Most weekly team meetings die one of two deaths.
The first is the slow death: the meeting drifts into a status readout, everyone recites what they did last week, nobody makes a decision, and after a couple of months someone suggests "making it async" — which is a polite way of canceling it.
The second is faster: the meeting becomes a debate club. One topic swallows forty minutes, the quiet people check out, the goals never get looked at, and the meeting earns a reputation as the place where time goes to disappear.
Both deaths have the same cause. It's not the people, and it's not that meetings are bad. It's that the meeting has no structure doing any work. A weekly goals review that sticks — one that people would actually protest losing — has a specific job: look at the numbers, call the status of every goal, and actually solve the biggest problems, in a disciplined 90 minutes, every single week.
Here's how to run one. This structure is a synthesis of what the best-run companies converge on — whether they arrived there through OKRs, EOS-style operating systems, or trial and error — with the dogma stripped out.
The ground rules (before the agenda)
Three rules make everything downstream work. Skip them and no agenda will save you.
1.Every goal, metric, and task has exactly one owner. Not a team, not a pair — one name. Shared ownership is how things fall between chairs while everyone assumes someone else has it. If you can't name the one person who owns a goal, you've found your first issue to solve.
2.Status gets called before the meeting, not during it. Each owner spends five minutes before the meeting updating their numbers and calling their goals: on track, at risk, or off track. The meeting then starts with data instead of typing. This one habit converts the meeting from "collecting information" to "acting on information" — and it's what lets a 90-minute meeting spend the bulk of its time solving problems instead of reciting updates.
3.The meeting reviews; it doesn't do the work. You'll be tempted to solve everything in the room. Don't. The meeting's job is to decide what matters and who owns it. The work happens during the week.
The agenda: six stages, 90 minutes
Run the stages in order, every week, same order. The repetition is the point — predictability is what lets the meeting get fast.
1. Wins & headlines 5 min
Open with what went right: a customer win, a shipped feature, a personal headline. Two reasons this isn't fluff. First, it front-loads energy before you spend the better part of an hour on problems. Second, wins are information — a closed deal or a happy customer often carries signal the rest of the meeting should know about.
Keep it fast. One or two sentences per person, no discussion. Anything that sparks a conversation goes on the issues list.
2. Scorecard 10 min
Review your handful of weekly numbers — each with one owner and a target. And here's the part most teams get wrong: the owner reads out their own number. Not the facilitator running down the list, not a dashboard silently shared on screen — the person who owns the metric says the number, the target, and whether it's on or off, out loud. It takes ten seconds per number, and it's the same accountability ritual as the goals stage: your number, your voice. A facilitator reading someone else's miss lets the owner hide inside the spreadsheet; an owner reading their own miss owns it in front of the room.
The only question per number: on target or off target? Green numbers get a nod and you move on. That's not neglect; it's the reward for being green.
Off-target numbers get exactly one decision: does this need discussion? If yes, it goes to the issues list for stage five. It does not get discussed now. This is the discipline that keeps the scorecard review at ten minutes instead of thirty.
No weekly scorecard yet? Pick 5–8 numbers that would tell you the business's health if you were stranded on a desert island — revenue signals, pipeline, customer health, delivery pace. Imperfect numbers reviewed weekly beat perfect numbers reviewed never. (Or skip the blank page: our scorecard starter pack has department metric libraries ready to pick from.)
3. Goals review 15 min
Now the quarter's goals. Each owner — and only the owner — calls their status: on track, at risk, or off track. The facilitator doesn't call it for them; ownership means you say your own number out loud in front of your peers. That tiny ritual carries most of the accountability in this entire system.
Two rules keep this stage honest:
- A status called this week is the only status that counts. "It was on track last time I checked" is not a status; it's a memory. If an owner hasn't updated a goal since the last review, treat it as unknown — and stale-but-green is the most dangerous state a goal can be in, because it looks fine right up until it isn't.
- At-risk is a gift, not a confession. The whole point of a weekly cadence is catching a slipping goal in week 4 instead of week 12. If your team gets punished for calling at-risk, they'll stop calling it, and your review will be a theater of green until the quarter-end surprise. Thank people for early flags. Then put the flag on the issues list.
No solving here either. Call the statuses, capture the risks, move. (And if your statuses are hard to call because the goals themselves are mushy, that's a writing problem, not a meeting problem — see how to write key results people can't game.)
4. To-dos 5 min
Review the action items from last week's meeting. Each one is binary: done or not done. No partial credit, no story time.
Track your done rate. A healthy team completes roughly 80–90% of weekly to-dos. If yours is chronically below that, you don't have a laziness problem — you have a sizing problem (to-dos too big for one week) or a commitment problem (people agreeing in the room to things they don't believe in). Either way, that's an issue. List it.
A to-do that rolls over more than twice is telling you something. Ask what.
