You've sent hundreds of applications since the layoff. You're an experienced engineer, most of the roles looked like a fit, and you have a handful of interviews to show for it. So you read the silence for clues. Is it the résumé? The gap? The market?
The silence can't tell you. No reply looks the same whichever cause is behind it. Five separate things could be behind it: the roles you target, what your experience shows for those roles, how your résumé shows it, the route your application takes, and the market. Four of them you can test, and the fifth you can only account for.
How to figure out what's going wrong when most applications get no reply
If you've applied to hundreds of roles and reached only a handful of interviews, you can't work out what's going wrong from the silence itself. Most applications get no reply, including ones from people who fit. Sort what you've sent by kind of role and by how it reached the company, and look at where the interviews you did get came from. Then change one thing at a time: which roles you target, what a person who hires sees in your résumé against one posting, and whether the application arrives through a person. Keep applying while you do it.
That method is my own reasoning. No study tested it, and I can't promise that any of these tests produces an interview. They point you in a direction, which is more than the silence does.
I'm also not going to tell you whether your numbers are good or bad. I don't know of a study that measures how often experienced software engineers hear back, so there's no benchmark here for your own rate. The rates that do get published come from other groups, and they shift with who was studied and what counts as a response.
Ashby, a hiring-software company reporting its own customers' data from January 2021 through March 2026, says approximately 7 to 8% of applications resulted in an interview in 2021 and between 3.6% and 4.7% do now, depending on role type. That counts applications, not applicants. Ashby doesn't say who those customers are, so it isn't a market rate.
In a U.S. experiment run more than a decade ago, researchers sent made-up résumés to real postings, and applications from experienced, college-educated women to administrative jobs got a callback 10.4% of the time. A callback there meant any response, with or without an interview request. Another experiment, run around the start of the pandemic, sent more than 83,000 fictitious applications to entry-level jobs at 108 of the largest U.S. employers, and 24% of them were contacted within 30 days. Entry-level jobs at very large employers are a long way from experienced engineering roles, and "contacted" meant any contact attempt.
None of those is a benchmark for you. What they share is that most applications led nowhere in every one of those settings. Silence is the ordinary result, so it can't diagnose anything.
Sort what you've already sent
You already have data. Take the list of what you've sent and sort it three ways: by kind of role, by how the application reached the company, and by whether the posting stated a hard requirement you don't meet.
Kind of role means title and level. How it reached the company means cold through a job board or careers page, through a recruiter who contacted you, or through a person you know there. A hard requirement is something the posting calls required and you can't claim, such as a location, a clearance, or years in a specific stack.
Then look at where the handful of interviews came from. If they all came from one kind of role, or none came from cold applications, that's a lead. If a large share of what you sent was ruled out by a stated requirement, those silences are explained, and they say nothing about the rest.
With a handful of interviews, a cluster is a lead and not proof. It tells you which test to run first.
It's also why I prefer tests whose result you can observe over comparing reply rates between two batches of applications. When replies are this rare, one batch drawing a reply and the other drawing none proves little. A hiring manager telling you what they concluded from your résumé is something you can see. So is a reply from a named person.
Test which roles you're targeting
There are two ways targeting goes wrong. You can aim too narrowly, at one title in one kind of company. Or you can aim at roles your last two jobs don't obviously lead to, where a stranger reading your résumé has to guess why you're applying.
The sort shows you which one to suspect. If almost everything you sent is one title, add the adjacent roles your recent work does lead to and keep sending the original ones. If a lot of it sits a step away from your last two jobs, put one of those postings next to your résumé and ask whether someone who's never met you would see the connection.
What you'd see is an interview, or a recruiter's reply, from a kind of role you weren't applying to before. That would tell you your targeting was part of the problem. Hearing nothing from the new roles wouldn't clear your targeting, because hearing nothing is the ordinary result.
The research here is thin and it isn't about engineers. In a small experiment run more than a decade ago with unemployed job seekers in Edinburgh, showing people alternative occupations broadened the set of jobs they considered and increased their interviews, mainly for those who otherwise searched narrowly and had been unemployed for a few months. It doesn't promise that widening works for you. It's a reason to test targeting before you rewrite the résumé again.
Test what a hiring reader sees in your résumé
If you're qualified for the roles you apply to and keep getting no response, you can't see from your side what hiring teams aren't seeing. A résumé and a job description don't show a company's private screening criteria, how strong the other applicants were, or why one employer didn't reply. What you can find out is what a person who hires for that role concludes from your résumé against one posting. Ask one to read both and tell you what they'd take from it. In my view, that read is how you separate experience that lacks what the posting asks for from experience that has it and doesn't show it. And no response isn't proof that your résumé is the problem, because employers who use screening software say themselves that it screens out qualified people.
