More than half of unemployed Americans who looked for work in 2025 sent out resumes or filled out applications, while fewer than 1 in 4 contacted friends or relatives, according to the US Bureau of Labor Statistics.¹ Yet in LinkedIn experiments covering more than 20 million members, the new connections most likely to lead to a job move were acquaintances, not close contacts.²

How to start a job search, in one sentence: pick a target, write a weekly plan, and work applications and people side by side while you track what happens. Here are 7 steps, ordered for a first month:

1. Pick a target, then widen it one notch
2. Write a weekly plan with numbers in it
3. Tell people you’re looking, from close friends to acquaintances
4. Rewrite your resume and profile for the target
5. Practice interviews before you need them
6. Log every application and review the week
7. Keep a steady pace instead of sprinting

Job search programs work when they build skills and motivation together

A 2014 meta-analysis of 47 job search programs found that those combining practical skills with motivation, such as goal setting and enlisting support, raised employment, while programs that offered only one showed no clear effect. The programs were training courses for unemployed people, such as job clubs and job search workshops.³ Their ingredients can still be borrowed for a search you run yourself.

The study

Moderate evidence

Inside the 2014 review of 47 job search programs

Researchers Songqi Liu, Jason Huang and Mo Wang pooled 47 experimental and quasi-experimental evaluations covering 9,575 job seekers, counting who moved from unemployment into full-time work. Overall, participants were more likely to find work than comparison groups. Programs that paired search skills, such as finding leads and presenting yourself in resumes and interviews, with motivation, such as confidence building, goal setting and support from family and peers, raised employment; skills-only and motivation-only programs showed no clear difference from controls.³

The review reports its results as odds ratios, not employment rates, so the gain in plain percentages is unknown, and results varied widely from program to program. Only 6 skills-only and 9 motivation-only programs were available, so the case for combining the two is weaker than the case that programs help at all.³

The programs helped younger and older job seekers more than middle-aged ones, and people out of work for less than six months more than the long-term unemployed.³ Economists who study government programs are more cautious: a 2024 review of job training and job search assistance in developing countries notes that the first generation of evaluations found typically small impacts, and that in many cases the binding constraint may be a lack of job openings rather than a lack of skilled workers.⁴ The closest test of a do-it-yourself version is the plan-making trial in step 2 below.

The first month: how to start a job search in 7 steps

Start with a target and a written weekly plan, then work applications and people side by side, tailor your resume, practice interviews and review your results every week. Randomized experiments bear on each of the first three steps, though none tested a step exactly as written here: one in Edinburgh on widening the search, one in South Africa on plan-making and one on LinkedIn on which new connections led to job moves.

1. Pick a target, then widen it one notch

Write down the job titles you want, the level, and where and how you want to work, then add one related occupation that uses similar skills. In a 2013-14 experiment with 300 unemployed people in Edinburgh who searched for jobs in 12 weekly sessions, half were offered a search tool that suggested related occupations based on labor market data. They searched more broadly and reported about 44 percent more job interviews than the control group, from a starting average of about 0.6 interviews a week.⁵

That result was only borderline statistically significant, came mainly from people who had been searching narrowly, and did not show up as more people finding jobs, though the study was too small to detect that reliably.⁵ A later randomized trial by the same economists, with long-term unemployed job seekers and published in 2025, found positive effects on finding a stable job, larger for people out of work longer.⁶ In the 2014 meta-analysis, programs that encouraged proactive moves, such as widening the range of jobs considered and contacting employers directly, were more effective than programs that did not.³

2. Write a weekly plan with numbers in it

Start from the time you actually have in a typical week, then turn the target into a schedule: which days you will search, where and how, plus weekly goals for hours, openings found and applications sent. That is the template tested in a 2015 field experiment with 1,097 unemployed people aged 18 to 35 in Gauteng, South Africa, in which one group filled it in at a government job search workshop.⁷

The plan-makers sent about 15 percent more applications than people who attended the workshop alone, without spending more hours searching. Across follow-ups 5 and 12 weeks later, an average of 11.5 percent of a control group had a job, and the plan-makers’ employment rate was an estimated 4.9 percentage points higher. Their lead over the workshop-only group on employment was not statistically clear, and adding weekly text reminders, or a friend or relative who was sent the plan’s goals, did not improve results. The setting, young job seekers in a labor market with high structural unemployment, may limit how far the numbers travel.⁷

Goal setting also stands out in the 2014 meta-analysis: programs that taught people to set concrete job and search goals, and to check progress on them regularly, were more effective than programs that did not.³

3. Tell people you’re looking, from close friends to acquaintances

List people who might hear of openings, from former colleagues and classmates to contacts you rarely speak to, and tell each of them what you are looking for. Among unemployed Americans in 2025, contacting friends or relatives was far less common than sending applications.¹

Bar chart with 6 bars; the largest is Sent out resumes or filled out applications at 56%.

