The "Exact Phrase" CV Exchange: How to Swap Your Duties for Job Description Verbs Without Losing Your Authentic Voice

Your CV describes exactly what you did. The job description asks for exactly what you did. The system rejects you anyway.

This happens constantly. Not because you lack the skills. Because you called it "managed client accounts" and they called it "client relationship management."

Around 70% of large companies use applicant tracking systems². These systems compare your words against theirs. When the words don't match, you don't progress.

The fix isn't stuffing keywords into your summary. It's swapping your duty language for their duty language, one phrase at a time, while keeping the substance true.

01What this problem really is

ATS platforms convert your CV into structured data. They pull out job titles, employers, dates, and skills. Then they score you against the job description¹².

The scoring often relies on explicit matches. "Project management" as a noun phrase gets tagged as a skill. "Managing projects" as a verb phrase might not¹². Same work. Different label. Different outcome.

Newer systems use natural language processing to interpret meaning¹². But the sophistication varies wildly between tools. And employers can configure systems to prioritise their exact wording¹⁵.

The result is a linguistic bottleneck. Your CV has to be machine-readable in format and machine-matched in vocabulary before a human ever sees it.

02Why it happens

ATS weren't built to understand you. They were built to reduce a pile of 500 applications to a shortlist of 20.

When recruiters configure these systems, they often pull keywords directly from job descriptions³. The system then looks for those keywords in your CV. If you paraphrased, it might not register.

Generic advice tells you to "add keywords." So people do. They sprinkle terms into their skills section or summary. But the real problem is in the bullet points, where duties are described in language the system doesn't recognise³¹¹.

AI modules now layer onto ATS to summarise CVs and extract skills¹⁸. This adds another filter. If the AI summary doesn't surface the right phrases, you may be scored down before anyone reads your actual text.

Formatting compounds the issue. Tables, columns, graphics, and headers can confuse parsers¹³. Even if your wording is perfect, a misread layout means your keywords never get counted.

03How it affects job seekers

You apply. You wait. Nothing.

One widely cited figure suggests up to 75% of qualified candidates get rejected due to formatting and readability issues alone¹¹. That number comes from a generic advice source and should be treated cautiously. But the pattern is real: small technical errors block substantive fit from being seen.

The problem is worst for career changers and non-traditional candidates. If your previous role used different terminology, your experience may be structurally invisible to the ATS.

There's also the authenticity trap. You're told to mirror job description language. So you do. But now your CV sounds like everyone else's. Recruiters reviewing shortlists see the same phrases repeated across applicants¹⁴.

And when AI tools generate CVs wholesale, the homogenisation gets worse. Candidates end up with keyword-dense documents that pass filters but fail to distinguish themselves when a human finally reads them.

04What to do instead

The "exact phrase" exchange is a three-step drill. It replaces vague duty language with employer-specific verbs while preserving the truth of what you actually did.

1. Identify your generic duty bullets

Look for phrases like "responsible for," "helped with," "managed," or "handled." These are red flags. They describe activity without matching employer vocabulary.

2. Extract the exact verbs and phrases from the job description

Read the posting line by line. Note the specific terms they use for skills, responsibilities, and outcomes. "Stakeholder engagement." "Data analysis." "Cross-functional collaboration." Write them down.

3. Rewrite each bullet using their language, anchored by your metrics

Take your original duty and translate it into their phrase. Then add a number or concrete outcome.

Responsible for sales reporting.

Owned client reporting and dashboard creation, improving data accuracy by 20% and reducing reporting time by two days per cycle.

The phrase "client reporting" comes from the job description. The metrics come from you. The result is a bullet that satisfies ATS matching and proves substance to a human.

Do this for your top five to seven bullets per role. Focus on the ones that align most closely with the job's core requirements.

05Common mistakes to avoid

• Keyword stuffing without context. Listing 'project management, stakeholder engagement, cross-functional collaboration' in a skills block doesn't help if your bullet points say none of those things. ATS often score based on contextual use, not isolated mentions¹¹.
• Using synonyms and hoping. 'Managed customer relations' might not register as 'customer relationship management.' The grammatical difference matters to parsers¹². When in doubt, use the noun phrase from the job description.
• Copying verbs you can't back up. If the job description says 'led strategic initiatives' and you attended a few planning meetings, don't claim the verb. Phrase swaps must reflect your actual scope. The ICO requires that automated decisions in recruitment can be contested, which means you may need to defend what you wrote⁷.
• Ignoring formatting. Submit in .docx unless the employer specifies otherwise. Avoid tables, columns, text boxes, and graphics. Use standard headings like 'Experience' and 'Education'¹³.
• Over-tailoring to the point of exhaustion. You don't need a fully rewritten CV for every application. Build a master CV with multiple bullet versions for common role types. Swap in the relevant version for each application.

06A realistic example

Original bullet from a marketing coordinator's CV:

Helped manage social media accounts and worked with the team on campaigns.

Job description phrase: "Social media strategy and campaign execution."

Rewritten bullet:

Contributed to social media strategy and campaign execution across three platforms, supporting a 12% increase in engagement over six months.

The original was vague and passive. The rewrite uses the exact job description phrase, clarifies scope (three platforms), and adds a metric (12% engagement increase).

The work described is the same. The language is now aligned. The candidate hasn't exaggerated. They've translated.

07Key takeaway

The "exact phrase" exchange isn't a trick. It's a translation.

You're not fabricating skills. You're describing real work in the language the system expects. Every swap should pass a simple test: could you explain this in an interview without backpedalling?

ATS and AI screening favour explicit matches. Formatting errors kill visibility. Generic advice to "add keywords" misses the real problem, which is that your duty language doesn't match their requirement language.

Fix that. Keep the substance. Add the numbers. Sound like yourself, in their words.

08Frequently Asked Questions

09Sources

• ¹ https://www.jobscan.co
• ² https://www.selectsoftwarereviews.com/blog/applicant-tracking-system-statistics
• ³ https://www.indeed.com/career-advice/resumes-cover-letters/automated-screening-resume
• ⁷ https://privacymatters.dlapiper.com/2026/04/uk-ico-report-on-automated-decision-making-in-recruitment/
• ¹¹ https://www.indeed.com/career-advice/resumes-cover-letters/resume-keyword-scanners
• ¹² https://www.candidately.com/glossary/nlp-recruitment
• ¹⁴ https://www.cipd.org/en/knowledge/bitesize-research/fairness-ai-recruitment/
• ¹⁵ https://dl.acm.org/doi/full/10.1145/3696457
• ¹⁸ https://www.telegraph.co.uk/money/jobs/career-advice/how-to-get-your-cv-seen-by-a-human/