How to Structure a Resume: A Beginner's Guide to Formatting, Bullets, and ATS
Your resume isn’t getting rejected because of your experience. It’s getting rejected because a recruiter, or the software standing in for one, couldn’t find the shape of that experience fast enough to bother looking closer.
I used to sort a stack of resumes into “screen” and “pass” piles before I’d finished reading a full sentence on any of them. Not because I’m some savant. After you’ve screened resumes for a living, your eye stops reading top to bottom and starts pattern-matching on shape: where the strongest line sits, whether the dates line up, whether the page looks like eleven fonts got into a fight. Structure decided the outcome before the words got a vote.
Structure decided the outcome before the words got a vote.
The moment that stuck with me: two candidates, nearly identical experience, similar years in the field. One got a screen. One didn’t.
The difference wasn’t better sentences. It was that I could find what I needed on the first pass with one resume, and with the other, I couldn’t, so I moved on. Multiply that by a full day of screening and wording stops being the thing that saves you.
I spent eight years as a technical recruiter before I became a designer. Resumes were the raw material of that entire job: reading them, rejecting them, occasionally fighting for one a hiring manager wanted to pass on. Now I’m on the other side of the desk, building my own resume and reviewing portfolios as a working product designer.
I’ve used the exact structure in this post, the same skeleton, the same bullet logic, through every version of my own career since. It’s gotten me interviews at every level, from an entry-level design job to roles paying up to $200k a year. The wording changed every time. The shape didn’t.
Naming the limit up front: this is one recruiter’s pattern, built from hundreds of resumes I screened professionally and my own resume rebuilt more times than I’d like to admit since. It’s not a universal law of hiring. Every recruiter and every ATS instance weighs things a little differently. But the pattern held often enough, across enough different roles and companies, that I trust it more than most of what’s circulating on LinkedIn.
In this guide
I’ll walk through the skeleton every resume should share, with an interactive diagram you can hover through, how to write bullets that survive a two-minute skim, and how to format the PDF itself so an ATS or an AI screening layer doesn’t mangle it before a human ever sees your name.
The skeleton underneath every good resume
A resume skeleton is the container, not the content. It’s the answer to “where does this information live” before you’ve written a single sentence about it.
Name goes here. Dates go there, right-aligned. Skills live in a scannable row, not a paragraph.
Get the skeleton right and mediocre sentences still read fine. Get it wrong and even great accomplishments get lost in the shuffle.
Here’s the shape I mean. Hover or tap any of the nine numbered marks below to see why that specific part of the page is built the way it is.
Interactive skeleton — hover or tap any numbered part for why it goes there
Most of what a recruiter does on a first pass is closer to scanning than reading. A widely cited 2018 eye-tracking study from TheLadders found recruiters spent an average of 7.4 seconds on an initial resume review, though the exact number has been disputed. A trade publication called ERE.net dug into the study’s methodology and pointed out that the sample size and recruiter experience level were never disclosed, which is a fair critique. What the study did find, and what’s held up better than the raw timing figure, is that resumes with simple layouts, clear section headings, bold titles, and bulleted accomplishments consistently outperformed resumes with clutter: multiple columns, no white space, missing headers, dense unbroken paragraphs.
That’s the same F-pattern that governs how people read any dense block of text on a screen, not something unique to resumes. Eyes sweep the top, drop down the left edge, dip into whatever catches attention along the way. That’s exactly why the skeleton above puts your name and target title at the top, locks dates to the right margin, and puts the strongest accomplishment first in every role.
Reverse-chronological, most recent role first, is the format roughly 90% of job seekers already use, and it’s what both recruiters and ATS software are built to expect. This isn’t the place to get creative.
If you’re early in your career and the work-experience column feels thin, that’s fine too. Education, class projects, and volunteer work can legitimately carry a full page for someone a year or two out of school. Recruiters expect that shape from an entry-level resume and don’t penalize it for being thin on job titles.
Bullets that survive the skim
Your resume isn’t getting rejected because of your experience. It’s getting rejected because a recruiter, or the software standing in for one, couldn’t find the shape of that experience fast enough to bother looking closer.
Once the skeleton’s right, the actual sentences matter less than beginners think, but they’re not nothing. Two frameworks do most of the work, and they’re not competing with each other.
