By Peter Miller · Published March 24, 2026 · Last updated: September 14, 2026
11 min read
You're not failing because you forgot a buzzword. You're failing because the file treats skills, tools, and technologies like one interchangeable list. Workday and Greenhouse don't. They tag soft and hard skills as keywords, then look for product names and stack terms on dated jobs.
If you're applying tonight, check your resume for free against the posting before you upload. A pretty skills cloud that never touches an employer block still parses as tags, not experience.
This teardown isn't a dictionary of every programming language. It's how ATS interprets skills vs tools vs technologies on the same file a recruiter opens after the parser runs.
Copy the After shapes below. Swap in your stack. Don't leave the thirty-name footer on screen one.
Do these three things first
- Highlight three required tools from the posting. Each one needs a dated bullet under a real employer, not only a skills row.
- Move stack terms (Python, AWS, Kubernetes) from the footer into the job where you used them, with Month Year dates.
- Upload the single-column PDF and run the free checker against tonight's posting before you hit Apply.
Three buckets, three parser jobs
The standard these pairs are judged against is the posting plus what a recruiter can verify in thirty seconds, not a generic resume template. Skills answer whether you claim a capability. Tools answer which product you touched. Technologies answer which stack or language sat under that work.
Skills are tags. Tools are nouns on dated lines. Technologies are stack terms tied to a role.
I've screened Greenhouse queues where the skills footer listed Salesforce twelve times and the experience block still showed zero dated tool use under any employer.
Skills cover what you can do: stakeholder management, financial modeling, conflict resolution, user research. The parser stores them as keyword tokens. They help matching when the posting uses the same phrase. They do not prove you did the work unless a bullet shows an object and a date.
Tools are named products: Salesforce, Jira, Excel, Zendesk, Figma, HubSpot. Applicants often dump every tool they have ever opened into one row. The parser still tags them. The recruiter asks when, and under which title.
Technologies are languages, platforms, and frameworks: Python, SQL, AWS, React, Docker. Technical postings hunt these terms twice: once as tags, once inside bullets that show what you shipped. A footer-only mention reads like a wish list.
And the split is not cosmetic. Taleo and iCIMS often keep skills in a keyword field while experience lines carry employer, title, and dates. When tools never leave the skills field, the screen that sorts by years of Salesforce may show blank even though the word is on the file.
For a layout that keeps those fields in order, use a Workday-safe single-column format . Two-column templates still send tool names into the wrong column.
This will not get you past a role you are not qualified for. It stops a qualified file from dying because the only proof sat in a footer cloud.
How ATS interprets skills vs tools vs technologies on six real shapes
Each pair is a composite. Copy the After structure, not the product names. Dates and metrics inside bullets are illustrations of good shape, not industry averages.
Marketing coordinator with a skills dump
Before: Skills: communication, teamwork, HubSpot, Mailchimp, Canva, social media, SEO, Google Analytics, content strategy.
After: Skills: stakeholder updates, campaign briefs, A/B test readouts. Tools on bullets: Built HubSpot nurture flows for three product lines, March 2024 to present. Ran Mailchimp tests on 40k-list segments; lifted click-through on the spring launch.
The Before line is a tag cloud. HubSpot in Skills without a dated job is a keyword, not experience. The After line keeps soft skills short and puts each tool on a bullet with dates and an object.
Software engineer with technologies only in the footer
Before: Technologies: Python, JavaScript, React, Node, AWS, Docker, Kubernetes, PostgreSQL, Redis, GraphQL.
After: Software Engineer, FinTech Co., January 2023 to present. Shipped React and Node payment APIs on AWS ECS; cut failed checkout events 9% in Q2 2025. Owned PostgreSQL migrations and Docker builds for three services.
Ten stack terms in a footer tell the parser what words exist. One dated block tells the desk what you shipped. Kubernetes in a list is a tag. Kubernetes in a bullet with a service count is evidence.
Customer support lead mixing tools and soft skills
Before: Skills: empathy, Zendesk, Salesforce Service Cloud, de-escalation, WFM, customer obsession.
After: Support Lead, SaaS vendor, June 2022 to August 2025. Managed Zendesk queues for 18 agents; CSAT 94% over four quarters. Built Salesforce Service Cloud macros that cut handle time 12% on billing tickets.
Empathy and customer obsession are poster words. Zendesk with headcount and CSAT is a lead story. If the tool name is not on a dated line, Greenhouse stores it as a skill tag and the human still asks what you actually ran.
Project manager listing tools without outcomes
Before: Tools: Jira, Confluence, MS Project, Smartsheet, Agile, Scrum, Kanban.
After: Technical Program Manager, Health IT, September 2021 to May 2025. Ran two Scrum teams in Jira across 14 integrations; cleared a regulatory release six weeks early. Confluence runbooks for on-call cut duplicate tickets 15% in 2024.
Agile in a tools row is a methodology tag, not a product. Jira on a bullet with team count and a release object is what a hiring manager can repeat in a phone screen.
Data analyst with Excel buried in a cloud
Before: Skills: data analysis, Excel, SQL, Tableau, Power BI, Python, statistics, attention to detail.
After: Data Analyst, Retail Co., March 2023 to present. SQL and Python pipelines for weekly inventory; Tableau dashboards for 12 district managers. Excel models for promo lift; flagged $2.1M stockout risk before peak season.
