Two Weeks, One Spreadsheet, Zero Dollars: Running Your Own Content Retrospective

Somewhere in your content archive is a pattern you have never noticed. A format that quietly outperforms everything else. A topic your audience tolerates rather than wants. A publishing slot that has been dead for a year while the calendar keeps feeding it. Most teams sense these patterns exist; few ever sit down and find them, usually because an audit sounds like it requires new software, a consultant, or a budget line.

It requires none of those things. Native platform analytics plus a spreadsheet and roughly two weeks of part-time attention will surface more actionable truth than most paid audit tools. Here is the working method.

Week One, Days One and Two: Build the Ledger

Open a spreadsheet and create one row per post for your last hundred — across whichever platforms matter most to your business. Columns to include:

• Post date and platform
• Format (video, carousel, static image, text, link)
• Primary topic, in plain language ("hiring tips," "product updates")
• CTA type (none, follow, click, comment, DM)
• Reach or impressions
• Engagement rate (native analytics will give you this)
• Clicks, if the platform reports them
• Saves and shares, where available
• One-word gut rating: strong, fine, weak

That last column matters more than it sounds. It creates a record of instinct you can later test against evidence — often the most educational part of the exercise.

Days Three Through Six: Pull the Data Without Getting Lost in It

Export or transcribe the numbers from each platform's native analytics. Do not normalize, benchmark against industry averages, or build a scoring model. The single most common way teams stall an audit is by over-engineering the methodology before looking at anything. Raw numbers, entered honestly, are enough.

Two practical notes for US small teams: LinkedIn and Instagram both allow CSV exports of post-level data; for platforms that do not, a shared screen and a coffee will get a hundred rows entered in an afternoon. Assign the transcription to whoever is newest to the team — the familiarity with the archive is a bonus, not a chore.

Week Two, Days One and Two: Sort Until Something Is Strange

Now the analysis, in deliberately simple form:

1. Sort by engagement rate, top twenty and bottom twenty. Read the top and bottom groups slowly. What do the winners share? Format? Topic? CTA? Tone? Write down every hypothesis, even the ones that feel obvious.
2. Repeat the sort by shares and saves. Reach measures what the algorithm did; shares measure what humans chose to pass along. Divergence between the two lists is often where the real insight hides — high-reach, low-share posts are algorithm flukes, while modest-reach, high-share posts are your actual word-of-mouth engine.
3. Filter by topic. Which subjects earn disproportionate engagement? Which produce consistent silence? Most teams discover one or two topics carrying the account and a long tail of filler diluting it.
4. Filter by CTA. Compare posts that asked for something against posts that asked for nothing. The results are rarely what teams expect.

Days Three and Four: Answer Three Questions

Resist the urge to produce a forty-slide deck. The audit pays off when it collapses into three answers:

• What should we make more of? Name the top two formats and topics by evidence, not preference.
• What should we stop making? Every retired content type frees capacity for the winners.
• What do we genuinely not know? Honest gaps become next quarter's experiments.

One page, circulated to everyone who touches content, beats any dashboard.

Day Five: Turn Findings Into a Rulebook, Not a Memo

Convert the answers into three to five explicit publishing rules — "every week includes at least two posts in the top-performing format," "link posts are limited to Fridays," "every carousel ends with a save prompt." Rules force the audit to change behavior. Then calendar the next audit for ninety days out. The ritual, repeated, is what compounds; a single audit is a snapshot, but a quarterly one is a feedback loop.

Why Free Beats Fancy Here

Paid audit tools are useful at scale — hundreds of posts, dozens of accounts, competitive benchmarking. But for a team publishing a few times a week, the bottleneck was never computation. It was the hour of honest attention nobody had scheduled. Two weeks, one spreadsheet, and a standing date on the calendar will tell you more about your audience than any dashboard you have not opened since onboarding.