SaaS Pricing Change Radar

Track competitor pricing. Understand what changed.

Monitor public SaaS pricing pages, extract evidenced plans and published prices, keep private cross-run snapshots, and spot pricing, feature, quota and trial changes. Built for SaaS founders, product managers, product marketing teams and competitive researchers.

No LLM subscription or extra API key. No separate dashboard or login. Run on demand or reuse an Apify Task with a daily or weekly Schedule.

Quick start

1. Keep the example or add a product name and a public pricing URL.
2. Click Start. The first successful observation establishes a baseline.
3. Run again with the same URLs, monitor name and rendering settings to compare against the latest successful observation.
4. Open the JSON report, flat CSV, Markdown report or dataset in Output.

{
  "pages": [
    {"product": "Linear", "url": "https://linear.app/pricing"},
    {"product": "Buffer", "url": "https://buffer.com/pricing"}
  ],
  "monitorName": "competitors",
  "renderMode": "browser"
}

Up to 20 URLs per run, processed sequentially. Exact duplicate normalized URLs are analyzed once. The default browser mode renders JavaScript, waits briefly for loading and excludes hidden elements. It observes the default English desktop state without clicking billing, currency, region or usage controls.

What you get

• Product name, requested and resolved source URLs, UTC observation time and collection method.
• Plan names, directly observed amounts, currency when explicit, billing period, seat/unit text and annual-billing labels.
• Free and unpublished/contact-sales plans. Unknown values stay null, unknown or not_public; no annual conversions or inferred enterprise quotes.
• Observed card feature labels with per-item inclusion status and evidence, quota/limit text, trial and discount excerpts, contact-sales links where present.
• Per-plan source excerpts and source URLs; each offer has its own evidence. A source HTML hash makes collection drift inspectable.
• Private history across separate runs. Each monitored URL retains the latest 30 successful observations.
• JSON results, a flat CSV with one row per plan/offer, Markdown and a monitoring summary.

A dollar sign alone does not establish USD. The Actor leaves currency unknown unless the page identifies the relevant dollar currency. It does not convert currencies.

Understand changes and confidence

comparisonStatus is baseline, no_change, changes_detected, incomparable or unavailable.

Published-price increases/decreases compare the same normalized plan name and offer context: currency or symbol, billing period, unit, billing model and billed cadence. Output includes the before/after offers and their evidence, previous/current observation times and confidence.

Added/removed plans, feature wording changes, quota changes, trial/discount changes and changed billing contexts are review candidates (reviewRequired: true). Layout changes, A/B tests and wording can resemble commercial changes. These candidates are not verified claims that an entitlement was removed. Whitespace and case changes are ignored. Renamed plans may appear as addition/removal candidates. Confidence values are rule-based quality indicators, not calibrated probabilities.

The time window is when a change was observed, not the vendor's effective change date. A baseline is never presented as historical change. Failed/empty extractions do not replace the previous successful snapshot. Changing the rendering or card selector makes the next comparison incomparable.

Feature names in features are observed card labels. featureDetails[].inclusion is included or excluded only when an explicit accessible text label supports it; otherwise it is unknown. Unlabelled check/cross icons are not interpreted as proof of an entitlement. The warning unverified_feature_inclusion calls this out. Review source pages for contractual feature availability.

History and privacy

History is stored in a named Apify key-value store under the executing account, using authenticated user and Actor identifiers plus monitorName. Users cannot supply another account identity or select another history store. The Actor requests limited permissions. No embedded developer token or shared external database is used.

Different monitors within your account have separate histories. Account members with access to your storage can access that account's reports. Do not put credentials in product names or page URLs. Reports are subject to your Apify storage retention and access policies. Delete the named store shown in SUMMARY.historyStoreId to reset history; deleting it removes the baseline. Only the latest 30 snapshots per URL are retained, not an unlimited archive.

A server-side lock prevents concurrent updates to the same monitor. A busy monitor run fails clearly with no product events. Its lock releases on completion or expires after interruption. For best results, schedule non-overlapping runs. Resurrected runs do not reanalyze or charge again; start a fresh run to collect new evidence.

Schedule and notifications

Save the input as an Apify Task and create a daily or weekly Schedule. Reuse its monitor name and source URLs. A schedule is created only when you enable it; installing this Actor does not start recurring costs.

Optionally set changesWebhookUrl to your public HTTPS endpoint. The Actor sends JSON only when it observes change signals, including review candidates. Baseline/no-change runs send nothing. The payload contains event: "pricing.changes_observed", run ID, monitor name and per-page changes. Treat webhook URLs containing secrets as private input. Delivery is bounded to one attempt; inspect WEBHOOK-RESULT for errors and use native Apify webhooks if you need platform-managed retry delivery.

