What a 15-minute weekly portfolio review looks like

From the SoleOS answers series — written about our own product space; grounded in published definitions and documented behavior, never invented numbers.

A 15-minute weekly portfolio review means looking at the same five things for every product you run — MRR and its 30-day change, new signups, active users, churn, and what actually moved since last week — at a fixed time, every week, no exceptions. It works because a short review you do consistently catches problems while they're still small; an occasional deep-dive only happens after something's already gone wrong. The review exists for one reason: to decide where next week's focus block goes.

Why weekly and short beats sporadic and deep

Solo founders running several products tend to default to one of two failure modes. Either they check nothing until a payment fails or a support email spikes, or they binge-audit one product for two hours, feel productive, and don't look at the other four for a month. Both feel like attention. Neither is.

A fixed 15-minute weekly slot works because it turns "how's everything doing" into a habit instead of a decision you have to re-motivate yourself into each time. Fifteen minutes is short enough that skipping it has no excuse, and long enough to notice a trend line bending the wrong way. If the review can't fit in 15 minutes, you're tracking too much per project or reviewing too many projects to hold in your head — both are signals, not reasons to extend the timer.

Trends are also only visible against a baseline. A single deep-dive tells you where a product is; a weekly cadence tells you where it's headed, which is the thing you actually need to act on.

The five things to check per project

Same five per project, same order, every time:

1. MRR and its 30-day change. Not just this week's number — the direction over the last month. A flat $340 MRR that was $310 a month ago is a different story than one that was $410.
2. New signups. Trials started, or accounts created, depending on the product. This is your top-of-funnel pulse.
3. Active users. Whatever "active" means for that specific product — logged in, completed a core action, opened the app. Pick a definition once and don't redefine it week to week, or the trend becomes meaningless.
4. Churn. Cancellations or lapsed accounts since the last review. Small numbers matter here — on a product with 40 subscribers, losing three in a week is worth a note even if the dollar amount looks tiny.
5. What moved since last week. The one line that isn't a metric: a launch, a bug, a review, a spike in refunds, a competitor mention. This is the context that explains why the four numbers above look the way they do.

That's it. No funnel breakdowns, no cohort tables, no attribution modeling — that's what the occasional deeper session is for, not the weekly one. For a working definition of which numbers are worth this kind of recurring attention versus which ones are noise, see what actually to track across a portfolio of apps.

The one decision this review exists to make

Every weekly review should end with a single sentence: "Next week's focus block goes to [project], because [reason from the five things above]." That's the entire output. Not a to-do list, not a strategy memo — one allocation decision.

A review that doesn't produce a decision is just a status check, and status checks don't move anything forward. If you finish 15 minutes with "yeah, things look about the same everywhere," you haven't reviewed — you've skimmed. Force the sentence. If nothing stands out, the honest version is "focus stays where it already was, because nothing changed enough to justify moving it" — which is itself a decision. With several products running at once, your bottleneck was never information — it was where your one weekly focus block gets spent. The five-metric scan exists purely to feed that one choice.

Keep the format lightweight

The format matters less than the discipline. Two options that both work:

• A doc with one line per project. Project name, MRR, 30-day change, signups, active, churn, one-line note on what moved — a single screen for ten products. A plain spreadsheet is genuinely enough if you're running one or two things; the tab-switching cost only bites once you're pulling from three or four separate dashboards to fill it in.
• A dashboard that already has the numbers pulled together. The advantage isn't a better number — it's not re-logging into five separate billing and analytics dashboards to assemble the same five fields every week. That reassembly cost is what usually kills the ritual by month two, not lack of discipline.

Whichever you use, resist adding fields. The moment the review needs a "notes on notes" column or a second tab, it's stopped being a 15-minute ritual and turned into the deep-dive it was supposed to replace.

Questions to ask during the review

As you scan the five metrics per project, three questions do most of the work:

• What's growing? Not "what looks fine" — what has a genuinely upward line on MRR, signups, or active users over the last 30 days. That's a candidate for more focus, not less.
• What stalled? Flat or declining for two or more weeks running. Stalled isn't automatically bad — some products are meant to be low-maintenance cash flow — but it should be a conscious call, not something you noticed by accident three months late.
• What needs a decision? Not a task — a decision. "Should this pricing page get rewritten" is a task. "Is this product worth another focus block at all, or should it go on maintenance mode" is a decision. If a product has been stalled for a while with no plan to fix it, that's the moment to ask whether it's worth keeping active — see when it's time to kill a side project for how to make that call without sentimentality.

Avoiding analysis paralysis and vanity metrics

Two ways this ritual quietly breaks:

Vanity metrics creep in. Total downloads, cumulative signups, lifetime revenue — numbers that only go up and never tell you what happened this week. They feel good to look at and tell you nothing to act on. Every metric in the five above is a rate or a 30-day trend, never a running total. If a number in your review can't go down, it doesn't belong in the review.

Analysis paralysis creeps in when the review tries to explain everything. Fifteen minutes is enough time to notice churn ticked up, not to diagnose why. If a number looks wrong, the review's job is to flag it, not solve it on the spot — write "churn up on Project X, dig in this week" and move on. That dig-in is a separate, deeper session; the weekly review's only job is triage.

Doing this across several products without it ballooning

The failure mode with five, eight, or ten products isn't lack of discipline — it's that the per-project cost doesn't stay fixed. Logging into five separate payment dashboards and three separate analytics tools to fill in one line per project turns a 15-minute ritual into an hour, and an hour-long ritual gets skipped the first busy week. If you're managing more than a couple of products, how to actually manage multiple side projects without losing your mind covers the broader operating rhythm this review sits inside.

The practical fix is getting all the numbers in one place before the timer starts, so the 15 minutes is spent reading and deciding, not hunting and logging in. Whether that's a spreadsheet across two products or a dashboard that already pulls revenue and usage together across ten, the constraint is the same: the review has to stay cheap enough per project that adding a new product doesn't push the whole ritual past 15 minutes. SoleOS publishes this guide because a consolidated weekly view across every project — with an AI-written summary of what moved — is exactly what it's built to do; see what that looks like on a live demo of a portfolio dashboard.

Frequently asked questions

Is 15 minutes really enough time for a real review?

Yes, as long as the numbers are already gathered before you start the timer. Fifteen minutes is a ceiling on reading and deciding, not on collecting data — if the numbers aren't in one place, the review runs long or gets skipped, which is the failure mode to guard against.

What if a product doesn't have real signup or churn numbers yet — it's pre-revenue?

Track the closest top-of-funnel signal instead: waitlist joins, landing page visits, whatever action is nearest to "someone showed interest." The same five categories still apply before there's a subscriber base — churn just becomes "did anyone who signed up go quiet."

Should I do this review on the same day every week?

Yes. Week-over-week comparison depends on comparing like periods, not Friday to Monday to Thursday. Pick the day least likely to get bumped and protect it the way you'd protect a client call.

Does a weekly review replace deeper monthly or quarterly planning?

No — it feeds it. The weekly review is short-term triage: where next week's attention goes. Bigger calls, like repricing a product or shutting one down, deserve a longer, less frequent session with more context than 15 minutes allows. The weekly ritual just makes sure you notice the signal in time to schedule that longer session before it becomes urgent.