Why "Improved Performance by 30%" Hurts Software Engineer Resumes
Adding metrics isn't automatically a win. Without context, numbers don't build trust, they quietly erode it.
Thejus Sunny
Engineering + hiring perspective
If you've been told "just add metrics" to your resume, this might surprise you:
A quantified bullet can still hurt your resume.
In fact, bullets like this often get ignored, or quietly discounted, by hiring managers:
• Improved performance by 30%
On the surface, this looks strong. It has a number. It claims impact. In practice, it raises more red flags than it resolves.
This post explains why.
The problem isn't the number, it's credibility
Hiring managers don't read resumes like marketing copy. They read them like bug reports.
When they see:
"Improved performance by 30%"
their brain immediately asks:
- Performance of what?
- Measured how?
- Compared to what baseline?
- On which system, at what scale?
- Was this sustained or a one-off benchmark?
If none of that is answered, the number doesn't build trust, it erodes it.
Why this bullet quietly fails in real reviews
Let's be blunt: hiring managers have seen hundreds of inflated performance claims.
When a metric is presented without context, it triggers one of three reactions:
Assumed exaggeration
"This sounds like resume padding."
Assumed triviality
"30% of something small probably doesn't matter."
Assumed misunderstanding
"They may not understand what they're measuring."
An unscoped metric is worse than no metric at all.
No metric = neutral. Bad metric = credibility loss.
Common variations of the same mistake
These all suffer from the same problem:
- "Reduced API latency by 40%"
- "Improved system performance by 2x"
- "Optimized database queries by 35%"
- "Increased throughput significantly"
They look different. They fail for the same reason: no verification path. A hiring manager cannot reconstruct what happened.
The practical takeaway
If your resume has bullets like:
- "Improved performance by X%"
- "Reduced latency by Y%"
- "Optimized system efficiency"
Ask yourself: could a stranger explain what I actually did without guessing?
If the answer is no, the metric isn't helping you.
One last thing
This isn't about being "more impressive." It's about being more believable. Hiring managers don't reward the biggest numbers, they reward the clearest thinking.