The first comes from Laszlo Bock, Google’s former SVP of People Operations, who laid it out in his book Work Rules! and in a LinkedIn post that’s been circulating ever since: Accomplished [X] as measured by [Y], by doing [Z]. X is the outcome, not the task. Y is a number that proves it: a percentage, a dollar figure, a count. Z is the mechanism, what you actually built or changed to get there.
The second comes from Harvard’s career services office, and it’s a sentence-construction rule rather than a content rule: start every bullet with an action verb, not a job description. “Responsible for managing client relationships” is passive and forgettable. Something like “managed a 12-client portfolio, growing retention 18 points year over year” does the job an actual sentence should do.
The two rules stack instead of compete. Action-verb-first is how the sentence starts. XYZ is what fills it in.
A bullet that satisfies both reads fast and proves something at the same time. Here’s what that looks like on paper:
| Before (a task, no proof) | After (verb-first + XYZ) |
|---|---|
| Responsible for managing the company’s social media accounts. | Grew Instagram engagement 42% in six months by shifting the posting schedule to match peak audience activity. |
| Helped resolve customer support tickets. | Resolved an average of 35 tickets a day, cutting first-response time from six hours to 45 minutes by building a canned-response library. |
| Worked on redesigning the onboarding flow. | Redesigned a subscription app’s onboarding flow, cutting drop-off at step three by 18% through simplified form fields and inline validation. |
Swap weak openers, helped, assisted, participated in, responsible for, for verbs that claim ownership: led, developed, analyzed, implemented, streamlined, reduced, generated, optimized, secured. If you didn’t do it alone, say so honestly somewhere else on the page. But say what you did, specifically, first.
Format the PDF so the robots can actually read it
An ATS, an applicant tracking system, is the software that sits between your submitted resume and the human who might eventually read it. What is ATS in practice? It’s not one thing.
Jobscan’s 2025 usage report, built by manually checking every Fortune 500 company’s live career page, found 97.8% of them run a detectable ATS. Workday is the most common at around 39%, SAP SuccessFactors is next at roughly 13%, and Greenhouse, Lever, iCIMS, Taleo, SmartRecruiters, and Ashby all show up constantly outside the Fortune 500 too.
The point isn’t which platform. It’s that parsing software is reading your resume before, or alongside, an actual person, and it reads very differently than a human does.
| Risky pattern | Why it breaks parsing |
|---|---|
| Tables or text boxes | Parsers can slice straight across the page width or skip a text box’s contents entirely, since it’s treated as a separate object layer. |
| Scanned or image-only PDF | The ATS sees a picture, not words. Parsing fails completely without OCR applied first. |
| Non-standard section headers (“My Journey”) | The parser uses header text to classify what follows. A creative header can get content miscategorized or dropped. |
| Icons instead of text | There’s no underlying text for a parser to extract from a phone icon or a skill-level bar. |
| Hyperlinked-only labels (“LinkedIn”) | The clickable URL behind the label often doesn’t transfer. Write the actual URL out as visible text. |
Two-column layouts deserve their own callout, because it’s the one I get asked about most, usually from designers. ATS parsers read linearly: left to right, top to bottom. A two-column resume can get serialized wildly out of order.
Jobscan tested a design-heavy, columned resume against Lever’s parser and it failed to extract skills, contact info, and work samples, pulling out only the work-experience section. That’s real. But I won’t pretend a two-column resume is always a mistake.
Some designers use one deliberately, as a portfolio piece as much as a resume, betting the visual craft earns more than the parsing risk costs. That’s a legitimate, calculated tradeoff for a design-forward role where a human is likely to open the file directly. It’s just not the safe default, and if you make that bet, keep a plain single-column version ready to paste into an online application form.
What’s actually documented is less dramatic and more useful. Harvard Business School and Accenture surveyed 8,720 workers and 2,275 executives for a 2021 study called Hidden Workers: Untapped Talent. They found an estimated 27 million Americans are systematically screened out of hiring despite having relevant skills, and 88% of executives admitted their own screening process causes them to miss qualified candidates.
Nearly half of the companies surveyed configure their ATS to automatically filter out anyone with an employment gap of six months or more. Recruiters interviewed at Amazon, Google, and Microsoft pushed back on the idea that this happens autonomously.
