This free skills gap analysis shows what to learn next to get the job you want. Tell it where you are and the job you want. See the skills you have, the ones you are missing and how much each matters, then get a week-by-week plan to close the gap.
It fills in the work skills you already use. Leave it empty if you are a student or new to work.
Pick from 910 jobs, or type your own title and add a job post below.
Your plan is saved in this browser. The text of your resume, and what you send to the AI, are saved to ClapAssist so our tools can fill them in for you next time. See or delete.
We gave the tool on this page a made-up resume and two job titles, the way you would. The job now: “Administrative assistant”, which the job data reads as Secretaries and Administrative Assistants, Except Legal, Medical, and Executive. The job wanted: “Project manager”, read as Project Management Specialists. Mid level, 5 hours a week to learn, no job post pasted. Every figure below is what the tool returned from the O*NET 31.0 data. We typed none of it.
56%
14 of 26 skills are already at the level the job asks for. Label: Early days: a longer plan.
Why the label is not “Ready”: the job data says most people in this job have a bachelor’s degree, and this resume shows training, an associate’s degree or years of experience. The report puts that next to the score.
Active Listening, Critical Thinking, Reading Comprehension, Writing, Speaking, Active Learning, Monitoring, Reading people, Coordination, Time Management, Office admin and records, Customer service, Computers and software, Excel.
Most come from the work the current job uses every day. Excel counts because the resume names it.
Persuasion, Microsoft Project, Jira, SAP, Microsoft Visio.
Nothing on the resume shows them, so they count as missing until someone rates them.
The first step the report gives: Microsoft PowerPoint (important), from Basic to Working. The plan to close every gap takes 34 to 60 weeks at 5 hours a week. The long end counts all 5 skills with no level as missing; the short end leaves them out. So the fastest way to a true date is to rate those 5 honestly.
Kim Novak kim.novak@example.com Experience Administrative Assistant | Westbrook Logistics | Mar 2019 - Present - Kept the calendar, travel and meeting notes for a team of 9 managers - Planned the move of a 40-person office over one weekend, on budget and with no lost files - Tracked supplier invoices and purchase orders in Excel and chased late ones every Friday - Wrote the weekly status email that the operations director sends to the whole company - Set up a shared task board in Trello for office requests Education Associate degree in Business Administration, 2018 Skills Microsoft Excel, Microsoft Outlook, Microsoft Word, Trello, Zoom
Three marks, three weights and a score in two parts.
Have means your level already meets the level the job asks for. These skills go on your resume with proof, not on your study list.
Partly means you use the skill, but below the level the job needs. These are often quicker to close than brand-new skills, because you build on work you already do.
Missing means no source gave you a level, or you rated it None. Some are real gaps. Others are things you know but never wrote down. Rate them before you trust the dates.
Each skill carries a weight in the score: 3 for a must-have, 2 for an important skill, 1 for a nice-to-have. So one must-have gap costs as much as three nice-to-have gaps. Close the heavy ones first.
The score has two parts. The tools part covers the software, tools and hard skills the job names. The work skills part covers how you work and what you know: speaking, solving problems, the subjects the work draws on. Readiness is the average of the two parts, and the label follows the weaker one. Most people who have worked for a few years already have many work skills, so a strong work part cannot hide missing tools.
Readiness measures one thing: how much of the skill list you meet, weighted by importance, tools and work skills apart. It is not your chance of getting hired. It cannot see your interview skills, your network or how many people apply for the same job. Use it to choose what to learn, not to judge yourself.
Turn the report into proof an employer can see.
And the better way for each.
Pick from 910 jobs, from nurse and electrician to teacher, accountant and developer. Each comes with the work skills, knowledge and tools it uses and how important each one is, from the O*NET 31.0 Database of the U.S. Department of Labor. Paste real job posts and their skills are added and counted.
Your resume does most of the work: every job on it is read as its job in the data, with the level that job builds; your degree fills in what you studied; the tools you name count. One AI button rates the rest from your resume, and you can change any level in one click. You see a readiness score, how many skills match, and where each of your levels comes from.
