Most people change direction several times over a working life, and most of those changes start the same way: not with a plan, but with a Tuesday that finally goes badly enough.
That is a fine reason to start and a poor way to choose. What follows is the slower version — five phases, in order, each one making the next one cheaper.
Phase one: work out what you are actually after
Before you read a single posting, answer four questions honestly and in writing. What are you genuinely good at? What do you enjoy enough to do on a bad week? What do you want the work to produce? And where do those three overlap in a job someone will pay for?
Keep skill and strength apart while you do it. There is usually something you are excellent at and quietly sick of, and building the next role around it is the most common way to end up back here in two years.
Phase two: research before you commit
Pick three directions rather than one. For each, find out what the work actually involves day to day, what it pays at your level, and where demand is going. Job descriptions are free research: read twenty for the same role and the real requirements separate themselves from the boilerplate.
Then do the part most people skip and talk to two or three people already doing it. Fifteen minutes of someone describing their Tuesday will tell you more than a week of reading.
- Draw a two-column list: what these postings keep asking for, and what you can currently evidence.
- The left column with nothing beside it is your gap list — and it is usually shorter than it feels.
- Anything you can evidence from a different industry is a transferable skill. Name it in their language, not yours.
Phase three: build the materials once
You need three things: a CV that parses cleanly and is tailored per application, a letter that answers the posting requirement by requirement, and a profile that says the same thing to anyone searching.
Tailoring is the part that decides outcomes and the part that gets dropped first when you are tired. It is also the part software is genuinely good at now — Apply Engine reads the posting, rewrites the CV and letter against it and fills the form, and shows you all of it before anything sends.
Phase four: hunt on the routes that work
Job boards are one channel of several and the most crowded. Run them, but do not let them be the whole search.
- Make a list of twenty companies you would genuinely like to work for and watch their careers pages directly.
- Work outwards through people: your own network first, then alumni, then the communities in your field.
- Talk to two or three specialist recruiters who actually place your kind of role, and be specific with them about what you want.
- Take the volunteering, the side project or the talk that puts you in a room with people who hire.
Phase five: interview, then negotiate
Behavioural rounds reward structure: a brief situation, the task you owned, what you did, and what changed as a result, with a number in it wherever one exists. Prepare six of those and you can answer most of what you are asked.
Then negotiate, which most people skip out of relief. Find out what the role pays from more than one source, decide your number before the call, and justify it with the impact you have already had rather than with what you need. Ask about the whole package — bonus, flexibility, development budget, start date — because the parts that are not base salary are often the ones with room in them.
The point of the order
Each phase makes the next cheaper. Knowing what you want narrows the research; the research writes the CV; good materials make the hunt shorter; a short hunt leaves you enough leverage and enough patience to negotiate properly.
Done out of order, you end up applying widely to roles you have not thought about, with documents written for nobody in particular. Which is how most searches go, and why most of them take longer than they should.
