List the 10-30 competitor prices and offer pages that genuinely affect your sales, point a page-change monitor such as Visualping or Distill at them, and have AI turn each alert into a row in a price log: item, old price, new price, conditions and date. Catch offers that never reach a web page with a newsletter inbox and monthly ad-library checks.
The common failure isn't missing a change; it's watching too much. A monitor on a busy product page can send dozens of alerts a week about stock levels and banner images, and within a month nobody opens them. Start narrow, read the log weekly, and use it to inform decisions rather than trigger automatic price matching. AI summarises what changed; you decide what it means.
Choose the prices that actually move your sales
My illustration here is a bicycle repair shop with two mechanics that sells some bikes and parts but earns most of its money from servicing. Its owner drew up a watch list by asking one question of every item: if a rival changed this price, would a customer notice and would it change where they go?
| What to watch | Competitor | Where it's published | Why it matters |
|---|---|---|---|
| Basic service price | Sports retailer's service desk | Service page | The number customers quote back |
| Full service price | Sports retailer, mobile mechanic | Service pages | Biggest single job by revenue |
| Puncture repair | All three rivals | Price lists | Entry job that brings new customers |
| E-bike service add-on | Mobile mechanic | Price list PDF | Fastest-growing work |
| Service plan or membership | Sports retailer | Offers page | Locks customers in for a year |
| Two popular commuter bikes | Two nearby bike shops | Product pages | Where bike sales are won or lost |
That's 14 pages. It left out hundreds of parts prices, because nobody chooses a repair shop on the price of an inner tube, and those pages change so often they'd bury the useful alerts.
An illustrative independent bookshop learnt this the hard way. It set monitors on 20 bestseller pages at a large online retailer and received more than 300 alerts in the first week, almost all tiny price movements it could never match anyway. The fix was to stop watching the retailer's prices entirely and watch two local rivals' event and book-club offers instead, which were the things that actually pulled customers away.
Where rivals' offers show up, and how to catch each one
Prices on web pages are the easy part. Offers travel through four channels, and each needs a different net:
- Web pages (price lists, offers pages, product pages): page-change monitors, covered below.
- Email and text newsletters: sign up openly to each rival's list with a separate inbox used only for this, so offers don't get lost among your own mail. A mail filter can label them automatically.
- Ads and social posts: check the Meta Ad Library (current Facebook and Instagram ads, no login needed) and the Google Ads Transparency Center (Search, YouTube and Maps ads) once a month. Ads often carry offers that aren't on the website.
- In person: window posters, counter cards, what staff say. A monthly walk past, or asking new customers what they were offered elsewhere, covers it.
An illustrative florist found the newsletter inbox the most valuable of the four. Over one winter it showed that both main rivals opened Valentine's pre-orders in the second week of January with an early-bird discount, which the florist had never noticed because the offer only went to subscribers. The next year she opened her own pre-orders a week earlier.
Set up page monitors that don't flood your inbox
A page-change monitor visits a web page on a schedule and tells you when something changes. Two widely used ones for small businesses are Visualping and Distill. Both let you watch just part of a page rather than all of it, and Visualping adds AI summaries of what changed, even on its free plan, so you get "full service price changed from $129 to $99" instead of a screenshot to squint at. Setting a monitor up well takes about five minutes a page:
- Paste the page address and let the page load in the tool.
- Select only the price area. Both tools let you draw a box round, or click, the element you care about. Watching the whole page is the main cause of noise.
- Choose a sensible frequency. Daily is plenty for service prices. Hourly checks on a price that changes twice a year just burn through your allowance.
- Narrow the alert further if the tool offers a condition or keyword option, so that only a change to the price text triggers it, not a new photo.
- Send alerts to one place, such as a dedicated folder or label, not your main inbox.
- Name each monitor clearly: "Retailer - full service price" beats "monitor 7".
Test each monitor by checking its first two or three alerts against the live page. If an alert fires with no visible change, tighten the selection; if a real change slips through, widen it slightly.
A quick sum: which plan the watch list needs
The plans are priced on how many pages you watch and how often each is checked. From the vendors' pricing pages as of September 2026:
| Plan | Price | Pages or monitors | Checks |
|---|---|---|---|
| Visualping Free | $0 | 5 pages | 150 checks a month, at most hourly |
| Visualping Personal 1K | $14 a month, or $10 a month billed annually | 10 pages | 1,000 checks a month |
| Visualping Personal 5K | $35 a month, or $25 a month billed annually | 20 pages | 5,000 checks a month |
| Distill Free | $0 | 25 monitors, 5 of them in the cloud | 1,000 cloud checks a month, cloud checks at most every 6 hours |
| Distill Starter | $15 a month | 50 monitors | Cloud checks as often as every 10 minutes |
Now the sum. Five pages checked once a day for 30 days is 150 checks, which fits Visualping's free plan exactly. The bike shop's 14 pages checked daily need about 420 checks a month and more pages than the free plans allow in the cloud. The options were Distill Starter at $15 a month, Visualping Personal 5K at $25 a month billed annually, or Distill's free plan with 5 cloud monitors and 9 local ones. Local monitors run in your browser extension, so they only check while your computer is on and the browser is open. The owner went with Distill Starter: $180 a year to watch the prices that decide most of the shop's servicing income.