Attention to detail is noise. Excel with a dollar object and a season is a story. Spread tools across the jobs where you used them instead of one alphabetical paragraph.
Career changer keeping old tools in Skills only
Before: Skills: teaching, curriculum design, Google Classroom, Canvas, communication, project management.
After: Instructional Designer, EdTech startup, January 2024 to present. Migrated twelve Canvas modules; built Google Classroom pilots for 800 students. Prior role: High School Teacher, District 14, August 2018 to December 2023. Owned curriculum design for AP cohorts.
Teaching as a skill tag does not bridge into corporate L&D. Dated platform work under the new title does. For more on reframing old titles, see how to reframe old experience for a new role .
Copy-paste skills, tools, and technologies block (edit every bracket)
Copy-paste this into a notes doc, then onto your file. Fill brackets from one posting, not from a portal word bank.
Skills (short, capability words only):
- [stakeholder updates] · [financial modeling] · [user research]
[Job title], [Company], [Month Year] to [Month Year or present]
- [Verb] [tool name] for [object]; [metric or scope] in [timeframe]
- [Verb] [technology] to [ship/fix/build]; [result]
[Prior job title], [Company], [Month Year] to [Month Year]
- Repeat only tools/technologies you actually used in that role
Optional Skills footer (posting matches only, no duplicates):
Tools: [2–4 names from posting, each already on a bullet above]
Technologies: [2–4 stack terms, each already on a bullet above]
If a tool is not on a dated bullet, it does not belong in the footer.
Edge case: The posting asks for a tool you used five years ago. Put it on the older job with dates. Do not float it in Skills without a home.
Edge case: You used an internal tool with no public name. Describe the function in the bullet and mirror the posting's category label: CRM, ticketing, BI. Do not invent a vendor.
Pair quantified bullets with resume bullets that define the metric so the number has a label the desk can repeat.
What the weak versions share
Weak files share one habit. They describe categories instead of objects on dated lines. The parser tags the words. The recruiter still has nothing to verify.
If deleting every employer bullet leaves the skills row unchanged, the file is still a tag cloud.
Repeating the same tool in Skills, Summary, and a footer without a job line is keyword stuffing that modern parsers ignore after the first hit. Pick two placements: one bullet with dates, one optional footer match to the posting.
But mixing tools and soft skills in one comma line is worse than a long list. Communication, Salesforce, leadership, Excel reads like a word soup. Split capabilities from product names. Put each product on the job where you used it.
Abbreviation-only stack lines are another parse miss. JS, K8s, and TF in a cloud without JavaScript, Kubernetes, or TensorFlow spelled once forces string matching to guess. Spell the full term on first use.
Tables and icons for skills look clean in PDF and often flatten to garbage in Workday. Plain text bullets in 11-point Calibri survive. Month Year dates. Single column.
Listing every tool from the last decade dilutes the posting match. Recruiters skim for the three required names in context, not for thirty gray keywords. Cut anything you cannot date.
Duty bullets without tool names waste the line where proof should live. Responsible for CRM administration tells the parser nothing product-specific. Administered Salesforce cases for 400 enterprise accounts in 2025 gives a noun and a scope. See why weak bullet points get ignored for the same pattern outside the skills row.
Parse bullets before you add more keywords
A rewritten After line still fails if the checker shows tools only in the skills field. Upload the file you will send and run a free ATS check against the same posting. Confirm each required tool appears inside extracted experience text with dates, not only in a footer cloud.
Fix the bullets first. A longer skills list cannot invent dated tool use.
If the portal asks for a letter, name the same tools you put on bullets. Generate a cover letter that repeats two posting tools with one outcome each. Two tools are enough for this pass. Checker, then letter, then submit.
Put proof on dated lines tonight
That is the file a desk can screen. Skills still matter as tags. Tools and technologies only pull weight when they share a line with an employer, a title, and Month Year dates.
Open the posting. Pick three required tools. Move each one from the footer into the job where you used it. Trim the cloud. Upload the PDF and run the checker once.
Job searching is already hard. You do not need a longer skills list. You need one Greenhouse-readable bullet per required tool. When you are ready to tighten the wording, check your resume for free against the same posting before you send it.
Frequently asked questions
One labeled Skills section is fine if the posting asked for it. What matters is that tools and technologies also appear on dated employer bullets. A footer cloud of product names without dates is where parsers store keywords and humans still see no proof.
The label matters less than placement. Python in a skills row gets tagged. Python in a bullet under a dated software-engineer role gets tied to that job. If the only mention is a comma list at the bottom, most desks treat it as a tag, not as years of use.
Match the posting, then stop. Six to ten relevant names beat thirty outdated ones. Each tool you keep should also show up once under a job with a Month Year date. If you cannot date it, cut it unless the posting names it.
Spell the full term once on the page, then use the short form if the posting does. JavaScript on line one and JS in a bullet is safer than JS alone in a skills cloud. The parser is matching strings, not guessing your shorthand.
Soft skills need a verb and an object in a bullet. Communication in a skills row is a tag. Wrote client-facing release notes for twelve launches in 2025 is a skill with evidence. Mixing both in one comma line blurs what the desk can verify.