Job Search Strategies Ranked by Interview and Hire Rate

If you want to know where your job search hours buy the most, the hiring data gives a fairly clear order. An internal move comes first, if you have one available. A referral from someone who knows your work comes second. Introductions by agencies and recruiters come third. A tailored application on the company's own career site beats the same application through a big job board, and a sponsored job ad converts worst of all.

The gaps are large, but they are smaller than the popular numbers suggest. You may have read that a referral from a friend gives you a 40% to 60% chance of getting hired. The 40% is real, and it is the chance of an interview. Across the seven datasets below, the chance that a referred application ends in a hire sits between about 2% and 8%. That is several times better than a cold application, and it is still not a coin toss.

I have spent twenty years building software and a good part of them hiring the people who build it. I read hiring funnels the way I read system metrics: first what the number counts, then what it says. So this is a report more than an essay. For every strategy you get two numbers, the chance of an interview and the chance of being hired, each as a range across the sources that measured it. Then comes where I would put the effort.

Job search strategies ranked by interview and hire rate

Every figure is per application, for one specific job. The numbers in the last column refer to the source list at the end of the post.

What each range is made of

Internal move. Ashby [1] measured 42% of internal applicants reaching an interview and 32% of those reaching an offer, about 13% end to end. Its offer rate for internal applications rose to 17.9% by the end of 2024. Jobvite's counts [7] give 10.2% hired per internal application, and Gem [3] reports internal mobility converting at 32 times the rate of inbound applications, which works out to about 9%.

Employee referral. Interview: 40% of referred applicants reached an interview in Ashby's 38 million applications [1]. Ashby's 2026 benchmark [2] found 37% for same-function referrals and 26% for cross-function referrals in business roles, and 52% of referred candidates passing initial screens, against 35% overall. Hire: Greenhouse [4] implies at least 2%. CareerPlug's 2025 report [5] and Gem [3] both work out to about 3.1%. CareerPlug's 2026 franchise data [6] gives about 5%. Ashby recorded a 7.3% offer rate at the end of 2024 [1], and Jobvite's counts [7] give 7.8%.

Agency introduction. Ashby [1]: 42% reached an interview and 8% of those an offer, about 3.4%. Jobvite [7]: 3.9% hired per agency application.

Sourced by a recruiter or the employer. Ashby [1]: 25% reached an interview and 6% of those an offer, about 1.5%. Greenhouse [4]: recruiter-sourced candidates were 2.9% of applications and 9.7% of hires, about 1.6%. Gem [3]: sourced candidates were "nearly eight times more likely to be hired than inbound applicants," about 2.2%.

Postings shared directly by the employer. CareerPlug [5] found applicants from sources an employer shares by hand, such as industry, local and niche job boards, 14 times as likely to be hired as job-board applicants, about 5.5%. Jobvite's custom campaigns [7], targeted email and social campaigns run by employers, hired at 7.6%, on a small sample of 112 hires.

Going back to a former employer or client, or an earlier hiring conversation. This row is my estimate from experience, not a number from any dataset, because none of the sources measures this route per application. I put it this high for the same reason referrals work: the people you go back to already know your work, so they do not have to guess it from a CV.

Company career site. Greenhouse [4]: company job boards brought 23.8% of applications and 34.4% of hires, about 0.7%. CareerPlug [5]: 13.1% of applications and 26.8% of hires, about 1.1%. Jobvite [7]: 1.2%. CareerPlug's 2026 franchise report [6] found career site candidates hired "at nearly twice the rate of job board applicants."

General job board. Greenhouse [4]: external job boards brought 49.9% of applications and 22.6% of hires, about 0.2%. CareerPlug [5]: about 0.4%. Jobvite [7]: 0.5%.

Sponsored job ad. CareerPlug [5] is the only source: 23.1% of applications and 9.3% of hires, about 0.2%.