It’s a human-configured rules engine, not a robot making its own call. Someone decided a six-month gap was disqualifying. The software just enforces it.
88% of executives admitted their own screening process causes them to miss qualified candidates.
One tactic worth naming so you don’t try it: white-fonting, stuffing a resume with invisible keywords, sometimes an entire job description, in white or near-invisible text so a human never sees it but a parser might. It’s real.
Greenhouse’s 2025 AI in Hiring Report found 41% of surveyed job seekers admitted trying some form of keyword stuffing or prompt injection in a resume. But Greenhouse’s own detection across roughly 300 million resumes processed in the first half of 2025 found only about 1% actually contained white-text messages, worth flagging that Greenhouse sells the detection software, so it has some motivation on both sides of that number. ManpowerGroup, the staffing firm, separately reports catching hidden text in about 10% of the roughly one million resumes it scans a year, close to 100,000 resumes.
Either way, it backfires more often than it works. Workday, Greenhouse, and Lever are all reported to run automatic checks for low-opacity or invisible text now. Getting caught reads as dishonest, and it can mean outright disqualification regardless of whether you were actually qualified.
If you want a keyword to register, put it in an honest, visible, specific bullet instead. A parser and a human both read that the same way, which is the whole point of matching a job posting’s actual language once you start applying in a market this competitive.
Before you submit anything, run the copy-paste test. Select all the text in your PDF, copy it, and paste it into a completely plain text editor, not a word processor that’ll quietly reformat it for you. If sentences come out of order, a skill lands mid-way through a job title, or whole sections vanish, that’s roughly what an ATS is also seeing.
If you can’t even highlight the text in the first place, the file is a flattened image, not real text, and it will fail every parser regardless of platform. Free tools like Jobscan’s resume scanner will run a version of this check for you and flag specific formatting problems if you want a second opinion.
On file format: PDF and .docx parse about equally well on current versions of Greenhouse, Lever, and Workday. If you’re applying somewhere that might be running an older instance of iCIMS or Taleo, and there’s often no way to know which, .docx is the safer bet for anything with even a little layout complexity.
Reflective Coda
If you take one thing from all of this: structure first, wording second, keyword-stuffing never. The skeleton is boring on purpose. Boring is what gets read.
Every resume I’ve ever gotten hired off used some version of the shape in that diagram.
Every resume I’ve ever gotten hired off used some version of the shape in that diagram. Not because I lack imagination, but because a resume’s job isn’t to impress anyone with formatting choices. Its job is to get out of its own way long enough for the accomplishments to land. Save the creative risk-taking for your portfolio, where a human is guaranteed to actually look at it, and for the portfolio presentation once you’re in the room.
If you build the skeleton right, write bullets that prove something instead of describing a task, and format the file so nothing scrambles between your screen and theirs, you’ve done the actual job. The rest is luck, timing, and whoever’s reading it that day. You can’t control that part. You can control this one.
I still catch myself doing it, sorting a stack of resumes by shape before I’ve read a word, except now I’m looking at my own drafts instead of someone else’s stack. The habit never left. It’s just aimed differently now. Get the shape right, and the words finally get their turn.
Sources
- Eye-Tracking Study Shows Recruiters Look at Resumes for 7 Seconds (HR Dive, covering TheLadders’ 2018 study)
- Is the 6-Second Resume Scan a Myth? (ERE.net)
- My Personal Formula for a Winning Résumé (Laszlo Bock, LinkedIn, 2014)
- Create a Strong Resume (Harvard FAS Mignone Center for Career Success)
- 2025 ATS Usage Report: Fortune 500 Companies (Jobscan, Jul 2025)
- Why ATS Tables and Columns Break Your Resume Parsing (Jobscan)
- New Study Says “Hidden Workers” Are Being Excluded (Harvard Gazette, on the HBS/Accenture “Hidden Workers: Untapped Talent” study, 2021)
- The ATS Resume Rejection Myth (The Interview Guys)
- Job Seekers Are Hiding Secret Text in Their Resumes (The Interview Guys, citing Greenhouse’s 2025 AI in Hiring Report and ManpowerGroup data)
- Free Resume Scanner (Jobscan)
