- Home
- Guides & research
- Same skill, different words
Original research
Same skill, different words: how 2,209 job posts phrase the same skills
Recruiters search an applicant tracking system with the words from their own job post. So we asked: how consistently do job posts name the same skill? We collected every open posting from 36 tech companies on November 6, 2026, and measured it.
Key findings
- When two job posts in the same role both ask for cross-functional work, they share no common phrase in 52–62% of cases.
- Two product manager posts that both ask for user research share no phrase 66% of the time ("user research", "customer discovery", "customer research", "customer interviews").
- Some skills are standard: 100% of data posts that mention querying say "SQL"; 98% of PM posts about planning say "roadmap".
- Titles barely repeat: 72 designer posts used 61 different exact titles.
- 66% of posts state a number of years of experience; the median minimum is 5 years.
The data
| Role family | Postings | Companies |
|---|---|---|
| Software engineer | 1,419 | 35 |
| Product manager | 375 | 33 |
| Marketing | 201 | 32 |
| Data analyst / data scientist | 142 | 28 |
| Product / UX designer | 72 | 21 |
The companies: Affirm, Airbnb, Airtable, Anthropic, Asana, Brex, Calendly, Chime, Cloudflare, Coinbase, Databricks, Datadog, Discord, Dropbox, Duolingo, Elastic, Figma, Flexport, GitLab, Gusto, Instacart, Lyft, Mixpanel, MongoDB, Okta, Palantir, Pinterest, Reddit, Robinhood, Samsara, Scale AI, Spotify, Squarespace, Stripe, Twilio, and Vercel.
Finding 1: the same skill has many names
For each skill concept, we listed the common phrasings and checked every pair of postings in the same role family that mention the concept. The key number is "no shared phrase": how often two posts that want the same skill have not one phrase in common. If your resume uses one company's wording, this is how often it misses the other company's search.
| Role / skill | Posts mentioning it | How they phrase it (% of those posts) | No shared phrase |
|---|---|---|---|
| PM / user research | 15% (56) | user research 50, customer discovery 23, customer research 16, customer interviews 14 | 66% |
| Software engineer / cross-team work | 20% (280) | cross-functional collaboration 41, cross-functional teams 36, cross-functional partners 28, stakeholder alignment 6 | 62% |
| Data / cross-team work | 54% (76) | cross-functional partners 49, cross-functional teams 41, cross-functional collaboration 26, stakeholder management 9 | 55% |
| PM / cross-team work | 45% (170) | cross-functional variants and stakeholder terms | 52% |
| Marketing / cross-team work | 55% (111) | cross-functional variants and stakeholder terms | 53% |
| Marketing / planning | 49% (98) | roadmap 58, project management 30, prioritization 28, program management 14 | 52% |
| Marketing / communication | 54% (109) | communication skills 57, written and verbal 47, storytelling 28, communicator 16 | 50% |
| Software engineer / CI/CD | 20% (277) | "CI/CD" 65, continuous integration 29, deployment pipelines 20, continuous delivery 8 | 47% |
| Marketing / data-driven | 26% (53) | data-driven 62, analytical skills 42, data-informed 6 | 45% |
Pairs of postings in the same role family that both mention the skill. Source: Tailor My Resume analysis of 2,209 job posts, November 6, 2026.
The pattern is strongest for the "soft-but-searchable" skills that every resume claims: working across teams, research, planning, and communication. These are also the skills people most often describe in their own words ("worked with sales and legal", "talked to customers").
Finding 2: some skills have one standard word
Not every skill is a moving target. Where an industry has settled on one term, almost every post uses it:
| Role / skill | Posts mentioning it | Dominant phrase | No shared phrase |
|---|---|---|---|
| Data / querying | 87% (123) | SQL: 100% | 0% |
| PM / planning | 84% (316) | roadmap: 98% | 3% |
| Data / experimentation | 60% (85) | experimentation 77%, experiment design 45%, A/B testing 42% | 23% |
| Data / visualization | 39% (56) | dashboards 80%, Tableau 30%, data visualization 29%, Looker 16% | 26% |
So the advice is not "use every synonym". It is: for standardized skills, use the standard word; for the many-named skills, use this post's word.
