Why Am I Not Getting Interviews? Start With Requirement Coverage
Requirement coverage is how much of a job's actual requirements your resume shows. How to count yours in twenty minutes, and what to do with the number.
By Alexander Hoff · Founder
If you've sent somewhere between five and fifteen applications and heard nothing back, it almost always comes down to one of two things. Either you haven't sent enough applications yet to know anything, or your resume doesn't show enough of what the job actually requires. I call the second one requirement coverage, and you can measure it yourself in about twenty minutes.
One thing to clear up first, because if you search the term you'll land somewhere else. In software engineering, "requirements coverage" means how many of a system's documented requirements have a test case tied to them. Jama Connect and Perforce Helix ALM report it that way, and IBM DOORS builds it on the same requirement-to-test links. That meaning has nothing to do with job hunting. In a job search, requirement coverage means this: of everything a role requires, how much of it does your resume give a reader real evidence for?
What is requirement coverage?
Think of it as a simple ratio. On the bottom, everything the role requires. On top, the requirements your resume actually proves, where "proves" means a reader could point to a specific line on the page. It has to be a line they can see, not something you'd explain if asked.
The part most people miss is the bottom number. A job posting lists the requirements the company thought to write down, usually the must-haves and maybe a nice-to-have or two. It doesn't list the ones they didn't think to write down, and there are always some. A team that has lost two managers in eighteen months needs someone who can keep people from leaving. That never shows up in the posting, but it's a requirement, and the only way to find it is to research the company.
In a sample of 47 kits we've serviced, the median posting was about 890 words long and wrote down 18 requirements. The company research behind each one turned up, on average, another 8 that the posting never mentioned. So the real list is around 26 items, and roughly a third of it isn't on the page. Against that full list, the median resume left 8 items with nothing to back them up.
Nobody covers all 26, and you don't need to. The reason to count is that you end up with a short, specific list of what's missing instead of a vague sense that your resume isn't working. Some of what's missing is work you actually did and never wrote down. Some of it is work you haven't done. Those two problems have completely different fixes, and you can't tell which one you have until you count.
I've sent six applications and heard nothing. Does that mean something is wrong?
Probably not yet. Before you change anything, do the math on your own numbers.
One of our users, a director-level operations leader several months into a search after a layoff, had built nine applications through Telosi over the summer, and none of the nine had turned into an interview. She assumed the gap on her resume was the reason. But only six of the nine were old enough for anyone to have responded, and her interview rate across the rest of her search had been around one in ten. If you run that out, the chance of six applications producing zero interviews is about 53 percent. Getting nothing back on those six was the single most likely outcome, even if every one of them was fine.
That doesn't prove her applications were good. It means six of them can't tell you much either way. If you go looking for a problem in six data points, you'll find one, whether or not it's real.
What the nine did show was a targeting problem. They were spread across at least six different functions and nine different industries. No two roles were alike, and each application led with a different version of who she was. Nine applications is already a small number to learn from. Nine applications that each pitch a different person can't teach you anything at all.
The fix is boring. Pick one kind of role and one way of describing yourself, send the next ten applications that way, and then look at what came back. Ten applications with the same pitch will tell you something. Ten applications with ten different pitches won't.
It hadn't always been like this for her. Early in the search, applying fast had "worked really well," she said, "a bunch of interviews for the exact types of roles that I wanted," and then "that just stopped working." She had a name for what came after: "It's the black hole effect."
For what it's worth, the two extremes we see are one job seeker at around 220 applications and nothing to show for it, and another at 40 applications with five or six interviews and some final rounds. Two people aren't a study. But the one with five times the applications wasn't the one getting interviews.
How do I count my own requirement coverage?
You need the posting, your current resume, and a blank document. Budget twenty minutes once you've done it a couple of times, and closer to an hour the first time. Keep in mind that by hand you can only count the requirements that are written down. The unwritten ones take research, which I'll come back to.
- Paste the posting into the blank document and put each requirement on its own numbered line.
- Don't skip the "nice to have" list or the responsibilities section. Responsibilities are requirements that nobody labeled as requirements.
- Split anything compound. "8+ years in B2B SaaS and 3+ years managing managers" is two separate items, because you can have one without the other. Most postings come out somewhere around eighteen. If you're well under that, you probably skipped a section.
- Copy the bullet exactly as it appears. If no bullet says it, write "none."
- Resist the urge to paraphrase your own resume to make something fit. If you have to explain the connection, a recruiter reading it cold won't see it either.
- That number, over the total, is your gap for this posting.
- It will come out lower than the 8-of-26 figure above, because 26 includes the unwritten requirements you can't see yet. If you have a handful of "none" lines, the posting is probably worth an evening. If half the list is "none," it probably isn't.
- Pile one: things you actually did that your resume doesn't mention. This is the good pile. It's a writing problem.
- Pile two: things you haven't done. No amount of rewording fixes these.
- Set aside anything that's a hard number or a credential, like years of experience, a degree, or a certification. Those get filtered on directly, and they get their own treatment below.
- If most of the gap is in pile one, rewrite those bullets so each one uses the posting's own words for the requirement, then apply. Tailoring a resume without sounding generic covers how to do that without echoing the job description back at the reader.
