Marketers have a specific problem on X: they know how to write for a brand, and a personal account written like a brand is invisible. The posts that work for a marketer are the ones a brand could never publish. A teardown that names what is wrong. A campaign result with the cost attached. An opinion that a client would ask you to soften.
The fourteen posts below are written for a marketer's own account, grouped into teardowns, results, opinions and tools. Each is complete, with a note on why it works. The campaigns and numbers are illustrative. Yours, with the real costs, will do better.
Why marketers sound like brands
Years of writing copy that has to be approved teaches you to remove the specifics: the client name, the budget, the thing that failed. Those specifics are exactly what a personal account runs on. You do not have to name clients, but you do have to name numbers, and you have to say what you actually think about the tactic everyone is recommending this month.
The test for a marketer's post: would a brand account be allowed to publish this? If yes, it is probably too safe. If a brand would need three approvals, it is probably a good post.
Teardowns
A pricing page teardown
Tore down a well-funded startup's pricing page today. Three plans. Nine features each. Twelve of the 27 cells say "coming soon". You cannot sell a roadmap in a table.
- Why it works:
- A counted observation and a one-line principle. Counting the cells is the work; most people would have said "cluttered".
An ad teardown
Saw a billboard with a QR code on a highway. Someone approved that. Someone else drove past it at 70mph. The medium is the brief.
- Why it works:
- An everyday sighting turned into a point about channel fit. No brand is named, so the lesson travels.
An email teardown
An onboarding email I received today opened with "We're thrilled!" They are not thrilled. They are a cron job. Write the email the founder would send if they had five minutes.
- Why it works:
- Funny, specific and ends with a rewrite instruction. Marketers save posts that tell them what to do instead.
A good example, pointed at
The best unsubscribe page I have seen this year belongs to a tiny newsletter about fonts. One line: "Fair enough. The archive stays free." I almost resubscribed.
- Why it works:
- Praise teardowns are rarer than critical ones and get shared by the people praised. The quoted line does the teaching.
Results with context
A result is a post when it has a cost and a verdict. Spend, outcome, what you would do differently. Results without costs are brags; costs without outcomes are complaints. Anonymise the client and keep the numbers.
A channel result with the cost
Spent $500 on a newsletter sponsorship. 410 clicks, 38 signups, 3 paid in the first month. Cost per paid customer: about $167. Payback: 6 months. Would run it again at $400.
- Why it works:
- The full funnel in three lines and a decision at the end. Readers can compare it with their own numbers immediately.
A result that failed, diagnosed
Ran cold outreach for a client for six weeks. 800 emails. 11 replies. 1 meeting. 0 deals. The list was the problem, not the copy. Should have spent week one on the list.
- Why it works:
- Failures with a diagnosis are the most useful posts marketers write and the rarest. The diagnosis is a lesson people can act on.
A small change with the effect
Changed one word in a client's headline: "software" to "spreadsheet". Their users call it a spreadsheet. Signups moved within the week. Use the customer's word, not the category's.
- Why it works:
- One word, one principle. It avoids claiming a precise lift while still showing the direction and the cause.
A long-term result
Blog post I wrote in 2023 for a client still brings them 40 signups a month. The campaign it was part of ended that quarter. Some marketing is a bill. Some is an asset.
- Why it works:
- Time is the context here. The last line is a framing people quote in budget meetings.
Opinions on tactics
A tactic you stopped recommending
Stopped recommending giveaways. Every one I ran grew the list by a few thousand and the open rate by nothing. The prize buyers never became customers.
- Why it works:
- An opinion backed by repeated experience, not a hot take. The people who disagree will bring their own numbers.
A hot take with a diagnosis
hot take: most startups do not have a marketing problem. They have a "we cannot say what this is for in one sentence" problem.
- Why it works:
- The format signals an opinion is coming, and the opinion reframes a familiar complaint. Founders quote this at each other.
A rule you keep
Marketing rule I keep: if the headline needs a subheadline to make sense, the headline is wrong.
- Why it works:
- One line, instantly testable against any page the reader has open. Rules this short get saved and repeated.
Tools and comparisons
I don't need
I don't need another analytics dashboard. I need one that tells me which page the paying customers read before they paid.
- Why it works:
- The gap between what tools offer and the one question you actually have. It names a need, not a vendor.
A tool switch with what was lost and kept
Moved a client from a $300 a month email tool to a $30 one. Lost: a drag and drop editor nobody used. Kept: everything the campaigns actually needed.
- Why it works:
- Price difference plus an honest account of the tradeoff. Comparisons without a loss column read as ads.
A comparison with a verdict and a reason
Tested two scheduling tools for a month each. The cheaper one won because its analytics showed replies, not just likes. Replies are where the customers were.
- Why it works:
- The verdict is explained by a criterion the reader can adopt, which is more useful than the winner's name would be.
From campaign to post
- 01After every campaign, write three lines before the retrospective meeting: what it cost, what it produced, what you would change. That is a post before it is a slide.
- 02Anonymise the client and keep every number. The numbers are the post; the name was never the point.
- 03Once a week, tear down one thing you saw as a consumer: an ad, an email, a page. Count something in it. The count is your hook.
- 04Once a month, write the opinion you keep giving clients in private. If three clients have heard it, the feed should too.
- 05Read what people reply and answer with a number or a case, not a thank you. The reply examples for marketers show the shape.
- 06Keep the brand voice for the brand. On your own account, say the thing the approval process would have cut.
The brand-voice trap
Do
- Costs and outcomes together, with your verdict.
- Opinions you have earned by trying the opposite.
- Teardowns that count something and name a principle.
Don't
- "Excited to share" and anything else from the approved-copy vocabulary.
- Results without costs, or wins without the campaign that failed before them.
- Marketing advice in the abstract. Attach it to the campaign that taught you.
A marketer's personal account is an audit trail of judgement. Every teardown, result and opinion is evidence of how you think, which is what clients and employers are actually buying. The analytics guide is worth reading once so your own account's numbers get the same honesty you give a client's, and the hook patterns apply to marketers as much as to the people who build the products.
BuzzyX · X growth workspace
Grow on X with posts that sound like you.
Compose in your voice, schedule the week, find posts worth replying to and read real analytics from the X API. $29/mo after the trial, everything included.
Frequently
asked questions
Something else on your mind? Get in touch.
Can I post client results on my own account?
Post the numbers without the name unless the client agrees to be named, and check your contract for confidentiality terms. Costs, conversion counts and your verdict are usually fine anonymised. Revenue figures and customer lists are not yours to share.
Should a marketer's personal account promote the company they work for?
Sometimes, as a case or a lesson, never as copy. Posting your employer's campaign with your honest assessment of what worked is a good post. Reposting the brand account is not.
How critical can a teardown be?
Critical about the work, never about the people. Count what is wrong, say what you would do instead, and avoid naming small companies who cannot absorb the attention. Big brands and your own past work are fair targets.
What if I do not have campaign results to share yet?
Teardowns and opinions need no budget. Tear down one thing a week you encountered as a consumer and write the rule you keep giving people. Results come as you run things, and by then people will be reading.
Are hot takes a good format for marketers?
Occasionally, when the take is a diagnosis you have actually reached. A weekly hot take with nothing behind it reads as the brand voice in a different costume. One a month, earned, is plenty.