Somewhere between draft three and draft four, the plain bullets started looking boring. The template offered little arrows. Checkmarks for the skills section. Maybe a diamond or two. It looked sharper, so you kept it, and now a nagging voice is asking whether those cute symbols are quietly wrecking your applications.
The honest answer is more interesting than the usual "avoid all symbols" scare advice: standard bullets are completely safe, most common symbols survive fine in modern systems, and a specific handful of glyph choices genuinely can corrupt your text in ways you will never see. This guide shows you the actual failure mode, what corrupted bullets look like inside a parsed record, and the short safe list that ends the worry permanently.
What happens to a bullet at extraction time
When an ATS ingests your resume, the first step is text extraction: the software pulls the characters out of your PDF or Word file into plain text, which then gets carved into your candidate record. The full journey is described in what an ATS actually reads from your resume.
Every character in your file is stored as a code from a standard called Unicode, and extraction is essentially reading those codes back out. Common characters, letters, numbers, punctuation, and yes, the standard round bullet, are the most well-supported codes in all of computing. They extract perfectly, every time, in every system.
The trouble starts with three kinds of exotic characters:
- Obscure symbols from the far corners of Unicode: decorative arrows, ornamental stars, pointing hands, geometric dingbats. Support varies by parser, PDF generator, and font.
- Icon-font glyphs, where a template uses a symbol font (like the classic Wingdings trick) to display a checkmark. The file does not actually contain a checkmark character; it contains some other character wearing a checkmark costume. Extracted, the costume comes off and you get the underlying junk character.
- Emoji, which are multi-byte characters that older extraction pipelines can split into fragments of gibberish.
Note what is not on the risk list: styling. A bold bullet, a colored bullet, an indented bullet, all irrelevant, because parsers read character codes, not appearances.
What corrupted bullets actually look like
Here is the part almost no advice shows you: the actual damage, as it appears inside a parsed record. When an exotic glyph fails extraction, one of three things happens.
Replacement characters. The parser cannot map the glyph, so it substitutes a placeholder. Your elegant line:
➤ Led migration to new billing system
becomes:
? Led migration to new billing system
Mildly ugly, mostly harmless. This is the lucky outcome.
Garbage sequences. Icon-font tricks are worse, because the underlying character is real but wrong. A Wingdings-style checkmark often extracts as a random letter, so your skills section reads:
ü Excel ü Salesforce ü SQL
A recruiter viewing your parsed profile sees what looks like typos scattered through your record. Not fatal, but it makes careful work look careless.
Fused words. The genuinely damaging case. Sometimes the failed character disappears entirely, and with it the whitespace or line break it anchored. Adjacent words glue together:
Managed vendor contractsReduced processing time
Now the extracted text contains "contractsReduced," a word that exists nowhere. A recruiter searching the ATS for "reduced processing time" gets no match from your record. You had the exact phrase. The bullet ate it. And nothing anywhere will tell you this happened.
This fusing failure is why the bullet question deserves five minutes of attention instead of zero. It converts an invisible cosmetic issue into lost keyword matches, and keyword matches are how you surface in searches at all.
The safe list
You do not need to memorize Unicode. Here is the complete practical guidance:
Always safe:
- The standard round bullet, the default in Word and Google Docs.
- The plain hyphen at the start of a line. Utterly unbreakable.
- Numbered lists using ordinary digits and periods.
Fine in modern systems, acceptable risk:
- The simple dash variants and the small square bullet that mainstream word processors insert. These are common, well-supported characters.
Skip on a resume:
- Arrows, checkmarks, stars, diamonds, pointing hands, and any glyph you inserted from a symbol menu or copied from a website.
- Anything rendered via Wingdings, Webdings, or an icon font, including the built-in "define new bullet" symbol options that reach into those fonts.
- Emoji, everywhere on the document.
The trade is lopsided: exotic bullets add zero information and a nonzero chance of corrupting the text around your best keywords. A recruiter has never once thought "strong candidate, and those arrow bullets seal it." Spend your differentiation budget on the words after the bullet.
One related note: if your fancy bullets came bundled with a designed template, the bullets are probably the smallest risk in the file. Templates bring text boxes, columns, and graphics whose failure modes dwarf a bad glyph, and those are covered in do text boxes and shapes confuse ATS and the broader resume template question.
The 60-second corruption test
Test your actual file instead of trusting rules:
- Export your resume exactly as you plan to submit it.
- Open the PDF, select all, copy.
- Paste into a plain-text editor (Notepad, or TextEdit in plain-text mode).
Scan the paste for the three failure signatures: question marks or boxes where bullets were, stray letters at the start of lines, and fused words where line breaks used to be. Read your skills section character by character, because that is where symbol decoration concentrates and where keyword damage costs the most.
If everything reads clean, your bullets are safe, whatever they are. If you see junk, switch the affected lists to standard bullets or hyphens, re-export, and test again. Two minutes, permanent fix.
Check the whole extraction, not just the bullets
Bullet corruption is one member of a family: fonts that do not embed, text boxes that vanish, headers that scramble. They all share one trait, the file looks perfect to you and broken to the parser. Which means the real safeguard is not a rule list. It is looking at the extraction itself.
The free Bounce scan at careerbounce.io shows you your resume as raw extracted text, exactly what a parser pulls from your file, gibberish and all. Every corrupted bullet, every fused word, every vanished section is right there to see and fix. It is free, no signup, and it runs entirely on your device, so your resume never leaves your machine.
Run the scan before your next application. Clean bullets will not get you hired, nothing cosmetic will, but they make sure the honest, specific accomplishments you wrote actually arrive intact, readable by every bot and every human on the other side.