Common Resume Mistakes That Get You Rejected

Most data resumes that get rejected are not rejected because the candidate lacks skills. They are rejected because the page hides them. Hiring processes designed to find "perfect" candidates efficiently already exclude many qualified people, as Harvard Business School's research on hidden workers found (HBS: Hidden Workers). A resume full of avoidable mistakes makes that more likely. To make the common mistakes concrete, the code below runs a few simple checks on an illustrative junior analyst resume, before and after editing.

import re

# An illustrative junior data analyst resume, before and after editing (not a real person)
before = """Objective: Seeking a challenging position in a reputed organisation where I can utilise my skills.
Skills: Python, SQL, Excel, Power BI, Tableau, Machine Learning, Deep Learning, NLP, Spark, Hadoop, AWS, Azure
Experience: Responsible for creating reports for the sales team.
Was involved in data cleaning activities.
Worked on dashboard development in Power BI.
Responsible for handling ad-hoc requests.
References: Available on request."""

after = """Data analyst, SQL and Power BI, retail sales reporting
Skills: SQL, Python (pandas), Power BI, Excel
Experience: Automated the weekly sales report in SQL and Power BI, cutting preparation from 6 hours to 30 minutes.
Found and fixed duplicate orders in the returns join, correcting revenue by 3% for the quarter.
Built a regional dashboard used by 12 store managers in their Monday review.
Answered 40+ ad-hoc requests a month; turned the 5 most frequent into self-service views."""

CHECKS = {
    "objective statement instead of a summary": lambda t: bool(re.search(r"^objective:", t, re.I | re.M)),
    "'references available on request' line": lambda t: "available on request" in t.lower(),
    "duty phrases ('responsible for', 'worked on', 'involved in')": lambda t: len(re.findall(r"responsible for|worked on|involved in", t, re.I)),
    "lines after the skills line with no number": lambda t: sum(1 for line in t.splitlines()[2:] if line.strip() and not re.search(r"\d", line)),
    "skills listed (fewer, provable ones read better)": lambda t: len(re.split(r",\s*", re.search(r"Skills:(.*)", t).group(1))),
}

print(f"{'check':<58}{'before':>8}{'after':>8}")
for name, check in CHECKS.items():
    print(f"{name:<58}{str(check(before)):>8}{str(check(after)):>8}")
words = lambda t: len(t.split())
print(f"{'words':<58}{words(before):>8}{words(after):>8}")

check                                                       before   after
objective statement instead of a summary                      True   False
'references available on request' line                        True   False
duty phrases ('responsible for', 'worked on', 'involved in')       4       0
lines after the skills line with no number                       5       0
skills listed (fewer, provable ones read better)                12       4
words                                                           62      78

The "after" version is longer, 78 words against 62, yet it is far stronger: every line now says what happened and how much. Here is what each check is looking for.

Mistake 1: an objective statement

"Seeking a challenging position in a reputed organisation" tells the reader what you want, not what you offer, and every second resume says it. Replace it with a one-line summary: target role, core tools and domain, such as "Data analyst, SQL and Power BI, retail sales reporting".

Mistake 2: duties instead of results

"Responsible for creating reports" describes a job description, not your contribution. The checker found four duty phrases in the "before" version and none after. Rewrite each line as action, tool and result: "Automated the weekly sales report in SQL and Power BI, cutting preparation from 6 hours to 30 minutes." Employers say the same thing: in NACE's 2026 survey, they stressed that listing skills is not enough and that they want to see examples (NACE: skills to showcase).

Mistake 3: no numbers

All five lines after the skills line in the "before" version lacked a single number; none did afterwards. Numbers give scale and credibility: rows, users, hours saved, error rates, reports automated. If you have no exact figure, use honest scale or before-and-after. How to quantify achievements on your resume shows how to measure them.

Mistake 4: the skill dump

Twelve skills, including deep learning, NLP, Spark, Hadoop and two clouds, on a junior analyst resume invite the question "which of these can you actually do?" An interviewer will pick one and probe. Four skills you can prove are more convincing than twelve you have touched. List what the target role needs and what your experience lines demonstrate.

Mistake 5: wasted lines

"References: available on request" takes a line and adds nothing. MIT's career office notes that employers assume references are available and that you should not list them unless asked (MIT CAPD: resumes). Use that space for another result, or simply leave the page less crowded so the results you have stand out.

Why these mistakes cost more in data hiring

Data roles are, at heart, about turning messy information into clear, quantified conclusions. A resume is the first sample of that skill a hiring manager sees. When a candidate for an analyst role describes their own work without a single number, or lists twelve tools without showing what any of them produced, the reader draws a conclusion about how that person will present an analysis. The reverse is also true: a page that states results precisely, with honest scale and clear before-and-after comparisons, demonstrates the job before the interview starts. Treat your resume as a small analysis of your own career, with the same care you would give a report for a senior stakeholder.

A note on AI resume writers

Generative AI tools can tidy wording, but they also tend to produce exactly the generic phrases screeners skip, and they can invent results. Use them, if at all, to tighten sentences you have already written from real evidence, then check every claim against what you actually did.

Other mistakes the checker does not catch

• Formatting that screening software misreads: tables, text boxes, icons and multi-column layouts for key content.
• Inconsistent dates and titles that differ from LinkedIn or from what referees will say.
• Projects with no links, or links to empty repositories.
• Typos in company names, tools and your own email address.
• One generic resume for every role instead of a tailored version.
• Claims you cannot discuss. Anything on the page is fair game in an interview.

A fifteen-minute repair routine

1. Replace the objective with a one-line summary.
2. Rewrite every experience line as action, tool and result, with a number where honest.
3. Cut the skills list to what you can prove and the role needs.
4. Delete lines that add nothing, such as references.
5. Check formatting, dates and links, then ask someone to read it.

For the full structure of a data resume, see resume building for data roles.

Projects that give your resume real results to report are part of the Data Analytics course.

Sources

• Hidden Workers: Untapped Talent (Harvard Business School, Managing the Future of Work) — Harvard Business School
• The High-Impact Skills College Students Should Showcase on Their Resumes (National Association of Colleges and Employers) — National Association of Colleges and Employers
• Resumes (MIT Career Advising and Professional Development) — MIT CAPD