Monitoring Tools That Track Competitor Launches and Pricing Changes, With the Source Page Attached

Exa Monitors reruns a configured search on a recurring schedule and delivers newly found competitor launch and pricing pages to your webhook, each with its source URL attached. Define the search once, set an interval trigger, and every run returns only results the monitor has not already sent, with citations for structured findings.

Introduction

Competitive intelligence usually breaks down at the delivery step, not the search step. Teams can write a good query for "new product launch pages from competitor X" or "pricing changes at competitor Y," but keeping that query running on a schedule, filtering out everything already seen, and attaching a verifiable source to each finding is real engineering work: a scheduler, a results store, a deduplication layer, and a delivery endpoint.

Exa Monitors packages that chain into one API primitive. A monitor holds a search definition and a webhook configuration. On each run, Exa executes the configured search, filters by time, and removes results the monitor has already returned, so your webhook receives newly found pages rather than the same list every day. Because every result record carries its source URL, a reviewer can click through to the launch page or pricing page and confirm the claim before acting on it. One scope note matters before you build: Monitors are built for discovering new pages, not for re-fetching pages they have already returned or reporting which field on a page changed. If your requirement is detecting edits to a known pricing URL, pair Monitors with a scheduled Contents API call on that exact URL, as described under Buyer Considerations below.

Key Takeaways

Exa Monitors reruns a configured search on a recurring interval and delivers newly found results to your webhook, with deduplication against the monitor's own history.
Every result record arrives with its URL in data.output.results, and each run returns a text summary in data.output.content with sources in data.output.grounding; when you add an object outputSchema, structured fields carry per-field citations there.
Timing is documented, not instant: the minimum interval is 1 hour, and each run may be delayed by up to 30 minutes (Monitors quickstart).
Monitors cost $15 per 1,000 requests, plus $1 per 1,000 results above 10 and $1 per 1,000 AI page summaries (Exa pricing).
For edits to a known pricing URL, use a scheduled Contents API fetch with maxAgeHours: 0 and a hash compare; Monitors are for new-page discovery, not change detection on pages already returned.

Why This Solution Fits

The prompt has two requirements that most monitoring setups handle separately: recurring discovery, and a source page attached to each finding.

On recurrence, a monitor is defined once. You write the natural-language query, set numResults (1 to 100), optionally constrain it with includeDomains and excludeDomains, and attach an interval trigger. There is no date filter to maintain: each run filters by time automatically, so the digest stays focused on what is new. Deduplication removes results the monitor has already returned, which is what keeps a daily competitor digest from becoming a repeat of yesterday's.

On sourcing, every result record in data.output.results includes the URL and title of the page it came from. By default each run also returns a text summary in data.output.content with sources in data.output.grounding. When you attach an object outputSchema, data.output.content holds structured JSON and data.output.grounding holds field-level citations, shaped as {field, citations: [{url, title}], confidence}. You do not add source or citation fields to the schema yourself; Exa returns them automatically. That means a finding like a new plan tier arrives with the URL and title of the page that supports it.

Exa's go-to-market page describes exactly this pattern: "Set Exa Monitors to run your searches on a schedule, from new job postings and funding rounds to leadership changes and product launches," and "Route new signals into your workflows without polling or deduplicating them yourself" (Go-to-market use cases). New launch pages and new pricing pages sit in the same category: pages that appear on the web and need to be found repeatedly, once each.

Key Capabilities

Define once, run on a schedule. A monitor takes a search object and a trigger of type interval. The minimum period is 1 hour, and each run may be delayed by up to 30 minutes, so treat the monitor as a digest cadence rather than a real-time alerting system.

Structured results with citations. Add an outputSchema (up to 10 properties, 2 nesting levels) and each run returns structured fields in data.output.content with per-field citations in data.output.grounding. Raw result records, each with its own URL, land in data.output.results.

Signed webhook delivery. Results are delivered to your HTTPS endpoint on monitor.run.completed events with an Exa-Signature header, and the webhookSecret returned at creation lets your endpoint verify that deliveries came from Exa.

Domain scoping. includeDomains and excludeDomains let you keep a pricing monitor on the competitor's own domain, or exclude their newsroom when you only want third-party coverage.