Finding 3: how many tools a job post names
We counted named hard skills and tools from a fixed list of 183 (languages, frameworks, cloud, data, design, analytics, and marketing tools).
| Role family | Median tools named | Middle 50% | Posts naming none | Most named (% of posts) |
|---|---|---|---|---|
| Software engineer | 5 | 3–7 | 8.9% | Python 40, Java 38, AWS 30, Go 27, Kubernetes 27 |
| Data analyst / scientist | 3 | 2–5 | 4.2% | SQL 87, Python 83, R 37, Spark 16, dbt 14 |
| Product / UX designer | 0 | 0–1 | 59.7% | Figma 33, CSS 19, HTML 18, JavaScript 15 |
| Marketing | 0 | 0–2 | 51.7% | Salesforce 22, Marketo 15, SQL 10, AWS 9 |
| Product manager | 0 | 0–1 | 68.0% | SQL 12, AWS 9, Azure 9, GCP 8 |
Engineering and data posts are tool-heavy, so exact tool names matter most there. Product, design, and marketing posts rarely name tools — their keywords are skills and methods, which is exactly where Finding 1 shows the most variation.
Finding 4: almost every job title is unique
| Role family | Posts | Distinct exact titles | Titles used only once |
|---|---|---|---|
| Software engineer | 1,419 | 1,069 | 897 |
| Product manager | 375 | 312 | 273 |
| Marketing | 201 | 173 | 156 |
| Data analyst / scientist | 142 | 117 | 102 |
| Product / UX designer | 72 | 61 | 55 |
The core title is stable — 70 of 72 designer posts are some form of "Product Designer" — but seniority and team words change everything after it: "Senior Product Designer, Growth", "Staff Product Designer — Payments". A resume headline should carry the core title the post uses; recruiters who search by title (55.3% do, according to Jobscan's recruiter survey) usually search the core.
Other findings
| Signal (all five families, n = 2,209) | Share of posts |
|---|---|
| States a number of years of experience | 66.1% (median minimum: 5 years) |
| Includes a salary range | 58.9% |
| Has a "preferred" or "nice to have" section | 46.3% |
| Has both "required" and "preferred" markers | 41.7% |
| Mentions a degree | 28.2% |
| Says "or equivalent experience" | 12.0% |
Median minimum years of experience by family, when stated: software engineer 5, data 5, product manager 6, designer 6, marketing 7.
What this means for your resume
- One resume cannot match every post's wording. For the most common skills, half or more of post pairs share no phrase. Tailoring is not polish; it is how the same experience becomes findable.
- Copy the phrase, not just the idea. If the post says "customer discovery", say "customer discovery" — if you did it — even if your old team called it "user research".
- Use standard terms where they exist. SQL, roadmap, dashboards, experimentation: use the standard word.
- Lead with the core title. Match "Product Designer", not the team suffix.
- Check the gap before you apply. A free ATS keyword check shows which required terms you have "in other words".
Methods and limitations
Methods
- Collection. One request per company to the public Greenhouse job board API and Lever postings API on November 6, 2026 (23:48–23:49 UTC), with a pause between requests. 34 companies on Greenhouse, 2 on Lever.
- Cleaning. HTML removed. Company boilerplate (Greenhouse intro and conclusion blocks) excluded from skill matching. Duplicates removed by company, title, and body text.
- Role families. Assigned by job title with fixed rules. Manager, director, and head titles excluded from the engineer family; brand, graphic, and visual designers excluded from the designer family.
- Matching. Case-insensitive whole-word matching; hyphens and spaces treated alike. Short terms that are also common words (for example Go, R, Spark, Excel) matched only in clear contexts. A company's own product was not counted in its own posts.
- Concepts. Ten concept groups, each with a fixed list of phrasings. For every pair of posts in a family that mention a concept, we checked whether their sets of phrasings overlap. Results for pairs from different companies only were within about 3 points.
- Validation. Random match contexts were read by hand for ambiguous terms; three false-match sources were found and fixed before the final run.
Limitations
- Tech-company sample. All 36 companies are US-headquartered tech companies, mostly software and fintech. Large boards weigh more. Do not generalize to other industries.
- Lower bound. Only the phrasings on our lists were counted. Real wording varies more, so the "no shared phrase" figures are likely underestimates.
- Mentions, not requirements. A phrase anywhere in the description counts, including team descriptions.
- Small cells omitted. We do not report concept results for groups with fewer than about 25 posts.
- One-day snapshot. Job boards change daily; a new collection will give somewhat different numbers.
You may quote and reuse these findings with a link to this page.
Questions
Do different companies use different words for the same skill?
Often, yes. In our November 2026 analysis of 2,209 tech job posts, two posts in the same role that both asked for cross-functional work shared no common phrase in 52–62% of cases, and two product manager posts that asked for user research shared no phrase 66% of the time.
Should I put every synonym of a skill on my resume?
No. Use the phrase from the job post you are applying to, for skills you have. Where an industry uses one standard term, such as SQL or roadmap, use that term.
How many tools does a typical job post list?
It depends on the role. In our data the median software engineer post named 5 tools from a 183-term list, data posts named 3, and most product, design, and marketing posts named none — their keywords are skills and methods instead.