What employers want · 7 min read
How do I use the time when I'm not getting a job?
A structured search, small real work that leaves something to show, conversations that lead to referrals and volunteering have the best evidence. Courses pay off slowly.
The short answer
Use the weeks to make things a hirer can see. The best-evidenced uses of time between jobs are a structured search routine, small pieces of real work in your direction that leave an artefact, conversations with people who do the work, and volunteering that uses a skill. Randomised trials of a structured job-search programme found 53% of participants re-employed at four months against 29% of a control group. Courses help, but a meta-analysis of over 200 programme evaluations found training pays off over one to two years, not weeks, and only when it ends in something you can show.
On this page 5 sections
Being out of work is two problems at once: the money, and the question of what to do with the hours. This guide is about the hours. The evidence on what helps is clearer than most advice suggests, and most of it points the same way: structure, real work, and other people.
What the evidence says helps
| Use of time | What the evidence says | How strong |
|---|---|---|
| A structured, supported search | 53% re-employed at 4 months versus 29% of controls; fewer depressive episodes 2.5 years on | Strong (randomised trials) |
| Small real work that leaves an artefact | Work samples are among the better predictors employers have; a sample gives them what they already trust | Moderate |
| Conversations that lead to referrals | Referred candidates hired at about 1 in 16 against 1 in 152 for cold applications | Moderate (vendor data, consistent) |
| Volunteering that uses a skill | 27% higher odds of finding work; 51% for people without a high school diploma | Moderate (large, correlational) |
| Training and courses | Weak in the first year; pays off in the medium to long run | Strong (meta-analysis), slow |
A week that leaves something behind
The pattern that follows from the evidence is a weekly mix with five parts. It is deliberately small. The point is that each week ends with something new you can show or say.
- One piece of real work in your direction, an hour or an afternoon, that ends in a thing: a page, a model, a plan, a sample sent to someone who hires for it.
- One AI rep inside that work, not as a separate course. Let a tool do the first draft, then note what it did and what you changed. Two thirds of leaders now expect this.
- One or two conversations with people who do the work, each with one specific question. This is where referrals come from.
- A fixed block of applications with a count and a cut-off, using this week's evidence where a post asks for it.
- One line at the end of the week about what the work showed, so it's ready for the next application.
What to make
Pick the kind of thing a person in your target work would recognise. If you're aiming at operations, a tracker for a real kind of business. If it's analysis, a short answer to one question from public data, with one claim checked by hand. If it's writing or marketing, a sample for the reader they actually have. If it's building things, a one-page tool that does one job. AI builders mean most of these take an afternoon now, and the decisions you make about what the tool produced are part of the evidence.
What to avoid
- A long course with no project at the end. The certificate is a claim; the project is evidence.
- An open-ended search that fills every day. Fixed blocks protect both the search and you.
- Starting something that needs money or months before it shows anything. Make the small version first and show it to three people.
- Hiding the gap. Nearly one in four CVs now shows a gap of a year or more. The question a hirer asks is what you did with it.
On OffLadder, Try turns a direction into one small piece of real work with the tool to open and the text to paste, then keeps what you made as evidence. Your first read is free and needs no account.
Questions people ask
- Should I do a course while I'm out of work?
- Only if it ends in something you can show, and only alongside the search rather than instead of it. Card, Kluve and Weber's meta-analysis of over 200 programme evaluations found job-search help works fastest, while training looks weak in the first year and pays off in the medium to long run. A course with a project at the end is worth more than a certificate on its own.
- Does volunteering help you get a job?
- It's associated with it. A US federal study of over 70,000 job seekers across ten years found that unemployed people who volunteered had 27% higher odds of finding work, rising to 51% for people without a high school diploma. The authors couldn't prove cause, but volunteering that uses a skill also gives you a recent reference and a piece of work.
- How much time should I spend applying?
- A fixed block, not the whole week. The Michigan JOBS trials showed a structured search with a routine beats an open-ended one, and that it also protected people's mental health two and a half years on. Referrals convert far better than cold applications, so some of the time belongs to conversations.
- Do referrals really matter?
- Yes. In one study of about four million candidates, referred candidates were hired at about 1 in 16 against 1 in 152 for people applying through a careers site. On one recent hiring platform referrals were about 1% of applications and about 18% of hires. One conversation a week with someone who does the work is how referrals start.
- What should I make if I don't know what to build?
- Something a person in your target work would use. A tracker for a kind of business you know, a short analysis of public data in your field, a page that does one thing, a sample written for the reader they have. It should take an afternoon, not a month, and it should leave a link or a file.
Exercises are practice, not evidence of career fit or a qualification. How to read our guides and report a correction.
Sources
- The JOBS Project for the Unemployed · Michigan Prevention Research Center, University of Michigan
- What Works? A Meta Analysis of Recent Active Labor Market Program Evaluations · Card, Kluve and Weber, NBER
- Volunteering as a Pathway to Employment · AmeriCorps (Corporation for National and Community Service)
- 2024 Work Trend Index: AI at work is here. Now comes the hard part · Microsoft and LinkedIn
- 2025 Global AI Jobs Barometer · PwC
How this page is written
Sources support the factual background, not a guarantee that an exercise predicts career fit. Fictional practice material is labelled. Dates record content updates; no independent expert review is implied. More about OffLadder.