Turn raw alerts into a price log with AI
Alerts tell you something changed. A log tells you what, when and by how much, and after three months it shows patterns an alert never will. Keep it in a spreadsheet with these columns:
Date seen · Competitor · Item · Old price · New price · Change % · Offer type · Conditions · Ends · Source link · Our action
You can fill it by hand in ten minutes a week, or have AI extract the fields. The prompt below works in any chat assistant; paste one or several alert texts under it.
Below are change alerts from a website monitor about competitor prices.
For each alert, return one row with these fields, separated by " | ":
date seen | competitor | item | old price | new price | change % |
offer type (permanent change / temporary offer / new item / removed) |
conditions (e.g. "app only", "first visit", "from") | end date if stated | source link
Rules: copy prices exactly as written, including words like "from".
If a field isn't in the alert, write "not stated". Don't calculate a
change % if either price is missing.Sample output for two alerts (illustrative):
28 Sep | Sports retailer | Full service | $129 | from $99 | -23% | temporary offer | "from", autumn service event | 31 Oct | [link]
28 Sep | Mobile mechanic | E-bike add-on | $25 | $30 | +20% | permanent change | not stated | not stated | [link]
Look closely at the first row. "From $99" isn't the same as $99, and the -23% figure is misleading because the full service at that price probably excludes things the $129 version included. The prompt told the assistant to keep the word "from", which is how you spot it; on the page itself, the $99 version turned out to exclude a gear-cable replacement. The owner changed the change % cell to "not comparable" and added a note. This is the most common error in AI-built price logs: treating a headline price as a like-for-like price.
If you'd rather automate the whole thing, a three-step Zap does it: a new email arrives in the alerts label, an AI step extracts the fields, and a row is added to the sheet. Zapier doesn't count the trigger, but each action step that succeeds counts as a task, and an AI by Zapier step uses 1, 3 or 5 tasks depending on the model tier. At 30 alerts a month that's somewhere between 60 and 180 tasks, which is beyond the Free plan's 100 tasks and two-step limit, so it means the Professional plan at $19.99 a month billed annually. For most small businesses, a weekly copy-and-paste is better value; when Zapier gets too expensive explains where the break-even sits.
The Friday digest: five minutes instead of fifty alerts
Once a week, paste the new log rows and any competitor newsletters from the dedicated inbox into one conversation and ask for a digest:
Here are this week's new rows from my competitor price log and the
competitor newsletters I received. Write a digest of no more than
150 words for the owner of a [type of business]:
1. Changes that affect my top 5 services: [list them]
2. Temporary offers, with end dates
3. Anything that looks like a pattern across weeks (compare with last
week's digest, pasted at the bottom)
Don't recommend price changes. Flag anything you're unsure about.A sample digest for the bike shop (illustrative):
Retailer launched an autumn service event: full service "from $99" until 31 Oct, excludes cables. Mobile mechanic raised the e-bike add-on from $25 to $30, second rise this year. Nearby bike shop emailed a 10% off commuter bikes code, valid this weekend only; they've sent a weekend code three weeks running. Unsure: the retailer's service plan page changed wording but not price.
Two alternatives if copying and pasting doesn't suit you. A scheduled task in ChatGPT can run the same kind of prompt weekly, though if it searches the web itself rather than reading your log, it may find out-of-date pages; treat that version as a rough prompt to check, not a source. And if you already get a weekly summary of your own figures, adding the digest to it keeps everything in one read; a weekly business summary emailed by AI shows how to set that up.
Deciding what to do when a rival changes price
The log is there to inform decisions, and the worst use of it is automatic matching. Four questions decide most cases:
- Is it temporary? An offer with an end date is a promotion, not a new market price.
- Is it like for like? Check what's included before comparing.
- Does it touch your customers? A discount for first-time app users doesn't affect regulars.
