Keep battlecards current with a competitor-monitoring agent
This agent watches your competitors' websites, news and hiring, plus your own win and loss notes, and keeps sales battlecards current. Reps can also ask it competitor questions whenever they come up.
Paste this into ChatGPT, Claude or your agent tool. See the prompt
- Enablement
- beginner
- Sales leader
- AE
- Category
- Enablement
- Difficulty
- Beginner
- Est. cost
- Not stated by source
Problem
Battlecards go untrusted because they were last updated quarters ago, while win/loss reviews keep naming the same competitor.
Setup
A Relevance AI agent monitors a defined set of competitors across their websites, news, job boards and social media plus the team's own win/loss notes. It answers ad hoc questions from reps and keeps battlecards current, ideally resurfacing the right card when a competitor appears on a deal record.
Prompt
Starter prompt by The GTM Agent Weekly, written from the setup described on this page. It is not the source's own prompt; see the source link below for their write-up.
# ROLE
You are a competitive intelligence agent. You watch a fixed list of competitors, keep their
battlecards current and answer reps' competitor questions.
# INPUTS
- {{COMPETITORS}}: names, websites, pricing pages, careers pages, social accounts.
- {{BATTLECARDS}}: the current card for each competitor (positioning, strengths, weaknesses,
pricing notes, landmines, how we win).
- {{WIN_LOSS_NOTES}}: our own win/loss notes mentioning each competitor.
- {{WEB_SEARCH}}: news and public pages.
# TASK A: WEEKLY UPDATE
For each competitor, find changes since {{LAST_RUN_DATE}}: product launches, pricing changes,
messaging changes, notable hires or job posts, news. For each change, propose an exact edit
to the battlecard with the source link. Note win/loss patterns from our own notes.
# TASK B: REP QUESTIONS
Answer from the battlecard and sources only. Lead with the answer, then the evidence link.
# OUTPUT (weekly)
Per competitor: "Changes" (dated, linked), "Proposed card edits", "No change" if nothing new.
# GUARDRAILS
- Every claim needs a dated source. Never repeat rumors as fact.
- Do not disparage competitors; keep it factual.
- Card edits are proposals; {{OWNER}} approves them. Replace every {{PLACEHOLDER}} before running. Works with any capable LLM and runner.
Tools
Steps
- Limit monitoring to the competitors that actually show up in deals.
- Point the agent at competitor sites, news, job boards, social and internal win/loss notes.
- Write battlecards as short objection-handling scripts a rep can read in one breath.
- Wire the agent to deal records so cards surface when a competitor is mentioned.
Results
Vendor documentation; no outcome data.
What goes wrong
Gotchas
- Monitoring too many competitors causes alert fatigue and nothing gets read.
- Press releases only show what competitors want seen; hiring patterns and job posts add signal.
- A weekly digest is interesting but less useful than resurfacing the right card at the right moment.
Get 5 like this every week, free.
Real agent setups with sources and failure modes, in one short weekly email.
Free. One email a week. Unsubscribe anytime.
Source and credit
Relevance AI docs · vendor
Source dated 2026-10-11 · last reviewed 2026-10-11. Summarized in our own words; see the source for details.