Analytics dashboards fail the same way most tracking systems do: they offer far more data than anyone actually uses, so most people check them occasionally, absorb a vague impression, and change nothing concrete. A weekly review that works does the opposite — it's small, fixed, and ends in an actual decision every time.
The three things worth checking, every week
Your top post by engagement. Not to admire it — to identify specifically what about it worked (format, topic, timing) so you can deliberately try a variation, rather than hoping it happens again by chance.
Its decay curve, not just its total. A post that front-loaded all its engagement in the first day is a different signal than one that kept earning steadily — see extending a post's shelf life for what the second pattern specifically indicates and how to design for more of it.
Your best-time window for the coming week. Derived from your own data, not a generic chart (see the full case for this) — this directly shapes when you schedule what's coming next, not just an interesting number to note.
That's the whole review. Three numbers, one clear "what does this tell me to do differently" question per number.
Why more metrics usually make the habit worse
It's tempting to add follower count, demographics, reach trend, and posting frequency to every weekly check — all genuinely useful metrics, covered elsewhere on this blog. The problem isn't that they're not worth checking; it's that a review with too many numbers to look at every single week gets skipped the first time you're busy, and a skipped habit rarely restarts on its own. Keep the weekly review to three; move the slower-moving metrics (demographics, follower trend shape) to a monthly check instead, since they don't shift fast enough to need weekly attention anyway.
Ending every review with a decision
A review that doesn't change anything concrete isn't actually being used — end each one by naming one specific thing you'll do differently next week: repeat a format, adjust a time, retire something that clearly isn't landing. If a review three weeks running produces no decision, that's a sign either nothing's actually changing week to week (fine, check monthly instead) or the review itself isn't being taken seriously enough to act on.
Making the habit survive busy weeks
Attach the review to something that already happens reliably — right before your weekly content batch session, say, so it's structurally part of an existing routine rather than a separate thing to remember. A habit tied to an existing anchor survives far longer than one relying purely on remembering to do it.