The job’s tools first, then the must-haves, then the important skills, then the everyday work skills you practice as you go, the extras, and proof and applying. Each step has free courses and videos, a practice task from the job itself, and the line to put on your resume. Mark steps done, learning or skipped, and add the plan to your calendar.
Your levels are compared with all 910 jobs, at your level or one above. Jobs O*NET itself lists as related to yours are marked. See how alike the work is, how ready you are, the skills you share and the few to build, and set any of them as your target in one click.
Our own rules, printed in full. Nothing here is a guess dressed up as data.
| Source | Must have | Important | Nice to have |
|---|---|---|---|
| Work skills and knowledge (O*NET importance, 1 to 5) | 4 or more | 3.25 or more | 2.75 or more |
| Tools in the job data | – | “In demand” for this job, up to six, most important first (see below) | “Hot” across job posts and named by related jobs too, or any tool when you give job posts |
| Skills in your job posts | In half or more of the posts, as a requirement | Asked for in fewer posts | Only in nice-to-have parts |
Which tools come first: for the most searched jobs (data scientist, software developer, data analyst, web developer and a few more) a list we checked by hand, and for the everyday name you picked (“AI engineer” puts PyTorch before SAS). For every other job: 3 points if the job data marks the tool “in demand” for this job, plus 2.5 × the share of related jobs that name it, plus 1.5 × how much more often related jobs name it than all jobs do (Excel is everywhere; Tableau is a data tool). A short list or a list of skills in the job post box counts each skill as asked for. A choice in a post (“Tableau or Power BI”, “Pinecone, Weaviate or FAISS”) is one requirement that any one of them meets.
None, Basic, Working, Advanced. The level a job needs comes from O*NET’s level rating (0 to 7): under 2.75 is Basic, under 4.5 is Working, higher is Advanced. Tools asked for need Working. The level you choose changes it: entry level needs nothing above Working and nice-to-haves only at Basic; senior needs the must-haves at Advanced and the important ones at least at Working; mid level is the job data as it is.
Your level is the first of these that applies: your own rating; the AI’s reading of your resume (the button, if you press it); then the highest of: a work skill your current job uses (its level); each job on your resume, read as its job in the data (the job’s level after two years in it, one level lower before that, internships included); your degree (its main subject is Working for a bachelor’s and Advanced for a master’s or doctorate, nearby subjects one level lower, an earlier or lower degree one level lower; any degree gives reading, writing and learning skills); the tools your resume names (a tool is Working; Programming and Computers rise with the number of tools named); a line in your resume that shows a work skill (Basic, Working when three or more lines show it). Every estimate says where it came from, and your rating always wins.
| Importance | Weight |
|---|---|
| Must have | 3 |
| Important | 2 |
| Nice to have | 1 |
Readiness has two parts, scored apart: the tools (the software, tools and hard skills the job data or your posts name) and the work skills (work skills and knowledge areas). Each part = the sum of weight × (your level ÷ level needed, at most 1), divided by the sum of the weights. Readiness = the average of the two parts, and the label (Ready to apply from 80%, Close from 60%, Some way to go from 40%, Early days below) follows the weaker part. A part with fewer than two skills counts together with the other, as one list. A skill with no level from any source counts as none until you rate it. A job whose usual preparation is above what your resume shows (O*NET job zone: a master’s degree against your bachelor’s, for example), or a post that asks for a higher degree, is never marked “Ready to apply”: the report says what is missing.
How alike the work is: the cosine similarity between your levels and each job’s levels on the 35 work skills and 33 knowledge areas, each one centered on its average over all 910 jobs (so reading, speaking and critical thinking, which nearly every job needs, do not make a nurse look like a physicist), with a job’s level weighted by how important it is there; plus a share for the job’s tools you already use (85/15, or 75/25 when your resume names three tools or more). Single-line keyword hints stay out of it. Jobs O*NET itself lists as related to your current job or the jobs on your resume come first, marked, when their work is at least 25% like yours; then the rest by how alike the work is plus 0.6 × how alike the job is to the job you want, so the list leads toward your target (“Work much like yours” from 45%). The list shows your level (job zone: from the work you have done, or your degree) up to one above it or up to the level your degree opens; with a degree and no work yet, from one level under the degree up to it. You can choose otherwise. The number on each card is the readiness rule above on that job’s own list. The list needs at least four known levels.
