Determine Your Ideal Career Using Evidence You Already Have

Your ideal career is described in a document you already have

Open your CV, or your last three job descriptions if the CV is thin. Ignore the job titles. Titles are the least informative thing on the page, because two people with the same title do different work at different companies, and you are trying to find out about the work.

Instead, break each role into the tasks that took your hours. Aim for eight to fifteen tasks per role, written as verbs with objects: "reconciled the month-end ledger", "ran the standup", "wrote the tender response", "chased late suppliers", "trained the new starters", "built the reporting spreadsheet everybody used".

Then mark each task on two axes, and only two:

• Was I good at it? Judged by evidence rather than feeling. Did somebody hand you more of it, did the number move, did people ask you rather than the person whose job it was.
• Would I want to do more of it? Not "was it pleasant" but whether you would take a role built mostly out of it.

Four groups fall out. The tasks you are good at and want more of are your direction. The ones you are good at but do not want are the trap, because they are what recruiters will keep offering you and what your CV currently advertises. The ones you want but are not yet good at are your training plan. The rest you can stop worrying about.

Most people find the second group is the largest, and that is the real finding. Being pulled towards work you can do but do not want is the ordinary way careers go wrong, and it does not show up on any personality test.

What a personality test can and cannot tell you

Nearly every guide on this topic recommends the Myers-Briggs Type Indicator at step three. It is worth knowing what the research says before you build a decision on the result.

A review commissioned by the US Army Research Institute concluded that the instrument should not be used for career planning counselling, finding no evidence of utility for that purpose. Separately, researchers have found that the distribution of types within a given profession is close to the distribution in a random sample of the population, which is the finding that matters here: if accountants are not disproportionately one type, knowing your type tells you nothing about whether to become one. The instrument was not designed to measure aptitude or to predict performance, and its own publisher does not claim it does.

The publisher disputes the reliability criticism and reports test-retest correlations above 0.80 over periods up to fifteen weeks. Critics point to people who retake it and get a different type, which the dichotomous scoring makes likely for anyone sitting near a boundary. You do not have to resolve that argument to use the tool correctly.

What these instruments are reasonable for: starting a conversation about how you prefer to work, giving you vocabulary for something you already sensed, and prompting questions you would not have asked. What they are not for: choosing between two careers, explaining a rejection, or overriding what your own task audit told you. Treat the result as one more piece of self-report, weighted below the evidence of what you have already done.

Check demand and pay before you fall for a job title

A direction that nobody is hiring for is a hobby. Before you commit, spend an hour on the numbers, and use free official sources rather than a careers blog with an affiliate link.

• United States. The Bureau of Labor Statistics publishes the Occupational Outlook Handbook, which gives median pay, entry requirements, and projected growth by occupation.
• United Kingdom. The National Careers Service publishes job profiles with salary ranges, entry routes and required skills.
• Elsewhere. Look for the national statistics office or the labour ministry before you look at a jobs board, then use the boards for the reality check: how many roles are open in your city this month, and what do they ask for.

Read live adverts as data rather than as opportunities. Twenty adverts for the role you are considering will tell you the qualification that keeps appearing, the tool everybody wants, and the salary band that is real rather than aspirational. That is a better hour than any quiz.

Test it cheaply, in this order

The mistake that costs the most is buying an expensive test of a cheap hypothesis. People enrol on a two-year course to find out whether they like a kind of work that one afternoon would have told them about. Work up this ladder and stop as soon as you have your answer.

• A conversation, one hour. Find two people doing the job and ask what the worst part of the week is. Not what they love. What they dread, and how often it happens.
• A day beside somebody. Shadowing is cheap and most people say yes if you ask specifically and turn up on time.
• One small piece of real work. A freelance job, a volunteer task, a project at your current employer that nobody wants. This is the first rung that tests whether you can do it rather than whether you like the idea.
• A short course with an assessment. Weeks, not years, and only once the earlier rungs have held.
• A placement or a junior role. A real pay cut for real evidence.
• A qualification. Last, and only when the rungs below have all pointed the same way.

If you are still studying, the internship cover letter guide covers how to get onto rungs three and five, including the recruiting calendar that decides whether your application arrives in time.

What a switch costs you

The guides skip this, and it is the part that keeps people awake. Two moves are available and they cost very different amounts.

Changing industry while keeping your function is the cheap move. An accountant who moves from retail to healthcare keeps the qualification, most of the technical vocabulary, and the seniority. What resets is the domain knowledge and part of the network, and that comes back inside a year.

Changing function is the expensive one. You re-enter at or near the bottom of a ladder you did not climb, your salary band resets, and the ten years of credibility that made your opinion count no longer transfers. Some of it survives: judgment, writing, knowing how organisations behave, and any technical skill the new function also uses.

Price it before you decide. Write down the salary you would take, how long you can fund the gap, and what the first year would look like at that number. A direction that survives being costed is a decision. One that does not is a daydream, and finding that out on paper is cheaper than finding it out in March.

A worked audit

Priya Ramanathan spent six years in logistics, the last three as a shift supervisor at a distribution centre. Her CV said she managed a team of 22 and cut pick errors by 30 percent, so recruiters kept offering her larger teams on later shifts. The task audit put those tasks in the second group: good at, does not want more of.

The first group was smaller and looked unrelated. She had built the shift's reporting spreadsheet, which three other sites then copied. She had written the standard operating procedure the trainers used. She had spent two weeks untangling why the scanner data and the stock count disagreed, and enjoyed all of it. Nobody had ever paid her for any of that.

Twenty live adverts told her the shape those tasks take in the market and what the entry requirement was. Two conversations told her the worst part of the week. One unpaid weekend rebuilding a friend's stock report told her she still liked it at hour six. She costed the move, took a role a grade below her supervisor level, and was back to her old salary in eighteen months. The point is not the outcome; it is that four steps, none of which cost more than a weekend, replaced a decision she had been putting off for two years.

The worksheet

Print the grid, or use the .docx, which holds the task audit table, the switching-cost calculation, and the questions to ask in the two conversations.

Where this goes wrong

• Auditing job titles instead of tasks. The title is the label somebody else chose. The tasks are what you did.
• Marking "good at it" by feeling. Use evidence somebody outside your head could see.
• Treating a test result as a verdict. It is self-report about preference, not a measurement of aptitude.
• Skipping the costing. The number is the difference between a plan and a wish.
• Waiting for certainty. The ladder exists because certainty is not available. Rung one is a phone call.

Once you have a direction, the documents follow. If your history is thin, start with the first resume guide; if it is long but points the wrong way, rewriting the bullets around the tasks you want more of is what changes which roles you get offered.

Sources

• Myers-Briggs Type Indicator: review commissioned by the US Army Research Institute concluding the instrument should not be used for career planning counselling, with no evidence of utility found for that purpose; research finding the proportion of types within occupations close to that of a random population sample; the instrument's stated design, which does not include measuring aptitude or predicting job performance. The publisher disputes reliability criticism and reports test-retest correlations above 0.80 for periods up to fifteen weeks, which is recorded here alongside the criticism rather than in place of it.
• Labour market data: US Bureau of Labor Statistics Occupational Outlook Handbook for median pay, entry requirements and projected growth by occupation; UK National Careers Service job profiles for salary ranges, entry routes and skills.
• Career-choice guidance from the major career publishers, which converge on self-assessment, personality testing, research and hands-on experience, and which do not cover systematic task auditing, the cost of switching, or an order of testing by expense.