Baseline and average. Ashby [1]: 3% of unsolicited applications reached an interview, and 0.2% reached an offer at the end of 2024, down from 0.7% in 2021. Gem [3] puts inbound at about 90% of applications and about half of hires, around 0.3%. The average across all channels: Greenhouse [4] counted 203 applications per hire in 2025, Gem [3] "roughly one hire per 200 applications," CareerPlug [5] 180 in 2024, and CareerPlug's franchise data [6] 268,000 interviews and 70,500 hires from 6.2 million applicants, about 4.3% and 1.1%, in hourly service jobs.

How to read the table

Where a source publishes shares, not rates, the rate is my calculation. The formula is the average hire rate multiplied by the channel's share of hires divided by its share of applications. For referrals in CareerPlug's 2025 data: 180 applications per hire, and 2% of applications producing 11.3% of hires, so (1 ÷ 180) × (11.3 ÷ 2) ≈ 3.1%. Jobvite publishes raw counts, so its rates are hires divided by applications. Where Ashby reports two funnel steps, I multiplied them.

The interview ranges for the application routes are estimates. No source I could verify publishes interview rates by application route. CareerPlug [5] publishes one interview rate for all channels, 3%. A channel that produces twice its share of hires has some advantage before the interview and some after it. If all of it arose before the interview, the career site interview rate would be 3% × 2.05 ≈ 6.1%. Ashby's funnel [1] suggests about three quarters of the advantage arises there: referred candidates were 13 times as likely as inbound applicants to reach an interview, and less than 3 times as likely to turn that interview into an offer. That gives about 5%. Ashby's 3% for all unsolicited applications is the floor. The niche board range is the widest and the least certain.

The hire column mixes hires and offers. Ashby counts offers, which are close to hires but not the same. The other sources count hires.

The datasets describe different worlds. Ashby's customers lean toward tech. CareerPlug's are mostly small businesses, and its 2026 report covers hourly service franchises. Greenhouse covers thousands of organizations of every size. Jobvite's data is from 2016, a much tighter labour market: one in six applicants reached an interview that year. That spread is why the ranges are wide, and why I would trust the order more than any single number. Nearly all of the data is North American.

The rates describe channels, not cause and effect. Different people use different channels for different jobs. Moving an application from a job board to a career site does not automatically multiply your odds by three. CareerPlug's own reading [5] is that career site applicants are "more likely to have spent time considering their own fit for the role." The channel partly records how much thought went into the application.

Sourced by a recruiter is not a rate for messaging recruiters. It measures people an employer or recruiter chose and brought into a process. It says nothing about cold messages you send to recruiters.

How much more likely is a referral to be hired than a cold applicant?

The multiple depends on what the referral is compared with, and when. Every source puts it well above one.

The Ashby row is the one I find most telling. Between the third quarter of 2021 and the end of 2024, the offer rate per referred application fell from 12.1% to 7.3%. The rate for unsolicited applications fell from 0.7% to 0.2%. Both channels got harder. The gap between them roughly doubled.

Which job search strategies should you focus on?

Five, in this order. Each one is either higher in the table than what most people do by default, or it makes the next application better.

1. Check for an internal move before you look outside

If you are employed and thinking about leaving, look at your own company's open roles first. Every source that measures internal applications puts them at the top: 9% to 18% hired per application, several times any external channel. Some internal candidates are hired without a formal interview at all [1].

The reason is not mysterious. Your employer already has what every other employer is trying to guess from a CV: years of evidence about how you work. It will not fit every situation. If what you want to leave is the company itself, this row does not apply to you. If what you want to leave is the role, the manager or the pay grade, it probably does.

2. Get a referral from someone who knows your work

Referrals are the strongest external channel in every dataset. What makes them work is information. In three field experiments by Amanda Pallais and Emily Glassberg Sands, published by the NBER, referrals carried information about worker performance that employers could not see in the applicants' credentials. The experiments ran in an online labour market, so they explain the mechanism rather than give you a percentage.