5. Issues 45 min · the heart of the meeting
Everything the first four stages flagged — plus anything anyone raised during the week — is now sitting on one list. This is where the meeting earns its keep, and it's why the meeting is 90 minutes and not 45: half of it is reserved for genuinely solving problems, not just naming them. Teams that shortchange this stage end up with a tidy status meeting and the same issues every quarter.
First, prioritize. Don't work the list top to bottom. Pick the three most important issues. Forty-five minutes on three issues means real depth — fifteen minutes each to get past symptoms and into root causes. Most weeks, if you genuinely solve the top three, the list takes care of itself; a team that discusses ten issues shallowly is worse off than a team that kills three. If you solve the top three with time left, pull the next one.
Then, for each issue, run three moves:
- Name it. State the real issue in one sentence. Half the time the stated issue ("marketing and sales aren't aligned") isn't the real one ("we haven't defined what a qualified lead is"). Spend a minute getting to the root; it saves ten minutes of solving the wrong thing.
- Discuss it. Everyone says their piece once. Once is the magic word — it gets the quiet people heard and stops the loud people from lapping the room. When points start repeating, discussion is over.
- Solve it. A solve is a decision plus an owned to-do with a date. "Let's keep an eye on it" is not a solve; it's a deferral wearing a solve's clothes. If the issue is genuinely too big for the room, the solve is: who owns bringing back a recommendation, and by when.
Issues you don't reach stay on the list for next week — prioritized again from scratch. Some issues resolve themselves; the list has a way of composting.
6. Wrap-up 10 min
Three closing moves, a few minutes each:
- Recap the to-dos. Read back every new commitment: owner, task, date. This is where fuzzy agreements either firm up or fall apart — better now than Thursday.
- Decide what cascades. What from this meeting does the rest of the company need to hear? Assign someone to send it. Decisions that stay in the room didn't happen.
- Rate the meeting. Everyone scores it 1–10, privately if you can. A meeting that averages 8+ is doing its job. A meeting trending toward 6 is drifting toward one of the two deaths — and the ratings will tell you months before the attendance does. Low score? You know where it goes: the issues list.
Why this sticks when other formats don't
It's the same every week. Novelty is the enemy of cadence. Because the structure never changes, the meeting gets faster every month as the team internalizes the rhythm — and a new hire can be a full participant on day eight.
Every stage feeds the next. The scorecard and goals generate the issues; the issues generate the to-dos; the to-dos get checked next week. It's a closed loop. Formats that die are almost always open loops — problems get discussed but never land as owned commitments, or commitments get made but never reviewed.
Nothing depends on heroics. No stage requires a brilliant facilitator or a motivated Monday. The structure does the heavy lifting, which is precisely what lets it survive the weeks when everyone's tired.
Common failure modes (and the fix)
- The meeting keeps running long. You're solving during stages 1–4. Ruthlessly redirect everything to the issues list — that's what it's for.
- Statuses are always green, then the quarter misses. Your team has learned that at-risk gets punished. Fix the reaction, not the reporting. And check freshness: green called six weeks ago is not green.
- Same issues every week, never solved. Your "solves" aren't producing owned to-dos with dates. Audit last month's solved issues — if they didn't each end in a name and a deadline, they weren't solved.
- People multitask through it. The meeting isn't relevant to them — usually because it's a status readout (kill stage-time reporting; the pre-meeting update handles it) or because the issues discussed never touch their work (check whether the right people are in the room).
- The remote version feels dead. Remote reviews need more structure, not less: a visible shared agenda, one screen everyone follows, and explicit turn-taking in the discuss round. Ambient meetings die on Zoom; structured ones don't.
Copy-paste template
WEEKLY GOALS REVIEW — [Team] — [Date] Cadence: same day, same time, every week. 90 min. Before the meeting: every owner updates their numbers and calls their goal status (on / at risk / off). 1. WINS & HEADLINES (5 min) One or two sentences each. No discussion. 2. SCORECARD (10 min) OWNER reads out their own number: on target / off target. Off target → issues list. No solving here. 3. GOALS REVIEW (15 min) Owner calls each goal: ON TRACK / AT RISK / OFF TRACK. Not updated this week = unknown. At-risk → issues list. 4. TO-DOS (5 min) Last week's commitments: done / not done. Target 80–90% done. Chronic rollovers → issues list. 5. ISSUES (45 min) Prioritize top 3. For each: - NAME the real issue (one sentence) - DISCUSS — everyone speaks once - SOLVE — decision + owned to-do with a date 6. WRAP-UP (10 min) - Recap new to-dos (owner, task, date) - Cascade: what does the company need to know? Who sends it? - Rate the meeting 1–10
Prefer this as a working spreadsheet? The free Weekly Review agenda pack is this exact agenda with timed segments that cascade from your start time, a running issues list, and a meeting-rating log.
Run this for four consecutive weeks before you judge it. Week one will feel stiff. Week two, faster. By week four the team is finishing each other's stage transitions — and that's when you'll notice the real change isn't the meeting at all. It's that the week between meetings got calmer, because everyone already knows what matters and who owns it.
That's the review sticking.