Why keyword matching is so specific
Screening software is not reading for meaning. It is looking for exact or near-exact matches. "Stakeholder management" and "managing relationships with stakeholders" may mean the same thing to a human reader, but a basic keyword system will not count them as a match. Similarly, "Microsoft Excel" and "spreadsheets" are not the same as far as many screening tools are concerned.
This specificity catches a lot of candidates out. Someone with 10 years of relevant experience may score lower than someone with 3 years simply because the experienced candidate used different terminology to describe the same skills.
Where to find the keywords you need
The job description is your source of truth. Read it carefully and look in these places:
The job title and department name
If the role is titled "Business Analyst" and your CV says "Systems Analyst" throughout, that mismatch will cost you. Where relevant and accurate, align your language to the title they have used.
The "Required skills" or "Essential experience" section
This is the most heavily weighted section for keyword matching. Every item listed as essential should appear somewhere in your CV if you genuinely have that skill. Use the exact phrasing where possible.
The responsibilities section
The tasks listed under "What you will be doing" tell you the language the team uses day to day. If they say "sprint planning" rather than "agile ceremonies," use their phrasing.
The "Desirable" or "Nice to have" section
These are worth including too, even if you only have partial experience. A partial match is better than no mention.
Quick method: Paste the job description into a word frequency tool or just read it twice. The words that appear most often are usually the most important. Make sure each one appears at least once in your CV in a natural context.
Hard keywords vs soft keywords
Not all keywords carry equal weight. Hard keywords tend to matter more.
Hard keywords
These are specific, verifiable and easy for software to match: tool names (Salesforce, Python, Xero, SystmOne), certifications (PRINCE2, CIPD Level 5, AWS Certified), methodologies (Agile, Kanban, Six Sigma), and technical skills (SQL, machine learning, CAD).
Soft keywords
These are broader terms like "communication skills," "team player" or "leadership." They are still worth including where the JD uses them, but they carry less differentiation in keyword scoring because every CV has them.
Where to put keywords in your CV
Keyword placement matters. These are the most effective locations:
- Professional summary — this is often read first by both software and humans. Leading with your strongest keywords here is high value.
- Skills section — a dedicated skills list lets you include keywords clearly and concisely. Keep it to genuinely held skills.
- Bullet points in work history — embed keywords naturally into achievement-led bullets rather than listing them in isolation.
What keyword stuffing looks like, and why to avoid it
Keyword stuffing means including terms in ways that feel unnatural or dishonest, just to improve a score. "I have strong Python Python Python skills in Python development" is an extreme example, but subtler versions include listing tools you have barely used or inserting keywords in white text (invisible to humans but visible to software). This is both ineffective with modern screening systems and dishonest. A recruiter who reads past the software will notice.
The goal is not to trick the system. It is to accurately represent what you have done using the language the employer is already using to describe the role.
How many keywords is enough?
There is no universal answer. A reasonable target is to match at least 70% of the essential requirements listed in the job description. For a role with 10 listed requirements, that means 7 of them should appear clearly in your CV. The further below that threshold you fall, the less likely the software is to pass your application through.