They do explain the most useful number in the table. In Ashby's 2026 operations benchmarks, a referral from someone in the same function as the job led to a hire 5.2% of the time, against 3.1% from someone in a different function. An engineer vouching for an engineer can say something specific. A friend from finance vouching for the same engineer mostly confirms that they exist.

When I was hiring engineers, I saw the same pattern from the inside. One referral arrived with a single sentence from a senior colleague: the candidate had fixed a release process nobody had asked them to fix, in their first month. That sentence did more work than the whole CV, and we moved them straight to a technical conversation. Another referral in the same quarter came with "I do not know them well, they asked me to pass this on." Formally, both were referrals. The system counted them the same way. Only one of them carried a signal.

So when you look at a target company, start from the people. Annette Garsteck, a career coach in Cleveland, put the sequence plainly in a CNBC piece from September 2026: "Does your friend or your ex-work bestie know somebody at that organization? Can they introduce you? And can that person be the referral?" I would add one step before the ask. Ask for insight about the role and the team first. A person who has talked to you about the job can speak to your fit, and that is the referral that carries the signal.

3. Be findable by recruiters and agencies, and answer them

Agency introductions reached an interview 42% of the time in Ashby's data, the same rate as internal candidates, and 3% to 4% ended in a hire across Ashby and Jobvite. Recruiter-sourced candidates reached an interview 25% of the time and were hired in 1.5% to 2% of cases across Ashby, Greenhouse and Gem. Greenhouse calls recruiter sourcing the highest-converting channel among its top five sources. Both groups were selected by someone before the process started, which is why they convert.

You cannot apply to this channel. You can be findable in it. A LinkedIn headline that names your role and your core skills in the words recruiters search for, a profile that tells them in ten seconds what you do, and a reply to every relevant recruiter message within a day. If you work in a field where specialist agencies place people, one or two good relationships with them are worth more than a hundred applications.

4. Apply on the company's career site with a tailored CV

For the cold applications you still send, and you will send many, the route matters. Three independent sources put the career site at about 0.7% to 1.2% hired per application and general job boards at about 0.2% to 0.5%. In each of the three, the career site converts at two to three times the job board rate.

Tailoring belongs to the same move. I know of no large public dataset that isolates its effect on hires, so I will not give you a number. The logic does not need one. At 200 applications per hire, a reviewer reads a CV that uses the language of the job description and leads with the experience it asks for further than a CV that does not. A generic CV tends to be read until the first line that does not fit. People who tailor also tend to apply to roles they match, and choosing those roles is half of what tailoring is. For the matching part, see how many jobs you should apply for.

5. Look for the postings employers share directly

CareerPlug's multiple for employer-shared sources, 14 times the job-board hire rate, is the largest in its whole report, larger than the one for referrals. Jobvite's targeted campaigns point the same way. The mechanism is probably audience, not software: an employer who shares a posting on an industry board, a local board or a professional community is fishing where the right people already are, and the people there are fewer. The jobs and the applicants differ across these sites, so treat this as a reason to look beyond the largest boards, not as a promise that switching sites multiplies your odds by fourteen.

What should you do with the big job boards?

Keep using them, for what they are good at. General job boards convert worst per application, yet they still produce a large share of all hires: 43% in CareerPlug's 2025 data, and 57% for company and external boards together at Greenhouse. Low rate, huge volume.

The volume is also the problem. Greenhouse counted 244 applications per job in 2025, up from 116 in 2022. Daniel Chait, CEO of Greenhouse, described it to Fortune in July 2026: "They'll often get hundreds or thousands of applications in just a day or two, and they're not able to keep up with it." A job board is an excellent place to discover that a role exists. It is a crowded place to be read.

Sponsored job ads deserve a separate line, because they convert the worst of all: 23% of applications and 9% of hires in CareerPlug's data, about 0.2% per application. A promoted listing is optimised to collect many applicants. That is the employer's goal, not yours.

Which popular strategies have no reliable numbers?

Some of the most repeated advice cannot be ranked, because nobody has published a credible rate for it.

