively

Two different data flows share one workflow. The log captures every daily price for every product, unconditionally. The alert fires only past the threshold. Don't conflate them; the log's value is the unfiltered history, the alert's value is the filter.

Pattern: JSON-LD over CSS selectors

Retail pages redesign their markup constantly, and a CSS selector pinned to a <span class="price-now"> breaks the day they ship a new theme. The structured price in JSON-LD changes far less. Parse that first and fall back to a selector only when it's absent.

Pattern: multi-day undercut flag

A one-day price drop might be a flash sale. A five-day drop is a strategy. Add a Code node that reads the history sheet and flags products a competitor has undercut you on for N consecutive days, then escalate those separately from single-day alerts.

n8n nodes you'll use most

Getting started

1. List your competitor product URLs and your own price in a Google Sheet.
2. Build a daily Schedule trigger into an HTTP Request node per page.
3. Extract the price, parsing JSON-LD first and a CSS selector as fallback.
4. Read the last logged price and compute the percentage change.
5. Add a Filter so only changes above your threshold continue.
6. Send the change and your price to OpenAI for a recommended response.
7. Alert on the real moves and log every daily price for trend context.

For the broader monitoring picture, n8n competitor monitoring automation covers tracking launches and content beyond price, and automate ecommerce workflows with n8n puts price monitoring in the context of a full store automation stack. When you want the AI layer ready-made, the Competitor Price Intelligence template does the threshold and recommendation work for you.