Ask a portfolio publisher what reporting costs and they'll say "an hour in the morning." Watch the actual morning and it's this: six dashboards, four CSV exports, one master spreadsheet, and a growing suspicion that AdMob and Play Console are describing different companies. Multiply by every weekday, add the weekly deep-dive and the month-end reconciliation, and the honest bill lands near 17 hours a week — not analyzing, just assembling.
The tax has a second, worse cost: by the time the sheet balances, the decision energy is spent. Data that takes hours to assemble gets assembled — and then merely admired.
The daily review answers exactly three questions, in order, and then ends:
Two things are banned. No editing — the daily glance flags, it never fixes; mid-week budget surgery is how weekend noise becomes Monday regret. And no touring — if nothing is exceptional, the review is allowed to be boring and 90 seconds long. Boring means healthy.
The weekly pass is where changes happen — it's the enforcement arm of the scale/hold/kill framework: every campaign gets its verdict confirmed or changed, budget steps execute, watch-clocks tick, and exactly one experiment per hold-tier campaign gets queued. Flags raised by the daily glances get resolved here, with a full week of data instead of a morning's adrenaline.
Once a month, altitude goes up: the app-level portfolio table from the True ROI Playbook (which apps deserve to exist), the country P&L (where the map moved), and the revenue-mix check (is reality still matching the design). Sunset decisions, budget reallocation between apps, and monetization-mix corrections live here — decisions too heavy for a Tuesday, too important for "someday."
The traditional review is a tour: open every dashboard, look at every chart, hope something jumps out. Tours fail for a structural reason — human review runs on a schedule, but the moves that matter don't. A geo breakout that starts Wednesday afternoon waits four days for a Monday deep-dive; by then the window has half closed.
Anomaly-first inverts it: baselines watch every series continuously, and the review starts from what deviated. The demo portfolio's ES spike — revenue running +22% against baseline — was flagged four minutes after first sync. A weekly tour finds that pattern eventually; anomaly detection finds it while shifting budget still captures the wave. (What machines catch that calendars miss is its own story — we wrote it up on the blog.)
This cadence works in a spreadsheet — with one honest asterisk. The 4-minute daily assumes the numbers are already assembled when you sit down. If your morning starts with six logins and a CSV ritual, the routine is 4 minutes of judgment on top of an hour of plumbing. Automate the plumbing — with us or with anything — and the playbook keeps its promise.
Connect your first platform in minutes and see what your portfolio actually earns — after spend, every day.