macmenes Vantage-Monitor — Recurring Competitive Watch
Competitive facts decay. This skill stands up a repeatable watch over the set, captures a baseline, and on each run reports only what changed and what to do about it — so the landscape report and battlecards never go stale.
Web-only. Brief format = template E in ../vantage/references/deliverable-templates.md.
What to watch (per competitor)
| Signal | Source | Why it matters |
|---|---|---|
| Pricing & packaging | /pricing (+ Wayback diff) |
Price moves = strategy shifts; battlecard-critical |
| Homepage hero / positioning | Homepage H1+subhead | Repositioning = vulnerability or new threat |
| New ad campaigns | Meta Ad Library, Google ATC | New offers/angles to counter or swipe |
| New content / launches | /blog, /changelog, /news, sitemap lastmod |
Product & content momentum |
| Reviews & sentiment | G2/Capterra/Trustpilot counts + recent text | Emerging strengths/complaints for battlecards |
| Funding / M&A / news | WebSearch <brand> news |
Resourcing, threat level, market moves |
| Social cadence shifts | Profile activity | Channel strategy changes |
Method
- Baseline. On first run, snapshot each watched signal with value + source URL
- capture date into a
COMPETITIVE-WATCH-BASELINE.md(or update the existing landscape report's Evidence Log).
- capture date into a
- Diff. On each subsequent run, re-fetch and compare to the last snapshot. Use the Wayback Machine to confirm when a change happened where the live page hides it.
- Report only deltas. Produce the monitoring brief (template E): what changed, on which channel, the source, and the recommended response. No-change items get a one-line "no material change" — don't pad.
- Escalate. Flag changes that invalidate a battlecard claim or the positioning map as action-required.
- Refresh artifacts. Offer to update the affected battlecard / landscape section.
Standing up the recurring run
This skill defines what to check. To actually run it on a cadence, use the harness's scheduling — don't reinvent it:
/schedule— a cloud cron agent (e.g. weekly) that re-runs this watch and reports deltas. Best for hands-off, runs-without-you monitoring./loop— an in-session recurring run when you want to watch actively for a while.
Recommend a cadence by volatility: weekly for active markets / live deals,
monthly for stable categories, ad-hoc before a launch or big pitch. Suggest
the specific command, e.g. "Want me to /schedule this competitive watch weekly on
Mondays?" — and only offer if there's a real ongoing need.
Output (each run)
# Competitive Watch — [date] (since [last date])
## Material changes
| Competitor | What changed | Channel | Source | When (Wayback) | So what |
|-----------|--------------|---------|--------|----------------|---------|
## Action required (invalidates a battlecard / map)
- …
## New entrants / exits
- …
## No material change
- [competitors with nothing notable]
## Recommended responses
1. …