I Played Detective on My Own Feed for a Week — the Case Files Were Uncomfortable

Last month I did something I'd been avoiding for a long time: I scrolled through twelve months of my own content like I was a marketing agency being paid $5K to review it.

No skipping. No softening. I opened a spreadsheet, poured an iced coffee, and went post by post like my feed belonged to a stranger whose retainer check was due Friday.

What I found stung. But it also changed how I plan everything going forward.

The Rules of the Audit

First, the ground rules I set for myself, because "just look at your feed" is a recipe for an afternoon of vanity scrolling:

• I was not allowed to defend anything. Every note went in the spreadsheet as-is. No "well, actually, that post did fine because..."
• I reviewed in chronological order, oldest to newest, so tone drift would show up naturally instead of me cherry-picking.
• I labeled every post with a format (reel, carousel, story dump, text post), a topic, and a gut rating from 1–5 on whether it sounded like me.

That last column — the "does this sound like me" rating — is where things got interesting.

Finding #1: My Tone Drifted and I Never Noticed

Around month five, my captions started sounding like a LinkedIn post wearing a Halloween costume. Polished. Safe. Full of phrases like "excited to share" and "thrilled to announce."

I never made a conscious decision to sound like that. It happened the way it always happens: one post performed well, I unconsciously copied its energy, and the copy of a copy of a copy became a stranger.

When I read those posts back-to-back, I genuinely asked myself, "Who wrote this?" The answer, apparently, was a person trying very hard to be taken seriously by people I'd never meet.

Finding #2: Three Formats Were Dead and Still on Life Support

I had a weekly series I'd started two years ago. It was my baby. It was also, per the spreadsheet, averaging a third of my usual engagement for eight straight months.

I kept producing it because it felt responsible. Like flossing. But the data said my audience had moved on, and honestly, so had I — I was phoning it in by month three of the series anyway.

The audit also surfaced two other zombie formats: a monthly Q&A nobody asked questions for, and a style of carousel that took me six hours to make and earned a shrug from the algorithm.

Killing all three freed up roughly a full workday per week. A full day. I'd been paying a hidden tax out of loyalty to my own past ideas.

Finding #3: The 20% Was Doing Everything

This is the part every creator hears and no creator believes until they see their own numbers: a small cluster of posts generated the overwhelming majority of my reach, saves, and — most importantly — inbound inquiries.

For me it was the long-form, opinionated posts where I told an actual story with an actual stake in it. Not the trending audio. Not the aesthetic flat lays. The posts where I said something I could get in trouble for at a dinner party.

So the new math is simple: more of the 20%, less of everything else, and the freed-up hours go toward deeper versions of what already works.

The Moment That Got Weird

Here's what nobody warns you about: somewhere around month eight of the scroll, you stop being the auditor and become the person being audited. I wasn't looking at posts anymore. I was looking at a record of my energy, my burnout, my attempts to impress people, and the months where I clearly posted out of obligation.

I actually teared up a little at a caption from last November — a raw post I'd almost deleted before publishing. Reading it as a stranger, I thought, "This is the person I'd follow." Then I scrolled two weeks forward and met the corporate-sounding imposter who took over after a big brand deal.

That whiplash — seeing your brand through a stranger's eyes — is the entire value of the exercise. You can't unsee it, and you shouldn't want to.

The Checklist I Now Run Every Quarter

Screenshot this or copy it into your notes app:

1. Export or scroll 12 months of posts, oldest first. No skipping.
2. Tag every post: format, topic, effort level (low/med/high), and a 1–5 "sounds like me" score.
3. Pull the top 10 performers by saves and shares — not likes. Saves tell you what people actually valued.
4. Identify your 20%. What do your best posts have in common? Topic, tone, format, length?
5. Name the zombies. Any format running 3+ months below your average? Write down why you're keeping it. If the answer is "nostalgia," schedule the funeral.
6. Read your last 10 captions in a row. Would a stranger describe your voice the way you'd describe it? If not, find the exact post where the drift started.
7. Check your bio and pinned posts against your best-performing content. Do they promise the same thing?
8. Pick one thing to kill and one thing to double before the next audit. Two changes, not twenty.

The Verdict

The audit took me two evenings and saved me from another year of guessing. My feed is smaller in ambition now and bigger in impact — fewer formats, a clearer voice, and a calendar built around what actually works instead of what I wish worked.

The mirror is unflattering. But so is every good client report, and you'd never let a client keep paying for content that wasn't working. You deserve the same honesty you'd give them.

Grab the coffee. Open the spreadsheet. Your feed has been trying to tell you something for months — it's time to finally listen.