Battlecards have the worst decay-to-stakes ratio of any sales asset: they inform live deals - the moments where wrong information costs revenue directly - while rotting on the same weekly clock as everything else competitive. The rival's pricing moved, their pitch changed, they shipped the feature the card says they lack, and the rep quoting the card finds out from the prospect. This article is the system that fixes it: cards structured for updating, fed automatically by the competitive intelligence program's diffs, with freshness the field can verify at a glance.
The battlecard decay problem
The traditional battlecard lifecycle: built in a burst before a launch or QBR, circulated as a document, decayed silently, distrusted within a quarter, abandoned by the field in favor of tribal Slack threads. The failure is structural, not effort-shaped - a document updated by occasional heroics cannot track surfaces that move weekly. The inputs that should update it already exist in a working intelligence program: the Meta Ad Library diff catches pitch changes, the hiring side of competitor tracking catches strategic shifts, site-change monitoring catches pricing moves. The system below is just plumbing those feeds into the card.
A card structure built for updating
| Section | Contents | Fed by |
|---|---|---|
| Who they are | One-paragraph identity + momentum read | Teardown refresh |
| How they pitch | Current claims and angles, quoted | Ad-library diff |
| Pricing posture | Model, published points, discount behavior | Site-change watch |
| Where we win | Evidenced weaknesses → talk tracks | Teardown weaknesses section |
| Where they win | Honest strengths and the counter-position | Teardown + field feedback |
| Landmines | Objections they seed and the responses | Field feedback loop |
The structural decision that enables everything else: sections are collections of claims - discrete, sourced, dated - rather than narrative prose. A claim can be patched, retired or confirmed by a feed; a paragraph has to be rewritten by a person. "Where they win" earns emphasis: cards that pretend the rival has no strengths get discarded by the first rep who loses to one, and honest strengths with counter-positioning is what the field actually needs.
The feed: diffs into cards
With claims as the unit, the watching program's outputs map to card operations. A new dominant angle in the library diff patches "how they pitch" with the creative cited. A pricing-page change patches the pricing section same-day - the update that most directly saves deals. A momentum shift from the hiring watch updates the identity read. Each patch is a mission-shaped operation: evidence in, claim updated, source attached, timestamp refreshed - and the ambiguous cases (does this creative shift really change the pitch summary?) queue for the card owner per the usual gate pattern rather than auto-committing.
Evidence in the card itself
Making freshness visible
Trust requires visible freshness: each section stamped with its last-verified date, so a rep sees in one glance that pricing reflects this month while the landmines section is due a pass. The stamps also drive the maintenance loop - sections aging past their threshold (pricing weekly, identity quarterly, the rest in between) queue for re-verification automatically, which converts "keep the battlecards updated" from a standing guilt into a small scheduled workload. Cards with visible freshness get used; documents with a single "updated March" footer get screenshots in Slack asking "is this still true?", which is the system failing.
Closing the loop with the field
One input no library exposes: what the rival's reps actually say in deals. The field hears it weekly - the seeded objections, the pricing flexibility, the demo emphasis - and that intelligence dies in call notes unless the system collects it. The lightweight loop: a per-competitor intake (a Slack channel or a form field on closed-lost), mined on a schedule for landmine and strength updates, with the card crediting the source deal. Field-sourced claims complete the card's evidence spectrum - public surfaces for what the rival broadcasts, field reports for what they whisper - and reps who see their reports land in the card keep reporting, which is the loop sustaining itself. The runnable version lives on the Competitive Intel shelf.
Frequently asked questions
Why do battlecards go stale so fast?
They summarize surfaces that move weekly - pricing, pitch, creative angles - in a document updated episodically. The fix is structural: claims-based cards fed by the standing competitive diffs, not better intentions.
What structure should a battlecard use?
Six sections of discrete, sourced, dated claims: identity and momentum, current pitch, pricing posture, where we win, where they win honestly, and landmines - each section wired to the feed that updates it.
Should battlecards admit where competitors win?
Yes - cards that pretend otherwise get abandoned by the first rep who loses to an unlisted strength. Honest strengths with counter-positioning is what the field uses and trusts.
How does evidence appear in a card?
Every claim carries its artifact and date: the ad, the posting, the page capture, the sourced field report. Sourced claims commit reps in live calls and self-police - vanished evidence flags its claim for review.
What keeps the system running without a dedicated analyst?
The feeds are standing missions - library diffs, site watches, hiring sweeps - patching claims automatically with ambiguous cases queued for the card owner, plus a lightweight field-intake loop mined on schedule.
Sources
Every playbook on this blog ships as a runnable mission.
Open a workspace and the playbook library is waiting - describe the outcome and the agents carry it end to end, on your plan's monthly credits.