How to Quantify Achievements on Your Resume

To quantify achievements on your resume is to turn "improved the reporting process" into something a reader can weigh: how much, compared with what, for whom. Employers ask for exactly this. In NACE's 2026 survey, employers said listing skills is not enough; they want examples of how candidates used them (NACE: skills to showcase). Numbers are the most compact form of that evidence. They also invite questions, so they have to be right.

What to measure in data and tech work

Most achievements in data roles fall into a few measurable categories:

• Time: hours saved, report preparation cut, pipeline runtime reduced, time from request to answer.
• Scale: rows processed, sources integrated, users of a dashboard, teams served, requests handled.
• Quality: errors found or prevented, data quality checks added, incidents reduced, accuracy improved against a baseline.
• Money: cost reduced (for example cloud spend), revenue protected or influenced, with care about attribution.
• Decisions: a recommendation adopted, a process changed, a product decision informed by your analysis.

Pick the measure that matches what the work was for. A dashboard's value is usually who uses it and what they decide with it; a pipeline's is reliability and runtime.

State the change correctly

The same change can be stated in several honest ways, and choosing the right one matters. The code below shows three examples: one illustrative, two using real data.

import pandas as pd


def describe(label, before, after, unit=""):
    """The same change stated three honest ways: absolute, relative, and before-to-after."""
    diff = after - before
    rel = diff / before
    print(f"   {label}: {before:g}{unit} -> {after:g}{unit} | change {diff:+.3g}{' points' if unit == '%' else unit} | relative {rel:+.0%}")


print("Q1 time saved by automating a weekly report (illustrative inputs)")
before_h, after_h, weeks = 6.0, 0.5, 48
describe("hours per report", before_h, after_h, " h")
print(f"   per year: {(before_h - after_h) * weeks:.0f} hours saved over {weeks} working weeks")

print("\nQ2 percent vs percentage points, with real survey data (2025 Stack Overflow survey)")
so = pd.read_csv("results.csv", low_memory=False, usecols=["Employment", "RemoteWork", "WorkExp"])
work = so[so.Employment.isin(["Employed", "Independent contractor, freelancer, or self-employed"]) & so.RemoteWork.notna()]
junior = work[work.WorkExp.between(1, 2)].RemoteWork.eq("Remote").mean() * 100
senior = work[work.WorkExp >= 11].RemoteWork.eq("Remote").mean() * 100
describe("fully remote, 1-2 years vs 11+ years of experience", round(junior), round(senior), "%")

print("\nQ3 the same trap with wage growth (Atlanta Fed, US job stayers)")
wgt = pd.read_excel("wage-growth-data.xlsx", sheet_name="data_overall", header=1)
wgt = wgt.rename(columns={wgt.columns[0]: "month"}).dropna(subset=["month"])
wgt["Job Stayer"] = pd.to_numeric(wgt["Job Stayer"], errors="coerce")
by_year = wgt.groupby(wgt.month.dt.year)["Job Stayer"].mean().round(1)
describe("median pay growth, 2022 vs 2026 so far", by_year[2022], by_year[2026], "%")

Q1 time saved by automating a weekly report (illustrative inputs)
   hours per report: 6 h -> 0.5 h | change -5.5 h | relative -92%
   per year: 264 hours saved over 48 working weeks

Q2 percent vs percentage points, with real survey data (2025 Stack Overflow survey)
   fully remote, 1-2 years vs 11+ years of experience: 20% -> 37% | change +17 points | relative +85%

Q3 the same trap with wage growth (Atlanta Fed, US job stayers)
   median pay growth, 2022 vs 2026 so far: 5.5% -> 3.5% | change -2 points | relative -36%

Example 1: time saved. Cutting a weekly report from 6 hours to 30 minutes is a 92% reduction, or about 264 hours a year over 48 working weeks. For a resume line, "cut weekly report preparation from 6 hours to 30 minutes" is clearer than "92% faster", because the reader sees the scale. The yearly figure is useful in a promotion case or an interview story.

Example 2: percent versus percentage points. In the 2025 Stack Overflow survey (Stack Overflow survey 2025), 20% of people with one or two years of experience worked fully remotely, against 37% of those with 11 or more. That is a rise of 17 percentage points, which is an 85% relative increase. Writing "remote work increased by 17%" would be wrong on both counts. When the quantity you are changing is itself a percentage, say "points" for the difference.

Example 3: the same trap with rates. US job stayers' median pay growth fell from 5.5% in 2022 to 3.5% in 2026 so far, according to the Atlanta Fed's tracker (Atlanta Fed: Wage Growth Tracker). That is a fall of 2 percentage points, or 36% in relative terms. Both are true; they give very different impressions. On a resume, prefer the version a reader can check against the before and after values.

Rules for honest numbers

1. Always know the baseline. "Reduced errors by 40%" needs a before-figure you could explain.
2. Prefer before-and-after values to bare percentages: "from 6 hours to 30 minutes".
3. Use points for differences between percentages, and percent for relative change.
4. Attribute fairly. If a team achieved the result, say what your part was: "built the model that fed the pricing change", not "increased revenue by 12%".
5. Round sensibly. "About 260 hours a year" is more credible than "263.7 hours".
6. Keep a record of how you got each number. Interviewers may ask, and you should be able to explain it in a sentence.

When you do not have exact numbers

Estimate honestly and say so in conversation if asked:

• Count what you can: reports automated, users, requests per month, sources joined.
• Time it now: run the old and new process once and measure.
• Use ranges or scale words backed by facts: "used weekly by store managers in 12 regions".
• Ask your stakeholders: "How much time does this save you each week?" Their answer is real evidence.

Never invent a number. A figure you cannot explain does more damage in an interview than having no figure at all.

Quantifying work that prevented problems

Some of the most valuable data work is invisible because it stopped something bad from happening: a validation check that blocks bad loads, a monitoring alert, a documented definition that ended arguments. Quantify these by what they prevented or how often they acted: "the check has blocked 14 bad loads since March" or "the alert caught three delayed feeds before the morning report". If nothing has gone wrong yet, describe coverage instead: "validation on all 22 source tables".

Turning numbers into resume lines

Use the pattern action, tool, result, scale:

• "Automated the weekly sales report in SQL and Power BI, cutting preparation from 6 hours to 30 minutes."
• "Added 25 data quality checks to the orders pipeline; failed loads dropped from about one a week to one a quarter."
• "Built a churn analysis used by the retention team to target 300 at-risk accounts a month."

For the rest of the page, see resume building for data roles, and for the mistakes these lines replace, see common resume mistakes. The same habit of measuring your work feeds directly into a promotion case.

Measuring and communicating results is practised throughout the Data Analytics course.

Sources

• 2025 Stack Overflow Developer Survey — Stack Overflow
• Wage Growth Tracker (Federal Reserve Bank of Atlanta) — Federal Reserve Bank of Atlanta
• The High-Impact Skills College Students Should Showcase on Their Resumes (National Association of Colleges and Employers) — National Association of Colleges and Employers