• Sent out resumes or filled out applications56%
• Contacted an employer directly51%
• Contacted friends or relatives24%
• Used a public employment agency14%
• Placed or answered ads12%
• Used a private employment agency8%

Respondents could name more than one method, and the figures exclude people on temporary layoff; October 2025 is missing because a federal government shutdown stopped data collection.¹

Acquaintances may matter more than close friends. A 2022 study in Science analyzed experiments in which LinkedIn randomly varied the connections its “People You May Know” feature suggested to more than 20 million members between 2015 and 2019. The new connections most likely to lead to a job move, meaning the member later worked at that contact’s company, were moderately weak ties: people with some contacts in common but not many. Past about 10 shared contacts, a new tie became less likely to lead to a job move.²

1. Close ties: many contacts in common; past about 10 shared contacts, a new tie was less likely to lead to a job move
2. Moderately weak ties: some contacts in common; the most job moves came through this ring
3. Distant ties: few or no contacts in common; less likely than the middle ring to lead to a job move

Earlier large correlational studies had found that, tie for tie, close contacts were more likely to lead to a job. The randomized analysis reversed that result, but only partly: measured by contacts in common, the payoff rose as ties got weaker only up to a point, then fell. The findings describe LinkedIn users, who lean toward finance, professional services and technology compared with the US workforce, and in less digital industries strong ties did more. Two of the five authors are or were LinkedIn employees with a financial interest in Microsoft, and the study had no external funding.²

Referrals are the formal version of this route. At one US financial services firm, in a study published by the Federal Reserve Bank of New York, referred candidates made up 6 percent of about 62,000 applications from 2006 to 2010 but 29 percent of hires, while job board applicants made up 60 percent of applicants and 24 percent of hires.⁸

These are one firm’s records, and referred applicants may differ in ways the data cannot see, so they show how referrals flow through hiring rather than proving a referral causes an offer.⁸ In the 2014 meta-analysis, programs that enlisted family, friends and fellow participants for encouragement and job leads were more effective than programs that did not.³

4. Rewrite your resume and profile for the target

Rework your resume summary and recent roles around the target from step 1, leading with concrete results a hiring manager for that role would care about, and bring your professional profile in line with it. In the 2014 meta-analysis, programs that trained people to present their skills concretely and relevantly in resumes and interviews were more effective than programs that did not. The review does not compare resume formats, so it tells you that presentation matters, not which layout to use.³

5. Practice interviews before you need them

Rehearse answers to common questions out loud and, if you can, run a mock interview with someone who will give honest feedback. In the 2014 meta-analysis, programs that built job seekers’ confidence, for example through role-played interviews, were more effective than programs that did not, and gains in search skills, confidence and search activity partly explained why programs worked.³

6. Log every application and review the week

Keep a simple log with the date, role, employer, channel (application, referral, recruiter or contact), response and next step, and book a weekly slot to compare it with your plan.

1. 1Set the week's targetshours, openings, applications
2. 2Search and reach outapplications and people
3. 3Log what happenedreplies, interviews, next steps
4. 4Review and adjusttarget, channels, materials

Then repeat from “Set the week's targets”

A 2016 meta-analysis of 138 randomized studies with about 20,000 participants, mostly on health goals such as weight, physical activity and blood sugar, found that prompting people to monitor their progress helped them reach their goals, more so when progress was physically recorded or reported to others. Job searching was not among the goals studied, so treat this as a borrowed principle.⁹

7. Keep a steady pace instead of sprinting

Set a weekly volume you can sustain for months rather than a burst of applications in the first week. A 2021 meta-analysis of 378 samples covering 165,933 job seekers found that how intensively people searched was only modestly linked to interviews, offers and employment (corrected correlations of 0.14 to 0.23), and did not predict the quality of the job they found. Measures of search quality and self-regulation predicted both those outcomes and job quality, which the authors flag as promising for future research. The studies were correlational, so they show links, not causes.¹⁰

Myth

The fastest way to a new job is to send as many applications as you can.

Fact

Searching harder is only modestly linked to results. In a 2021 meta-analysis of 378 samples of job seekers, search intensity was modestly linked to interviews, offers and employment, and not linked to job quality.

Expect most applications to go nowhere, and count that as normal rather than personal. At the financial services firm in step 3, 2.9 percent of applications to positions that were filled led to an interview and 0.6 percent to a hire.⁸ Use the weekly review from step 6 to decide whether to change the volume or the approach.

WiserHours

Set up in your first week

• Target roles written down, plus one related occupation
• A weekly plan with goals for hours, openings and applications
• A list of people to tell, from close friends to acquaintances
• Resume and profile rewritten for the target
• Answers to common interview questions practiced out loud
• A log for every application and conversation
• A weekly review slot in the calendar

Find this checklist online:
https://wiserhours.com/job-search/how-to-start-a-job-search/#checklist-how-to-start-a-job-search-setup

Expect a search measured in months

In the US, the median unemployed person in August 2026 had been looking for work for about 11 weeks, and 27 percent had been looking for 27 weeks or more, according to the Bureau of Labor Statistics.¹¹ Those figures measure how long people who are still unemployed have been searching continuously, not how long completed searches took.¹²

They also leave out people who search while employed, a large group that does comparatively well. A 2022 study in Econometrica, based on a US survey fielded every year since 2013, found on-the-job search pervasive. Employed job seekers were at least three times as effective as unemployed ones, a measure of job offers per unit of search effort rather than of how fast people found work, and they received better offers.¹³

Which steps rest on the strongest evidence?

The strongest evidence in this guide is Liu, Huang and Wang’s 2014 meta-analysis of job search programs, which favors pairing skills with motivation, and LinkedIn’s randomized experiments on which new connections lead to job moves. Most individual steps rest on single trials, comparisons between programs, correlational studies or research on other kinds of goals.

The bottom line

Two moves recur across the research in this guide: a concrete target with a written weekly plan, and people, acquaintances among them, who know what you are looking for. After that, log what you do, review it every week and change the target or the approach before you add volume. If you are out of work, a structured program that builds both skills and motivation is worth asking about.