AI Quarterly Review Guide 2026: The QBR That Changes the Next Quarter
Most quarterly reviews are theater: a slide deck of wins, a hand-wave at the misses, and everyone back to work with the same plan as before — which means the quarter taught the organization nothing it didn't already feel. A working quarterly review answers three questions with evidence: what did we actually accomplish, what did we learn, and what should change about the plan — and it ends with the next quarter's plan genuinely revised, not reaffirmed. Here's the operating procedure.
The Evidence Pack: Opinions Follow Data Into the Room
The review's quality is set before the meeting, in the evidence pack assembled a week ahead: 1) the goal scorecard — every quarterly goal and key result with its actual number against target — green, yellow, red, no adjectives; 2) the counter-evidence — the metrics the quarter's narrative would rather ignore (churn that rose, the feature nobody adopted, the channel that underperformed) — a review without counter-evidence is a highlight reel; 3) the decision log — significant decisions made during the quarter with their recorded reasoning and the prediction-vs-actual check — the review is where those predictions get graded; 4) the external layer — market and competitor movements that changed the ground under the plan. The pack's discipline mirrors the rapid-snapshot rule of any campaign debrief — data assembled while it's still objective, before the quarter's memories set — and it draws directly on the journal and metric infrastructure in our decision-making guide and the goal-tracking loop in our annual planning guide. Pre-reads go out 48 hours before; the meeting is for decisions, not first readings.
The Review Agenda: 90 Minutes, Three Segments
The working agenda for a team-level QBR: 1) Results (30 min) — walk the scorecard, red items first (the order matters: starting with wins sets a celebratory frame that disarms the whole meeting); each red gets its one-sentence attribution — target was wrong, execution slipped, or environment shifted — the three-cause taxonomy keeps the conversation diagnostic instead of defensive; 2) Learnings (30 min) — the quarter's three most valuable lessons, each stated as a reusable claim ("paid social underperforms for our enterprise segment" — a claim next quarter's plan can act on) with the evidence that supports it; the segment's rule: learnings come from every function in the room, not just the loudest; 3) Re-planning (30 min) — the next quarter's goals drafted live against what the first two segments revealed: keep, adjust, or kill for each current goal, plus the capacity check (what gets dropped to make room for anything added). The facilitation that keeps the meeting honest — timeboxes, red-first ordering, decisions before dispersal — follows the hosting discipline in our meeting craft guides's agenda method, scaled to ninety minutes.
Re-Forecasting: The Part That Makes Reviews Matter
The review's real output is the revised plan, and the re-forecast discipline has three rules: 1) goals change formally — a dead goal is retired in writing with its reason (the market moved, the bet failed, the priority changed) — quietly abandoned goals corrupt the planning culture's credit system; 2) targets are re-based on evidence — a goal that was 40% achieved with a broken assumption gets re-based from the actual trajectory, not re-pled harder (re-pled-but-unchanged targets are how teams learn that plans are fiction); 3) the capacity ledger balances — every addition to the next quarter names its displacement — the review that adds priorities without subtracting any has just scheduled the next quarter's overcommitment, and everyone in the room knows it. The re-planned quarter then gets written in the same one-page form as the original plan — goals, targets, owners, capacity — so the plan's versions can be compared at the next review. The goal-setting standards the re-based targets must still pass — measurable, challenging, influenceable — carry over unchanged.
Follow-Through: The Week After the Review
The review's value survives only if its outputs land: 1) the review memo within 48 hours — scorecard, the three learnings, the re-planned quarter, the retired goals — circulated to everyone who feeds the next quarter's work (the one-page format follows our one-pager guide); 2) the goal updates enter the tracking system — new targets, new owners, new review dates — the tracking cadence resumes from week one of the new quarter, not week four; 3) the learnings get assigned — each reusable claim either changes a standing practice or joins the test backlog — a learning nobody acts on is meeting theater. And the meta-habit: the review itself gets a two-minute retro at its close (what should this meeting do differently next quarter?) — the review process is itself a quarterly product. AI's supporting role:
"I'm preparing our quarterly review. Inputs: goal scorecard [paste], decision log [paste], notable metrics [paste]. Draft: 1) the review agenda with timeboxes; 2) the red-first scorecard narrative (one-line attribution per red item, marked fact vs hypothesis); 3) three candidate learnings phrased as reusable claims with supporting evidence; 4) the re-planning worksheet (keep/adjust/kill per goal, with a capacity ledger column). Flag where my inputs lack evidence."
Common Questions
How is a QBR different from a monthly review?
Depth and stakes: monthly reviews are steering checks (are we on track, what's blocking — an hour, operational), while quarterly reviews are navigation (is the plan itself right — half a day, strategic). The quarterly asks questions the monthly can't: whether goals are still the right goals, whether the quarter's evidence changes the year-level plan, whether capacity assumptions held. Teams that make quarterly reviews just "bigger monthlies" get the operational rhythm without the navigation — and drift for a year before noticing.
Who should be in the room?
Everyone who owns a piece of the quarter's plan, plus one voice from outside it — a cross-functional partner or a customer-facing lead whose perspective checks the room's insularity. Leadership attends to hear and decide, not to present. The two attendance failures: the review as spectator sport (twenty people watching five talk — send the memo instead) and the review without deciders (findings agreed, nothing changes — the meeting was a documentary). The working size is 6-10 people with a stake in the re-plan.
How do we handle a quarter where everything went wrong?
The all-red quarter is the QBR's most valuable session, and its rules differ: the attribution pass runs first and runs honest (was it target-setting, execution, or market — name it plainly, because the fix differs), the tone is set by the leader's opening (if the leader treats it as a trial, the room learns to hide evidence — the most expensive lesson a team can learn), and the re-plan is deliberately modest (re-build credibility with a winnable quarter, not a heroic one). The all-red quarter's review, done well, is where teams build the muscle that makes future reds survivable.
Can AI run our quarterly review?
AI assembles the evidence layer well: scorecard compilation, variance summaries, first-draft attributions marked as hypotheses, the memo's structure. It cannot supply the counter-evidence your metrics dashboards forgot to track, judge which learning actually generalizes, or make the keep-adjust-kill calls — those are the room's judgment with stakes attached. AI writes the pack and the memo; the meeting's honesty is human work.
Author: UseAIWriter Team | Updated: 2026-11-10 | Originally published on UseAIWriter.
Try Our AI Writing Tool Free
Want to generate high-quality content? Try our free AI writing assistant — no registration, no limits, no credit card required.
Try AI Writer Free →