Turn account data into a QBR that earns the renewal.
Value delivered, adoption vs stated goals, risks named first, the roadmap, and one outcome-tied expansion ask.
Inside: The value story, the adoption read, the roadmap, and a slide-by-slide outline.
Install it in one line, or paste it in.
~/.claude/skills/ and runs automatically when it is relevant.Connect your context. Set it to your motion.
it reads the account record, renewal date, ARR, and open cases automatically.
pulls real adoption against the goals instead of your estimate.
surfaces the open risks you may have forgotten.
This was built for a B2B SaaS org running quarterly reviews on named accounts. Set these to your stack:
Point it at the goals this customer actually stated. A QBR that measures against your goals instead of theirs is a pitch, not a review.
| Set this | What it is | Default / Example |
|---|---|---|
| GOALS source | where the customer's stated goals live | onboarding notessuccess planorder form |
| ADOPTION source | where usage lives | product analyticsa usage export |
| VALUE metrics | how this customer defines a win | hours savedtickets deflectedcycle time |
| RISK source | where open issues live | support ticketsescalationsCSM notes |
| RENEWAL field | the renewal date and ARR | Account.RenewalDateAccount.ARR |
| EXPANSION target | the next product, seat, or tier | modulesseat bandsusage tiers |
Everything the skill does, in full.
Takes what you know about an account and builds the quarterly business review the buyer actually wants: proof of value delivered, adoption measured against the goals they bought for, the risks named before they name them, a roadmap that shows you are looking ahead, and an expansion ask tied to an outcome. It ends with a slide-by-slide outline you can drop into a deck.
- 1Value-delivered story
Lead with outcomes, not features. Tie every win to a goal the customer stated at purchase. If a win has no goal behind it, mark it a bonus, not the headline. If a stated goal has no win yet, say so; the honesty buys the room.
- 2Adoption vs goals
Show adoption against the goals, not adoption in the abstract. "You wanted the whole revenue team live; 6 of 9 seats are active weekly" beats any raw number. Name the gap and whose it is to close.
- 3Open risks, stated first
List the risks before the buyer does: unadopted seats, an open escalation, a quiet champion, a contract term that no longer fits. Naming a risk yourself turns it from an ambush into a plan.
- 4The roadmap
Two lanes: what the customer will do next quarter, and what your product is shipping that they should care about. Each item has an owner and a date. A roadmap with no owners is a wish list.
- 5The expansion ask
Exactly one primary ask, tied to an outcome they already value. "You saved this much time on one team; the next team is not yet on it" is an ask. "Do you want to buy more" is not. Only make it if the value story earned it.
- 6Deck outline
Output a slide-by-slide outline: title, executive summary, goals recap, value delivered, adoption, risks and mitigations, roadmap, the ask, next steps. Ready to build, not to guess.
- No win claimed without the goal it maps to. Bonus wins are labeled bonus.
- Every adoption number shows its source, or is marked an estimate.
- Risks appear even when the quarter went well. A QBR with zero risks is a QBR that stopped looking.
- One expansion ask, not five. Make it only if the value story earns it.
QBR · sample account · renewal in 74 days GOAL (stated) RESULT ADOPTION Cut onboarding time ~30% faster (est) 6/9 seats weekly Standardize outreach Live, 1 team 1 of 3 teams Report on pipeline Not started 0 seats Open risks: 1. Champion quiet 26 days. Single-threaded. 2. 3 seats never activated since launch. 3. Reporting goal untouched; renewal talks to reference it. Expansion ask: One team proved the outreach win. The other two teams are not on it yet. Deck outline: 9 slides. Value story leads, ask on slide 8, next steps close.
The adoption counts, the time-saved figures, and any percentage come from the data you paste or the tools you connect. Nothing in the example above is a real customer number; it is illustrative. When the skill cannot see a number, it marks it an estimate and tells you which input would make it real.
Where an operator takes this next.
The manual read earns the room. Here is the version that keeps earning it every quarter.
The QBR argues itself before you ever open the deck.
Point a scheduled Claude task at Salesforce renewal dates so the outline is ready before the CSM has to ask for it.
Connect Amplitude so the adoption-vs-goals table updates itself instead of a CSM estimating "about 30% faster."
Send the open-risks section to Slack the moment it's built, so the exec sponsor sees it before the call, not during it.
One skill is the on-ramp.
A single skill does one job. Chained into a playbook, or run as a full build, it becomes a system. Here is where this one plugs in.