Rival Newsletter launched this week with an ambitious pitch: free competitive intelligence delivered via email every Monday, where each subscriber gets a completely different brief tailored to their specific company and competitors. The tool runs entirely on AI agents with no humans in the loop, processing six data sources to generate personalized digests that include actionable copy you can ship same day.

How Three Agents Work Together

The system relies on three distinct agent roles working in sequence. First, the context agent reads your own website—homepage, pricing page, features, and blog posts—and drafts five markdown documents covering your positioning, ideal customer profile, pricing model, brand voice, and differentiators. This knowledge base refreshes weekly and lives editable in your dashboard. Second, the signal agent monitors your three tracked rivals across Reddit discussions (scored by upvotes), Hacker News threads weighted by engagement, web news via Bing and Google News including press releases and executive moves, their website diffs for pricing or feature changes, blog posts with topic tags and one-line summaries, and changelog entries parsed into per-feature categories. The final piece is where it gets interesting. The editor agent combines your context documents with that week's signals to produce a tailored brief—not generic industry news, but specific patterns across your rivals contrasted against YOUR positioning. The example digest shows copy like: 'Webflow simplified plans. Spronta simplified pricing. Flat rates. No math required.' Each brief ends with concrete recommendations and ready-to-ship copy drafts you can use immediately.

Why Personalization Matters

The founders explicitly call out what they see as the fundamental flaw with existing newsletters: 'Most newsletters are written once and broadcast to thousands. They optimise for the average reader. That's the wrong shape for marketing intelligence.' Their counter-proposal is hyperpersonalized content at scale—different brief, different rivals, different recommendations for every subscriber. On privacy, Rival Newsletter takes a surprisingly strict stance. They explicitly promise no LinkedIn scraping, no logged-in extraction, and public sources only. Your tracked competitor list stays private—they don't share what you watch with anyone. The cadence is simple: one email per week on Mondays at 9am UTC, with daily alerts opt-in. Unsubscribe works in one click with no retention dark patterns. The tool currently handles three rivals by default and reads six source types out of the box. Reddit discussions mentioning your competitors get scored by upvotes for relevance. Hacker News threads including Show HN launches receive engagement-weighted priority. Their changelogs are parsed into individual feature entries with category tags, making it trivial to spot when a rival ships something relevant to your market.

Key Takeaways

  • Three agents work in sequence: context (reads you), signal (watches rivals), editor (writes the brief)
  • Six public data sources monitored: Reddit, Hacker News, web news, competitor websites, blogs, and changelogs
  • Each digest is fully personalized to your positioning—not a broadcast newsletter
  • Output includes actionable copy drafts you can ship same day

The Bottom Line

This isn't another AI hype project—it's a legitimate workflow shift. Competitive intelligence has been stuck in expensive analyst reports and generic monitoring tools for years. If Rival Newsletter actually delivers on the personalization promise, it could make $50k/year competitive intel subscriptions look very outdated.