.png)
It may be your highest-intent piece of employer messaging, and almost nobody treats it that way.
Your company has a review process for the homepage. The CEO's annual letter goes through six drafts and a legal read. A product launch email gets tested against three subject lines. The brand team has opinions about the shade of blue.
The job post gets copied from the last one, edited by a hiring manager between meetings, and published by whoever had the ATS tab open.
Now consider the reader. They did not encounter this message by accident. They opened it because they are deciding whether this opportunity is worth more of their time. That makes a job post unusually high-intent corporate communication.
Yet it is still commonly treated as paperwork: no clear owner, no persuasive standard, and little review beyond whether the responsibilities and requirements are technically correct.
The Usual Way
Open a random posting at almost any company and the structure is identical, because everyone inherited it from everyone else.
A paragraph about the company, written in the voice of an investor deck. A paragraph about the team, containing the words collaborative, fast-paced, and mission-driven. A responsibilities list, which describes activities rather than outcomes. A requirements list of eleven to fourteen bullets, most of them carried forward from a version written for a different market. A benefits paragraph listing things every competitor also offers. An EEO statement. Apply now.
Nothing in that document is wrong, exactly. It is a competent description of a vacancy. And that is precisely the problem, because the reader is not trying to understand your vacancy. They are trying to decide between it and the other four tabs open in their browser.
The growth question
A job post is not a description. It is a competitive argument, submitted into a market, on behalf of a role, against named alternatives the reader is actively considering.
That single reframe changes everything downstream. A description asks "have we covered the role accurately?" An argument asks "why would the right person choose this over what else is available to them this month?" The first question can be answered by a form. The second requires knowing something about the market, the person, and what makes this role genuinely different from the eleven others they are weighing.
Most postings never attempt the second question. Which is why most postings do not persuade anyone, and why hiring managers keep asking why the pipeline is full of people who are not right.
Your post is doing selection work whether you designed it or not
Here is the thing that should make anyone take the document more seriously: word choice in job ads has measurable effects on who applies, and the effects run deeper than most people assume.
The foundational study is Gaucher, Friesen and Kay's 2011 work in the Journal of Personality and Social Psychology. They found systematic differences in the language of job ads across occupations, with masculine-coded words like leader, competitive, and dominant appearing more heavily in male-dominated fields. When women read those ads, the roles were less appealing to them.
The mechanism is the part worth sitting with. It was not that women judged themselves less capable of the work. It was belonging. The language signalled who the job was for, and readers responded to the signal rather than the requirements. The finding has been examined and extended in the years since, including replication work in entrepreneurship settings.
Generalize past gender and the lesson is bigger. Every posting is filtering, constantly, through vocabulary, framing, and emphasis that nobody consciously chose. Words carried forward from a decade-old template are telling readers who belongs here. The question is not whether your post selects. It is whether you selected the filter on purpose or inherited it from a document whose author has left the company.
The four things the post has to establish
Strip away the conventions and a persuasive posting does four jobs.
What will I actually do? Not the responsibilities list. The real texture of the work: what a week contains, what gets built or fixed or decided, what the first ninety days look like. Specificity here is the single cheapest credibility upgrade available.
Why does it matter? What changes in the business, for customers, or in the world because this seat is filled. A role with no consequence attached is asking someone to trade years of their life for a paycheck and an adjective.
What will I get from doing it here? Skills, scope, ownership, exposure, trajectory. What someone will be able to say they did, and be able to do next, because they took this job rather than the other one. This is where nearly every posting goes silent, and it is the thing candidates most want to know.
Why should I believe you? The one everybody skips. Claims without evidence are noise, and readers have been trained by a thousand identical postings to discount them automatically. If you say autonomy, show what decisions this person makes without approval. If you say growth, name the last two people who held the role and what they went on to do.
Build the argument out of specifics, tradeoffs, and proof
Three things separate an argument from a description.