- Do you have spare capacity? If you're booked two weeks ahead, cutting prices loses money on work you'd have got anyway.
| Situation | Usual response |
|---|---|
| Temporary offer, you're busy | Hold. Note it, check the end date, do nothing. |
| Temporary offer, you're quiet | Run your own offer on a different item or add value (free check, faster turnaround) rather than copying the discount. |
| Permanent cut on a like-for-like item | Review your own price and what you include; explain the difference on your price list. |
| Rival raises prices | Consider whether you're underpriced. A rise across two rivals is a signal worth taking seriously. |
| New item or bundle | Ask two or three customers if they'd want it before copying. |
Here's the bike shop's decision on the autumn event. Its own full service is $119. Matching at $99 on the 40 full services it expects before the end of October would give away $800, and the shop was already booked eight days ahead. It held its price, added "includes new gear and brake cables" in bold to its price list, and put a line on the booking page about same-week turnaround. October's full-service bookings came in two below the previous October, well within the normal range.
If your own prices need to move, planning and announcing a price increase with AI covers the wording and timing. And for why a log of rivals' prices can't tell you what to charge on its own, see the limits of AI pricing research.
Services that don't publish prices
Many service businesses never put prices online, or show them only after you pick options in a booking system. An illustrative dog groomer found that her main rival's prices appeared only after choosing breed, coat condition and add-ons inside a booking app, so no page monitor could see them. Her workable routine:
- Once a quarter, step through the rival's public booking page by hand for her three most common breeds, stopping before any details are entered, and record the prices shown.
- Ask new customers who've used the rival what they paid, and record the answer with the date.
- Keep the answers in the same log with "customer-reported" in the source column, so she knows how much to trust each figure.
It's slower and rougher than automated tracking, but three data points a quarter from reliable sources beat an automated feed of the wrong numbers. Whether AI can help a small shop set prices and promotions covers what to do with the figures once you have them.
Lines not to cross when watching competitors
- Only watch public pages. Don't log in with fake accounts, get round paywalls or copy anything marked confidential.
- Check gently. Daily checks of a handful of pages are nothing like the aggressive scraping many websites' terms forbid. If a site's terms prohibit automated access, check it by hand.
- Never share or agree prices with a rival. Watching public prices and setting your own independently is normal; talking to a competitor about what you'll both charge is price-fixing under competition law almost everywhere. If you're in a trade group where prices come up, ask a solicitor where the line sits.
- Sign up to newsletters openly. Use a real name, not an invented customer persona.
Checking the tracking is worth keeping
After eight weeks, go through the log and count the decisions it informed. If the answer is none, you're watching the wrong things or too many of them. The bike shop's review found that 3 of its 14 monitors had never changed and 2 changed constantly without mattering; it swapped them for a rival's new e-bike service page and dropped to Distill's free cloud allowance for two of the quiet ones. An illustrative café that tracked the loyalty offers of three coffee shops found one useful thing in two months, a rival's switch from a stamp card to an app, and decided that quarterly checks were enough.
A good watch list is one you read every week in five minutes and act on every month or two. If you want the fuller picture of each rival rather than just their numbers, run a one-afternoon competitor review once a quarter and let the price log fill the weeks in between.
Further reads
- AI Supplier Management: Track Prices, Lead Times, and Risk — Apply the same watch-list habit to your suppliers' prices.
- How Meta and Google Use AI to Run Your Ads: A Plain-English Guide — Understand the ads you'll see rivals running in the ad libraries.
- Can a Small Online Shop Use AI to Compete With Bigger Brands? — Ways to compete that don't depend on being cheapest.
- AI Social Listening: Tracking What Customers Say About You — Track what customers say about rivals, not just their prices.
- How to Catch Outdated Information in AI Answers — Stop stale prices from AI search slipping into your log.
- How to Run a SWOT Analysis of Your Business With AI — Six stages and copyable prompts for a SWOT built on your own figures and reviews, with a podiatry clinic example and a TOWS step that turns it into actions.
- AI Menu Engineering: Which Dishes to Promote, Reprice or Drop — Run the classic menu engineering matrix with AI on your own till data, with a worked ten-dish example and a price-test calculation you can copy.
- How to Write a Salon Service Menu and Price List With AI — Price each service from your chair-hour cost, then use three AI prompts to structure, describe and stress-test the menu before it goes live.
- How to Write eBay Listings With AI That Sell Faster — Sold-price research, a title formula, item specifics and flaw-honest descriptions for eBay listings drafted with AI, plus the policy traps to avoid.
- How Small Amazon Sellers Use AI for Listings and Reviews — How a small Amazon seller can use AI for listings under the 2026 title rules, analyse reviews for fixes, and grow reviews without breaking policy.
- How Landscapers Use AI to Stop Underpricing Jobs — Find where your quotes leak money, price from your true cost per crew hour, fix the markup-versus-margin gap and let AI check each quote for forgotten items.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: Visualping and Distill pricing pages; Zapier help on task usage and Delay; Meta Ad Library and Google Ads Transparency Center; OpenAI help on scheduled tasks in ChatGPT.