Alternatively configure a native ACTOR.RUN.SUCCEEDED webhook and fetch SUMMARY or REPORT, then filter changeCount > 0. Run success can include failed pages: check individual status, error and warnings.

Pricing

The launch price is $0.01 per successful pricing page analysis ($10 per 1,000). Standard Store platform usage is included; there is no separate LLM or proxy fee. The live Store Pricing tab is authoritative. One custom pricing-page-analysis event corresponds to one successfully persisted, valid page analysis, including baseline and no-change observations. Multiple plans or changes on that page do not create extra events.

Error pages, pages with no verifiable pricing cards, budget skips and invalid inputs do not trigger that product event. A valid analysis may contain unknown fields or partial card coverage; inspect the public examples and warnings before scheduling at scale. There is no startup event or automatic dataset-row event.

Use Max cost per run to limit paid results. Apify's free-plan prepaid credit can be used to test the example without a separate LLM subscription; this is platform credit, not an unlimited free tier. Execution and build costs for your own Actor development are governed by your Apify account plan. Standard Store runs include platform usage in the event price. Standby mode is not supported.

Output and API

The default dataset has one JSON row per input page, including diagnostics. REPORT contains the same rows under pages; REPORT-CSV flattens plans/offers for Excel and BI; REPORT-MARKDOWN is the human-readable report. Nested arrays in CSV are JSON strings. JSON best preserves the evidence structure.

import {ApifyClient} from 'apify-client';
const client = new ApifyClient({token: process.env.APIFY_TOKEN});
const run = await client.actor('Serendy/saas-pricing-change-radar').call({
  pages: [{product: 'Linear', url: 'https://linear.app/pricing'}],
  monitorName: 'competitors', renderMode: 'browser'
});
const {items} = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

billable means the persisted analysis qualifies for a product event. Actual charges depend on the current pricing, remaining budget and Apify counters. It does not prove a customer payment.

Live output example

This shortened example was collected from Linear on October 19, 2026. It is an observed public page, not a fictional price-change scenario. The full output includes plan and quota evidence.

{
  "product": "Linear",
  "sourceUrl": "https://linear.app/pricing",
  "collectedAt": "2026-10-19T11:54:10.141Z",
  "status": "success",
  "comparisonStatus": "no_change",
  "plans": [
    {
      "name": "Basic",
      "offers": [
        {
          "amount": 10,
          "currency": null,
          "symbol": "$",
          "period": "month",
          "unit": "per user",
          "billingModel": "per_seat",
          "billed": "yearly",
          "evidence": {
            "url": "https://linear.app/pricing",
            "quote": "$10 per user/month Billed yearly Get started"
          }
        }
      ],
      "features": [
        "All Free features +",
        "5 teams",
        "Unlimited issues",
        "Unlimited file uploads",
        "Admin roles"
      ]
    }
  ],
  "changes": [],
  "warnings": [
    "ambiguous_dollar_currency",
    "default_visible_state_only",
    "card_coverage_not_guaranteed",
    "common_plan_names_only_unless_selector_supplied"
  ]
}

Scope and limitations

This is a conservative DOM-based MVP. Automatic detection accepts common plan names (for example Free, Starter, Pro, Business, Team and Enterprise) on separate cards with headings and visible prices. For unusual plan names, supply a card selector. Add-ons and comparison-table-only offers are outside automatic detection. Arbitrary pages are not guaranteed: comparison matrices, calculator-only pricing, iframe/shadow-DOM content, delayed states and unusual markup can return partial or no results. A valid extracted subset does not prove complete page coverage. Feature extraction focuses on card lists; long comparison tables are not reconstructed.

For custom card layouts, provide cardSelector on a page entry: a CSS selector matching each plan card containing an h2–h6 plan heading. Check the sample output before recurring use. HTTP-only mode may include CSS-hidden variants and is intended for advanced users who have checked the page structure. Automatic mode uses HTTP first and opens a browser only if no valid cards were found.

Source network errors, HTTP 4xx/5xx, access controls, oversized responses, missing cards and browser loading failures are reported. Captchas, login walls and access restrictions are not bypassed. No residential proxies, LLM or third-party enrichment services are used. Public web pages can change, vary by region, or restrict automated access. Enter only URLs you are permitted to access.

Pricing evidence is for research and review, not a vendor quote, purchasing guarantee or proof of all commercial terms. Please review source links before making a decision.

Support

Open an issue in the Store Issues tab with the source URL, run ID, expected plan and a redacted input. Never share API tokens or private webhook secrets.