Guides · 6 min read
Resume writing guides that answer real questions
Five short references on the parts of resume writing people get wrong most often: the structure, the parser, the verbs, the numbers, and the layout. Each one ends with real before and after examples and the questions people actually ask.
How to use these guides
These are not general career advice. They are about the document itself: how it is structured, how it is read, and how it is machine-parsed before a human ever sees it. That focus is deliberate, because almost every disappointing outcome with a resume traces back to a decision about the page rather than about the work.
The most common failure is not a weak career. It is a strong career rendered in a way that is slow to scan, ambiguous to a parser, or padded with fifteen years of duties that tell a screener nothing. Each guide below attacks one of those problems. They are written to be read once and then used as a checklist while you edit, which is why they stay short and heavy on examples rather than long and heavy on theory.
- Structure first. A resume is read in a fixed order, and most readers never deviate from it.
- Parseability before aesthetics. If a machine cannot read the file, a human never gets a chance to.
- Evidence over description. Every claim needs a verb and, where one exists, a number.
- Layout serves scanning. One column is the default because it is the only layout every reader agrees on.
The example below is a single bullet, and it is the shape most of the work in this set converges on. The first version is not wrong, exactly. It is just a description of effort, and a screener reading forty of these in a row cannot tell it apart from the other thirty-nine. The second version names what changed, for whom, and by how much, which means it can be compared against the job you are applying for. Almost every rewrite in these guides is some version of that move: replace a description of activity with a description of effect.
- Responsible for managing the onboarding process for new warehouse hires across three sites, working with HR and IT.
- Helped with the quarterly reporting, including building some of the spreadsheets used by the finance team.
- Rebuilt onboarding for three sites, cutting time-to-first-shift from nine days to two and removing the HR handoff for 60 hires a quarter.
- Automated the quarterly report, cutting four days of manual spreadsheet work per close and removing a recurring source of version errors.
Notice what did not change. The scope stayed the same, the employer stays unnamed, and the tone stayed plain. Only the content of the sentence improved, and it improved because the writer spent two minutes asking what the work actually produced. That is the whole method: no cleverness, just a question about effect that most people stop asking after the first job.
The set
Five guides, each about one decision. You can read them in any order, but they are written as a sequence because the later ones assume the earlier ones.
A sensible reading order
Read resume basics first, even if you think you already know the material, because it is the one that establishes the vocabulary the rest of the set uses. Sections have names. Entries have fields. Everything downstream assumes you can see those objects in your own document.
Then read ATS formatting. This is the guide people skip and the one that quietly costs the most interviews. A resume can be flawless on paper and score near zero because the dates never made it out of the PDF, and nothing on the page tells you that happened. The guide explains how to check rather than how to hope.
After that, work on action verbs and quantifying impact together, entry by entry, because they edit the same sentence. A weak verb with no number and a strong verb with a number are the same rewrite at different levels of effort. Doing them one after the other on the same bullet list takes barely longer than doing either alone.
Read when to use two columns last, and only if you are actively considering a sidebar. It is the only guide in the set that argues for breaking a rule, so it is the one most useful to read once you already know the rule well enough to want an exception.
What these guides assume
They assume you are applying somewhere that reads your resume properly, either through a human screener or through a parser that a human will later look at. They assume a mostly English-language, mostly office-based role, which is the context where the conventions below were tested. Where conventions differ, they say so: date formats change by region, photo expectations change by country, and public-sector and academic applications in some places use a plain, restrained format that a marketing-style resume will actively hurt you in.
They also assume you have access to some facts about your past work. Not a spreadsheet. Just the memory of what the number was, or the ability to work out that the team handled roughly four hundred cases a month. The guide on quantifying impact spends a lot of time on where those numbers hide, because for most people the barrier is not writing skill. It is that nobody ever mentioned writing the number down at the time.
A note on tooling
None of this requires a specific builder. It does require a tool that does not fight you: a document you can edit inline, a layout you can see before you export, and an export that produces a real text layer rather than a picture of a page. If your editor is showing you something the parser will not see, no amount of careful writing downstream will fix it.
A self-check you can run this week
If you do one thing from this set, do this. It takes under an hour, needs no software, and catches the majority of what is costing you interviews. Work through it in order, because each step depends on the one before it.
- Print it. A paper copy tells you the truth about density. If a section is hard to scan on paper, it is worse on a screen, and if you are relying on colour or rules to separate sections, the print has already told you those separators do not exist.
- Read only the first third of each entry. Truncating your own reading to the first three lines simulates the actual skim. If those three lines do not tell a stranger what you are good for, the rest of the entry is not being read and does not need to be there.
- Circle every number. You are looking for entries with no numbers, not entries with too few. Those are the ones to fix first, and they are usually the ones describing coordination, communication, or management, which is where people are vaguest.
- Search your file for the words from the job description. Not to stuff them in. To find out which of your real skills are described using different vocabulary than the posting does. A synonym is invisible to a keyword filter and invisible to you, which is the whole problem.
- Copy the text out of your exported PDF. Paste it into any plain text editor and read it in order. If the order is jumbled, or a date has merged into a company name, you have found a parse problem before a recruiter did.
The last step is the one people skip, and it is the one that matters most. A PDF that looks perfect in a viewer is frequently a picture of a page, and a picture of a page contains no text for anything to extract. Five minutes of checking removes an entire category of failure that no amount of careful writing will ever fix.
Questions people ask
Do I need to read all five guides?
No. Read the one that matches your problem. If you are not sure what your problem is, start with resume basics, then ATS formatting, because those two decide whether the rest of your content ever gets read. Action verbs and impact numbers are the last mile, and the two-column guide only matters if you are deliberately running an ATS-heavy search.
Are these guides specific to the Resumeboard builder?
The advice stands on its own. It describes how recruiters screen by hand and how applicant tracking systems parse files, neither of which cares what tool you typed the text into. The builder happens to implement most of it by default, so the examples map onto what you see on screen.
How long should a resume be?
One page is the right target up to roughly ten years of relevant experience. Beyond that, two pages is normal and expected, not a penalty. What matters more than the count is density: every line should carry information a screener can use, and no line should exist to fill space.
Do I need a photo, references, or a personal statement?
Not in most markets. Photos introduce bias risk and slow down some parsers. References are assumed and only requested at offer stage. A personal statement belongs in a cover letter. The space those three take up is better spent on evidence of what you did.
Why do these guides repeat the same advice about single columns?
Because layout is the single most common reason a strong resume is scored badly, and the failure is invisible from the inside. You cannot see a broken parse while you are looking at your own page. Each guide repeats the constraint only where it is load bearing, and links to the two-column guide for the cases where the constraint has real exceptions.
Write it once, then make it parse
The builder keeps your draft in this browser, edits the document itself rather than a form beside it, and exports a text layer a parser can actually read.