That comes from a Harvard Business School and Accenture survey of employers in the United States, the United Kingdom, and Germany, run just before the pandemic: 63% said they used screening software, and of those, 88% believed qualified high-skills candidates were screened out because they didn't match the exact criteria in the job description. That's a share of employers giving their own view. It says nothing about how many résumés any system rejects.
For the read itself, pick one posting you applied to and were sure you fit. Give the posting and your résumé to someone who hires for that kind of role. Ask what they'd conclude about you and what they'd assume you haven't done.
You'll get one of two answers. The first is that the experience the posting asks for isn't there. That's a targeting problem or a missing-evidence problem, and no rewrite fixes it. The second is that the experience is there and the page hides it, buried low on the page or described in terms the posting doesn't use. A rewrite does fix that one, and MentorCruise's guide to what a software engineer résumé needs to get interviews goes through the document itself.
One read gives you one hiring person's conclusion about one posting. It doesn't tell you what any company's screen did with your application. It's still more than the silence has told you.
Test the route your application takes
For some of your next applications, go through a person at the company. A former colleague who works there now is the obvious one. Ask them to refer you or to pass your résumé to the person hiring, and keep sending cold applications alongside.
What you'd see is whether a person replies: the colleague, then a recruiter or the hiring manager. If people respond when a person asks them to look, the route was part of the problem. If your colleague reads the posting and tells you it isn't a match, you've learned something about your targeting instead.
The evidence supports trying this and no more than that. At one U.S. financial-services firm, referred applicants were 6% of applicants and 21% of interviewees in hiring records from more than a decade ago, according to a 2012 Federal Reserve Bank of New York staff report on that firm. The authors make no causal claim, and they can't rule out that referred applicants differ from the rest in ways hiring records don't capture. A 2015 study of several large firms, one of them high-tech, also found referred applicants were more likely to be hired than non-referred ones. That finding is about being hired and not about being interviewed, and it's an association too.
What the market explains and what it doesn't
The market explains part of this. In mid-June, U.S. software development job postings on Indeed were about 27.5% below their pre-pandemic level, while postings overall were essentially level with February 2020, according to Indeed's Hiring Lab. Indeed also reports that 71% of the increase in software development postings between May 2025 and May 2026 came from senior roles. These are postings on one job site, which aren't hires, and "senior" is Indeed's own label.
The unemployment figures are less dramatic than the postings. In 2025, unemployment in computer and mathematical occupations averaged 3.3%, up from 2.8% the year before, in the U.S. Bureau of Labor Statistics' annual averages. That group is wider than software developers, and that year's average leaves out October, so the two years aren't strictly comparable.
So for an experienced engineer, the market is part of the answer and not all of it. There are fewer software postings than before the pandemic, the recovery is mostly in senior roles, and unemployment in computing occupations rose but stayed low. That's my reading of two limited measures. The market is also the one cause on the list you can't test, which is another reason to spend your effort on the other four.
What you can't change
You can't change how long you've been out of work, and the research on whether it counts against you is mixed. In a U.S. experiment with early-career fictitious applicants for sales, customer service, and clerical jobs, written up in a 2012 working paper, interview callbacks fell as the stated unemployment spell got longer, with most of the decline in the first eight months. The experiment with experienced women applying to administrative jobs found no relationship between callbacks and unemployment of up to a year. It did find that applicants aged 50 and older were significantly less likely to get a callback. Neither experiment covered software roles.
I'm not telling you the gap makes no difference. The evidence disagrees with itself, the gap is already on the page, and you can't shorten it. You can't change your age either. So neither is the first thing I'd try to fix. The roles, the résumé, and the route are yours to change.
Keep applying while you test
Testing isn't a reason to stop applying. In a study of Swiss unemployment-insurance records from over a decade ago, among people required to send far fewer applications a month than you're sending, one more application a month shortened the spell of unemployment by 4%, and the gains were smaller for skilled job seekers and for people who were already applying more.
That's a long way from your situation. I don't know of anything that measures what the next hundred identical applications return to someone who's already sent hundreds. So keep the volume up and let the tests change what goes out.
Where a mentor fits and where you don't need one
The résumé read needs a person who hires for the role, and you may already know one. A former colleague who hires for that role can do the read as well as a paid mentor can. Ask them first. A mentor is worth it when you have nobody like that to ask.
On MentorCruise, a résumé coach works with you over time, giving feedback on your applications and adjusting the approach based on results, where a résumé writing service delivers a finished document. You can browse résumé coaches and compare their experience and client feedback yourself. I have nothing that shows a mentor or a coach raises anyone's interview rate, and I'm not claiming one does.
If you don't have a hiring person to ask, browse and compare mentors on MentorCruise. The profile detail that matters for this is whether they've hired for the kind of role you're targeting.
