Fix the brief before it costs you the shortlist

Adding requirements to a job description does not raise the quality of applicants. It changes the composition of the applicant pool, and it changes it in a direction most hiring managers do not intend. Candidates who read a fifteen-item essential list and hold twelve of the items behave differently depending on how confident they are, and confidence is not evenly distributed across the people you want to hear from. The applicants you lose are disproportionately the ones who assess themselves honestly, which is not the trait you were trying to filter for.

The list also has a second effect that is easier to measure. A long essential list makes the description slower to read, less specific overall because attention is spread thin, and harder for your own screening to use, since a screener weighting fifteen essentials weights none of them. Ten pass-marks and five failures looks the same as five and ten if everything is essential.

The analyzer counts the essential requirements and flags the list when it grows past the point where the word essential still means anything. It does not tell you which to cut, because that is a hiring decision. It tells you that the decision is being avoided.

The test that works is a hiring test, not a writing test. For each item on the essential list, ask whether you would reject an otherwise excellent candidate solely for lacking it. If the answer is no, or if it is yes but you know you would not actually do it, the item is a preference. Write it as one.

The second test is about the first ninety days. A must-have is something the person needs on day one because there is no time or route to acquire it: a licence they cannot practise without, a language the team works in, a legal right to work in the location. Almost everything else, including most tools and most frameworks, is something a capable person picks up, and calling it essential is a shortcut that costs you candidates.

Doing this properly usually cuts an essential list roughly in half and moves the rest into a preferences section that hiring managers still get to weigh. Nothing is lost. The description gets shorter, the screening gets sharper, and the people who would have self-rejected apply.

Very little of this is deliberate. It is inherited phrasing, copied from a previous description, that carries a signal nobody intended. Terms that read as an age preference, recent graduate, digital native, or the reverse. Physical requirements attached to a role that does not need them. Culture language that describes a demographic rather than a way of working. Sports and military metaphors used as a proxy for drive. Requirements for continuous employment that penalise anybody who took a caring break.

The analyzer flags the phrase and says why it might narrow the pool, and then stops. It does not rewrite the description for you and it does not tell you the phrasing is unlawful, because whether a given phrase creates a legal problem depends on your jurisdiction, the role and the facts, and that is a question for your counsel rather than a text tool. It is an observation, and observations are what a human can act on.

The most useful thing a job description can do is stop the wrong people applying, and the way it does that is by including the facts that people actually decide on. A salary range or band. The location and the genuine working pattern, including how many days on site and whether that is negotiable. The interview process, how many stages and roughly how long. Who the role reports to. Whether the role is new or a replacement. Contract type and duration for anything not permanent.

Descriptions that leave these out do not get more applicants of the kind they want. They get more applicants of every kind, plus a set of conversations that end at the first screen because the salary was never going to work. Every missing fact is a screening call somebody has to run to establish something that could have been one line in the ad.

The analyzer lists what is missing from that set. It is usually the fastest improvement available on any given description.

The analyzer will process the text you paste to produce the feedback and will not retain it. Nothing is written to a Surhires account, no account is needed, and the text is not used to train a model or to build a data set of job descriptions.

A job description is normally your own content. Two things change that. If you are an agency working from a client's draft, the text may be commercially sensitive and covered by your terms with them, and pasting it into a free web tool is a decision to make deliberately rather than by habit. If the draft contains a hiring manager's name and contact details or internal band information, that is other people's data and internal information travelling with it. Strip both before pasting; the feedback is unaffected.

It will not score the description out of a hundred. A score invites optimisation against the scorer, and a description tuned to please a text tool is not a better description. The output is a list of observations with the reasoning attached, which you can disagree with individually.

It will not rewrite the description. It will not tell you whether a phrase is lawful in your jurisdiction. It will not predict how many applicants you will receive, because that depends on the market, the salary, the employer and the channel far more than on the wording. And it will not check the salary disclosure requirements that apply where you are publishing; that is a different tool on this site, and the two are worth running together.