Cold messages to hiring managers, networking events and speculative CVs. I found no general, per-contact hire rate for any of them. They can be sensible ways to discover roles or start relationships. Anyone who gives you a precise success rate for them is inventing it.

"85% of jobs are filled through networking." This comes from a 2016 LinkedIn post by recruiter Lou Adler, a survey of about 3,000 people, most of them in staff or management roles, about how they found their most recent job. It refers to "critical jobs" filled "via networking of some sort" and never defines networking. It is a self-reported description of a route, not a conversion rate. The tracking data measures something narrower and gets a smaller number: referrals were 11.3% of hires at CareerPlug and 11.6% at Greenhouse.

"Spend 70% to 80% of your time networking." A reasonable heuristic, and I lean the same way. It is not a finding: none of these datasets measures hours.

What does have strong evidence is that a wider professional network helps over time. In a randomized experiment on LinkedIn involving more than 20 million people over five years, during which 600,000 new jobs were created, weaker professional ties increased job mobility, with effects that varied by industry. The study gives no rate per message. It does tell you that the acquaintances you add today are part of where next year's referrals come from.

How to put the ranking to work this week

1. Pick roles that fit. Every rate in the table assumes the application is plausible. No channel rescues a role you do not match.
2. For each target company, check the people first. Anyone you know there, anyone who knows someone there, ideally in the same function as the role. Ask about the role before you ask for a referral.
3. Be findable. Update your LinkedIn headline and profile for the searches recruiters run in your field, and answer relevant recruiter and agency messages quickly.
4. Apply directly, tailored. Use the big boards for discovery, then apply through the company's career site with a CV adjusted to the requirements you meet.
5. Record the channel for every application, and look at interviews and offers by channel, not just applications sent. After a few dozen applications, your own numbers start saying things the averages cannot. If one channel produces interviews and another produces silence, the silence has causes you can diagnose.

Where these channels fit inside the full search, together with target definition, proof and interview preparation, is covered in how to land a job efficiently.

The ranking is a map of where people already vouch for you

Read the table from the top and a pattern appears. The channels that convert best are the ones where someone has already put their judgment on the line for you: your current employer, a colleague, an agency, a recruiter who picked you. The channels that convert worst are the ones where nobody has. The practical strategy follows from that. Spend your time where someone can vouch for you, and make every cold application good enough to be worth the long odds.

That order is how HintCraft is built to be used. Your target companies, your contacts with the company each one works at and what they can help with, and your applications with the CV version you sent sit in one system. Before you open an application form, you can check who in your contacts knows that company.

Sources

1. Ashby, Referrals, Talent Trends Report, May 2025. 38 million applications, 93,000 jobs, 2021 to 2024.
2. Ashby, Recruiting Operations Benchmarks, 2026.
3. Gem, Key takeaways from the 2026 Recruiting Benchmarks Report, December 2025. 165 million applications, 1.2 million hires, June 2021 to May 2025.
4. Greenhouse, The Hire Standard: Greenhouse benchmark report, North America, March 2026. 640 million applications, 2022 to 2025.
5. CareerPlug, Recruiting Metrics Report, April 2025. More than 10 million applications at more than 60,000 small businesses, 2024.
6. CareerPlug, 2026 Service Franchise Hiring Report, June 2026. 7.5 million applicants and nearly 90,000 hires at about 46,000 franchise locations.
7. Jobvite, 2017 Recruiting Funnel Benchmark Report, 2017. Hires and applications by source, 2016 data.
8. Pinpoint, Why referrals are 7x more likely to be hired, 2023. 4.5 million applications; method not published.
9. Amanda Pallais and Emily Glassberg Sands, Why the Referential Treatment: Evidence from Field Experiments on Referrals, NBER Working Paper 21357, 2015.
10. Stanford Digital Economy Lab, A causal test of the strength of weak ties, summary of the study published in Science, 2022.
11. Lou Adler, New survey reveals 85% of all jobs are filled via networking, February 2016.