- If you miss a hard floor by more than a year, or most of the gap is in pile two, don't apply. Spend the evening on a posting where you clear the floor.
- If it's borderline, apply, and write down which requirement you think will sink it. After ten of those notes you'll start to see a pattern.
Do this for three postings before you change anything. The requirements that keep coming up missing are the ones to act on. One miss on one posting is noise.
This is also the part of the process Telosi automates. Step 2, the Requirements Analysis, pulls the requirements out of the posting and adds the unwritten ones the company research surfaces. Step 3, the Experience Mapping, lines your resume up against that full list and tells you what nothing on the page answers. Both run inside an Application Kit.
Which missing requirements actually cost you the interview?
They're not all equal. Two kinds do most of the damage.
Years of experience, a required degree or certification, a size threshold like "managed a team of 20+", a specific system you must have run. In the nine applications above, six of the postings had an explicit years-of-experience requirement, anywhere from 8 to 15 years and most of them 12 or more, and her resume said "9+ years." The thing is, she actually had 11. So on the postings asking for 12, a screener saw a three-year gap when the real gap was one year.
She pushed back on this, and her objection is reasonable: "You shouldn't just go based off of years of experience, because they literally make that up from a template." She's right that the number is often arbitrary. It still gets used as a filter, because the recruiter running the search uses the field whether or not the hiring manager thought hard about it.
My own version of this: in my last search, the feedback I got more than once was that I didn't have direct AI experience. There was no keyword I could have added to fix that. What they were screening for was a combination of seniority and a specific domain, and I was close to it but not in it.
The cover letter uses the vocabulary of someone who has done the work, and the resume underneath it has no line showing they did. Her letters talked fluently about retention mechanics and sector economics, on top of a resume with no bullet about owning any of it.
A former director of talent operations and HR technology at a roughly 3,000-person software company described how the first pass actually works: "They don't look at the full resume. I think they look at those key parts at the beginning." If your strongest evidence is on page two, it might as well not be there.
If it isn't keyword screening, what is actually rejecting my applications?
People assume the applicant tracking system is rejecting them for missing keywords. When we looked at where her keyword coverage was low, the underlying experience was missing too. The wording was never the problem. What actually rejects an application, based on what people on the hiring side say when you ask them, is one of these.
- Knockout questions on the application form. Work authorization, location, salary expectations, years of experience. You answer these yourself in a structured field, and a disqualifying answer ends the application before a human sees it. This is the one place where a number really does auto-reject you, and it's a number you typed.
- Not showing up in the search. A recruiter filters the pipeline on a few structured fields (title, skills, location, years of experience) and then works down the list from the top. If you never make it onto the list, you get the same silence you'd get from a rejection.
- Ranked screening on high-volume roles. The same talent operations director, who ran the AI screening tools at his company, was clear about where they get used: "It works pretty okay with the higher volume positions, but if you were some VP, we certainly would not have that conversation."
- A human clearing the queue. A recruiter going through a stack of near-identical applications, quickly. A hiring manager at a media company put it this way: "I get the impression from our HR team... that they do a lot to filter through the slop. And that's not an easy part of their job."
All four of those reward the same thing, which is being obviously qualified in the first few lines of the resume, in the words the posting used. A co-founder of one of the major applicant tracking systems, as relayed to us secondhand, put the buyer's problem this way: his customers have plenty of applicants. What they can't tell is who is actually qualified, and who actually wants this particular job. Coverage answers the first question. What to research about a company is where the second one starts.
What else do people ask about requirement coverage?
What coverage number should I be aiming for?
I don't have a number for you, and I'd be suspicious of anyone who does, because nobody has measured it. What the count is good for is seeing the shape of the gap. Eight missing items that are all nice-to-haves is a very different situation from three missing items where one is a hard floor. For the two or three companies a year you really want, preparing an application for a company you actually want deserves real time. For everything else, the twenty-minute count is enough.
What if the problem is seniority, not coverage?
Then coverage won't fix it, and it helps to know that early. Going from Associate Director to Director is one step, and that's normal. Three of the nine applications above skipped two steps, either by skipping a title or by jumping in scope, like running a team of eight and applying to run an organization of sixty. A jump that size almost always needs a referral to get a look. A quick way to check where you stand is to find three people who currently hold the title at similar companies and look at what they were doing right before.
Does this work for a career change, where half the requirements are genuinely uncovered?
Yes, but you read the result differently. A big gap is normal for a career change, so the question becomes which requirements are inside the gap. Missing a bunch of nice-to-haves is survivable. Missing the one requirement the whole role is built around is not, no matter how good the letter is. Writing a resume for a career change goes through how to sort your evidence for that.
Should I add keywords for requirements I don't really cover?
No. A filter won't catch it, but a person will. Padding usually sails through the screen and then falls apart in the first conversation, when someone asks about the thing and there's nothing behind it. At that point you've lost their trust in the other 25 lines too. Leaving a "none" line alone costs you one requirement.
If I get coverage right, will I get the interview?
It gets you read, which is the part you control. After that it comes down to things you can't see: who else applied, what the hiring manager decided the role really needs this quarter, whether the position survives the next budget meeting. Plenty of fully covered applications lose to one of those, and the candidate never finds out which. Coverage gets you past the first screen. The research you do on top of it is what makes you the candidate they remember.