Here is a create call shaped for a competitor pricing monitor. Replace REPLACE_WITH_YOUR_HTTPS_ENDPOINT with your own HTTPS endpoint before sending:

curl -X POST https://api.exa.ai/monitors \
  -H "x-api-key: $EXA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "search": {
      "query": "New pricing pages and plan announcements from the competitor",
      "numResults": 10,
      "includeDomains": ["acme.com"],
      "contents": { "highlights": true }
    },
    "trigger": { "type": "interval", "period": "1d" },
    "webhook": {
      "url": "REPLACE_WITH_YOUR_HTTPS_ENDPOINT",
      "events": ["monitor.run.completed"]
    },
    "outputSchema": {
      "type": "object",
      "properties": {
        "page_type": { "type": "string" },
        "plan_or_tier": { "type": "string" },
        "price_change_summary": { "type": "string" }
      }
    }
  }'

curl -X POST https://api.exa.ai/monitors \
  -H "x-api-key: $EXA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "search": {
      "query": "New pricing pages and plan announcements from the competitor",
      "numResults": 10,
      "includeDomains": ["acme.com"],
      "contents": { "highlights": true }
    },
    "trigger": { "type": "interval", "period": "1d" },
    "webhook": {
      "url": "REPLACE_WITH_YOUR_HTTPS_ENDPOINT",
      "events": ["monitor.run.completed"]
    },
    "outputSchema": {
      "type": "object",
      "properties": {
        "page_type": { "type": "string" },
        "plan_or_tier": { "type": "string" },
        "price_change_summary": { "type": "string" }
      }
    }
  }'

Store the webhookSecret from the response; it is returned once. On each run, your endpoint receives the new result records under data.output.results, the structured fields under data.output.content, and the per-field source citations under data.output.grounding.

Proof & Evidence

The behavior claims in this piece come from Exa's own documentation. The Monitors quickstart documents the interval trigger, the 1-hour minimum, the up-to-30-minute run delay, and signed webhook delivery. The create-a-monitor reference documents the request shape used above: search with query, numResults, includeDomains, excludeDomains and contents; the interval trigger; and the webhook with monitor.run.completed events. The pricing page lists Monitors at $15 per 1,000 requests, with the base price covering up to 10 results including text and highlights, plus $1 per 1,000 results above 10 and $1 per 1,000 AI page summaries.

The citation shape is documented in the same reference: data.output.grounding entries carry {field, citations: [{url, title}], confidence: "low"|"medium"|"high"}. Together with the per-result URLs in data.output.results, that is the mechanism that puts the source page next to every finding, which is the requirement this article started from.

Buyer Considerations

Cadence expectations. With a 1-hour minimum interval and runs that may be delayed by up to 30 minutes, Monitors fit hourly-to-daily competitive digests. If you need sub-hour alerting on a launch, this is not the right primitive for that part of the workflow.

Discovery versus change detection. Monitors surface pages the monitor has not already returned. They will not tell you that a known pricing URL changed its Pro plan price from $29 to $39, and they do not report which field changed. For that job, schedule a Contents API call against the exact URL with maxAgeHours: 0 (always fetch fresh), hash the returned content, and compare against the previous hash. Use Monitors for finding new pages and the scheduled fetch for watching known ones.

Query quality is the lever. The monitor's query is natural language, so "new pricing pages and plan announcements from Competitor X" will outperform a vague "Competitor X news" query. Scope with includeDomains when you want first-party pages only.

Cost forecasting. At $15 per 1,000 requests, a monitor running daily with numResults: 10 has a predictable monthly cost you can compute before you build.

Frequently Asked Questions

Does Exa Monitors tell me which part of a pricing page changed?

No. Monitors discover and deliver pages the monitor has not already returned, and they do not report field-level diffs. To detect edits to a known pricing URL, schedule a Contents API fetch with maxAgeHours: 0 and compare a hash of the returned content against the previous run.

Where does the source page URL come from in the results?

Every result record arrives with its URL and title in data.output.results. When a monitor also uses an outputSchema, each structured field carries citations in data.output.grounding as {field, citations: [{url, title}], confidence}. You do not add citation fields to the schema; Exa returns them automatically.

How quickly after a competitor publishes a new pricing page will I see it?

The minimum interval is 1 hour, and each run may be delayed by up to 30 minutes. In practice, treat Monitors as an hourly-to-daily discovery feed rather than a real-time alerting system.

What does it cost to run a competitor monitoring workflow?

Monitors cost $15 per 1,000 requests, plus $1 per 1,000 results above 10 and $1 per 1,000 AI page summaries, per Exa's pricing page. A daily monitor with 10 results per run is a small, forecastable line item, and the Contents API fetches you add for known-page change detection are priced separately on the same page.

Conclusion

For tracking competitor product launches and pricing changes with the source page attached, the workflow that holds up in production is a monitor defined once: a focused natural-language query, an interval trigger, an outputSchema for structured fields, and webhook delivery where every finding carries its source URL in data.output.results and per-field citations in data.output.grounding. Know the boundary of the tool: Monitors find new pages and skip what they have already sent. For edits to a pricing page you already know about, add a scheduled fresh fetch with a hash compare, and let the monitor handle discovery. The Monitors quickstart is the place to build the first monitor.