Specificity, because generic claims are unfalsifiable and therefore ignored. "You will own the migration of a fifteen-year-old billing system to something maintainable, with a team of four and a two-year runway" tells a reader more than any paragraph about innovation ever could.
Tradeoffs, because a role that sounds uniformly good reads as either dishonest or unremarkable. Naming what is hard about the job is the strongest available credibility signal, and it does useful filtering work at the same time. The right person reads "the codebase is a mess and fixing it is the job" and leans forward. The wrong person leaves, at no cost to anyone.
Proof, because every claim you make is also being made by your competitors. Numbers, named programs, specific practices, real examples. Anything a reader could verify or picture is worth more than three sentences of well-crafted enthusiasm.
Subscribe for more ideas and insights
Cut everything that does not help someone decide
Most postings are long in the wrong places. The test for every element: does this help the right person make a decision?
The company boilerplate paragraph almost never does. Candidates who care can visit the site. The requirements list, at eleven-plus bullets, mostly does not, because half those bullets are aspirational and everybody knows it. Benefits that every competitor also offers do not differentiate, though genuinely unusual ones do. The word "rockstar," in any context, does not.
What survives the cut is usually shorter and always more specific. That is not a coincidence. Vagueness is what fills the space when you have not decided what you are arguing.
Your post is a public statement of posture, whether you intended one or not
There is a useful case study in what happens when a posting is forced to reveal something the company would rather not say.
Colorado's Equal Pay for Equal Work Act took effect on January 1, 2021, requiring employers to disclose compensation and a description of benefits in job postings, including for remote roles that could be performed from Colorado. A number of large national employers responded, not by disclosing pay, but by adding a line to their remote postings stating the role was not available to candidates in Colorado.
Sit with what that communicated. Faced with a choice between publishing a salary range and excluding an entire state's worth of candidates from a remote job, some companies picked exclusion, in writing, in the document candidates read most closely. Colorado's Department of Labor and Employment took a firm line, arguing that excluding Coloradans was a preference among applicants rather than a description of where the work could be performed, and warning of investigations and fines.
The legal question resolved. The reputational signal did not, because it never does. Every posting is a statement of posture. What you are willing to say, what you evade, what you would rather not put in writing. Readers infer from all of it, including from the parts you left out, and they infer fastest about pay.
The red-pen test
The fastest diagnostic ever devised for this document requires one minute.
Print the posting. Cross out your company name and logo. Hand it to someone and ask whether they can tell which of your three closest competitors wrote it.
If they cannot, you have not written an argument. You have written the category's generic template with your logo applied, and the reader will treat it accordingly, which is to say they will skim it and decide based on salary and brand recognition, the two axes where you are least likely to win.
Then run the harder version. Could your competitor publish this exact copy, unedited, and have it be equally true of them? If yes, then nothing in the document is doing competitive work. This is the same problem that sends companies looking for an EVP project when what they actually need is the thing the project was supposed to produce: a specific, provable reason to choose them.
What good looks like
The old measure of a job post is applications. More is better, and a posting that generates volume is deemed successful.
That measure is actively misleading. A posting can double applications by becoming vaguer, because vagueness lets more people imagine themselves in the role. You will feel productive and your hiring managers will be furious, because they are the ones reading the résumés.
The better standard has two parts, and they run in opposite directions. The right people should lean forward, recognizing something specific that fits what they want next. The wrong people should opt out, cleanly and early, before consuming your screening capacity and their own hope.
Which means a good rewrite can reduce your application volume and improve every downstream number: screen-to-interview conversion, hiring manager satisfaction, offer acceptance, and the odds that the person is still there in a year. If you rewrite a posting well and volume drops while quality rises, that is not a mixed result. That is the whole objective.
Nobody owns the job post. That is the opportunity. It is the single highest-leverage piece of writing available to a talent function, it requires no budget, and it is sitting unclaimed in an ATS field that everyone treats as paperwork.
