How to decide which jobs are worth applying for

Most advice about job applications answers the wrong question. It asks whether you could do the job, which is nearly always arguable, and almost never asks the question you are really facing at eleven at night with nine tabs open: is this particular posting worth the next hour of my life?

Those are different questions. The first has no stable answer — you can talk yourself into or out of almost any role. The second has a good one, and it can be reached in a few minutes if you know what you are looking at.

What follows is a method, not a formula. It produces one of three outcomes for every listing you read, and it is designed so that two of them cost you nothing.

A job posting is written by several people with different motives, and it shows. Somewhere in it are the two or three things without which nobody gets hired. Around them is a wider ring of preferences, and around that a layer of language that exists because the last posting had it.

Read for three kinds of line. Lines that describe what the person will be accountable for. Lines that name a specific, checkable thing — a tool, a domain, a language, a certification, a right to work somewhere. And lines that could appear in any posting for any role.

The first two matter. The third is noise, and treating it as a requirement is the single most common reason people rule themselves out of roles they would have got. How to read a job description goes through the separation line by line.

Take the accountabilities and the specifics you just pulled out, and against each one ask a narrow question: what, in writing, already demonstrates this?

Not what you could explain in an interview. Not what you would pick up in a fortnight. What is already on your CV, in a portfolio, in a public repository, in a reference someone would give — a fact a stranger could check.

This is a harsher test than it sounds, and it is the one that saves time. A requirement you can evidence is a reason to apply. A requirement you can only argue is a risk you are choosing to take, and knowing which you are holding changes how you spend the evening.

Doing this properly against a single posting takes about ten minutes the first few times and under five once it is habit. How to check your CV against a job description is the longer version of this step.

Every listing you read should end in one of three states, decided on the spot:

• Apply properly. You can evidence the accountabilities and most of the specifics. This one gets a tailored application and real effort.
• Apply if the week allows. You can evidence the accountabilities but not some of the specifics, and you can explain the gap credibly. Worth sending — after the first group, never instead of it.
• Skip, and write down why. The gap is in the accountabilities themselves, or the posting requires something checkable that you simply do not have, such as a right to work in a country you cannot work in.

The third outcome is the one people skip, and it is the one that makes the method work. A search where nothing is ever ruled out is not a search; it is a queue that grows faster than you can empty it.

Write the reason down in a few words. Two things happen. You stop re-reading the same posting on three different evenings, and after a fortnight the reasons form a pattern — usually a specific requirement that keeps appearing and that you could actually go and acquire.

Applications are not free. Each one costs reading time, tailoring time, the small tax of a form, and a piece of the attention you need for the interviews you are trying to reach. Sending more of them is not obviously better; it is obviously more work.

The case for volume rests on an assumption worth naming: that applications are interchangeable and that enough of them will eventually land. If that were true, the fastest applicant would win. In practice a tailored application to a role you can evidence beats several rushed ones, and the rushed ones are what stop you having time to tailor.

So a skip is not a failure to apply. It is an hour moved to the application that deserved it. How to prioritise job applications when your time is short is about what to do with the hours the skips give back.

One distinction is worth holding on to, because it decides more applications than any other. Some gaps are about evidence and some are about eligibility.

An evidence gap is a thing you have done that your CV does not yet show, or a thing adjacent to what you have done. Those are arguable, and often worth arguing — whether to apply when you do not meet every requirement is a question with a better answer than most people assume.

An eligibility gap is binary and nobody can be talked round it: a licence you do not hold, a language the role is conducted in, a location you cannot legally work from. Spending an evening on one of those is the clearest waste in a job search, and it is entirely avoidable by reading for it first.

Software is good at the mechanical half of this. It can read a long list of postings, pull out the accountabilities and the checkable specifics, compare them against what a CV actually evidences, and put the listings in an order that reflects that comparison. Done honestly, that removes the part of the work that is pure attrition.

It cannot tell you whether you want the job. It cannot see the internal candidate, the budget that is about to be frozen, or the team you would be joining. And a tool that returns a number without showing the lines of the posting the number came from has not helped you decide anything — it has asked you to trust it. Why a percentage is not a decision is about the difference between the two.

Read the posting for accountabilities and checkable specifics. Ask what already evidences each one. Decide, on the spot, between applying properly, applying if the week allows, and skipping with a written reason. Protect the first group's time with the third group's decisions.

None of it requires a tool. It requires the decision to be made once per posting instead of three times.