The job’s tools first (most asked for first), then the must-have knowledge and hard skills, then the important ones, then the everyday work skills such as speaking, writing and problem solving, which you practice while you learn the rest, then the nice-to-haves, then four steps to prove it and apply. Inside each group, the heavier skill comes first.
O*NET has no product manager, UX researcher, AI engineer or DevOps engineer. For these titles the tools come from real job posts: we read the posts for the title on the public company job boards our job finder uses, counted how many of the companies name each tool or skill, and keep the ones named by 15% or more of them (must have from half, important from a quarter). Words that name a company’s industry or product (SaaS, fintech) and hiring boilerplate (recruitment, benefits) are left out. The work skills come from the closest O*NET job, named on the report with the reason. The date the posts were read is on every skill.
“About the job” shows US figures from the U.S. Bureau of Labor Statistics, to everyone, marked as US data: the median yearly pay and the range most people earn (OEWS, May 2025), and the change in the number of jobs from 2025 to 2035 with the openings each year (Employment Projections 2025–35, released 27 Aug 2026). When the job data is more detailed than the BLS, the figures are for the wider group, and the page names it. We have no official pay figures for other countries, so we show none.
A starting guess you can change on every step: levels to climb × 15 hours for a tool, 20 for a work skill, 25 for a knowledge area. Steps run one after another at your hours a week; skipped steps drop out. A skill not rated yet counts as missing, so the roadmap shows a range (with and without those skills) until you rate them; past two years it says so in words instead of a far-off date.
Five steps, whatever the job.
First the tools the job you want names and you do not have yet, then the must-have knowledge and skills you are below the needed level in. That is the order the roadmap uses, from job data for 910 jobs and the real job posts you paste, so you spend your hours where they move you closest to an offer.
No. The analysis starts from what you already know, counts the skills your past jobs and degree gave you, and shows the real distance to the new job as a readiness score. Most changes are shorter than they look once the shared skills are counted.
It compares the skills a job needs with the skills you have, so you know what to learn next and in what order. This one does it for any of 910 jobs, or any job post you paste, and turns the gap into a roadmap with free learning links.
Write down the job you want and read two or three real posts for it. List the skills they ask for, then mark each one: have, partly or missing, with proof for every have. This page does the listing and the marking for you from your resume and the job data, and you correct any level in one click.
The score has two parts: the tools the job names, and the work skills and knowledge it needs. Readiness is the average of the two, and the label follows the weaker part, so missing most of the job's tools never reads as close. 80% or more on the weaker part is marked "Ready to apply". 60% to 79% is "Close: a few gaps to close", 40% to 59% is "Some way to go", and under 40% is "Early days: a longer plan". When the job asks for more schooling than your resume shows, 60% or more reads "Skills close: it needs more study first" instead. Apply widely once both parts pass 80%.
Each time you finish a group of steps, and each time you aim at a new job post. Your levels change as you learn, and posts change too. Your plans stay saved in this browser, so you open the plan, update your levels and see the new score.
It is built for one person at a time. For a team, each person runs their own check against the same job or post and shares the PDF or text download with you. There is no team dashboard.
Yes, with no account and no limit. The analysis, the roadmap, the calendar file and the downloads are free. The AI buttons have a small free daily allowance per connection; everything else works without them.
From the O*NET 31.0 Database of the U.S. Department of Labor (August 2026 release), used under the CC BY 4.0 license. It rates how important each work skill and knowledge area is for 910 jobs, based on surveys of people doing the work, and lists the tools the job uses. The jobs and skills are the same in most countries; add a job post from your own country to see what employers there ask for.
Type your own title and paste one to three real job posts for it: their skills become the list. Or press "List the skills with AI" to get the skills the job usually needs, then check them against a real post. Product manager, UX researcher, AI engineer and DevOps engineer are not in the job data either, so for them the tools come from real job posts we counted, dated on every skill.
Your resume shows the tools and skills you name (we read it in your browser with the same reader as our resume scanner). Your current job fills in the work skills that job uses every day. Everything else you rate yourself: None, Basic, Working or Advanced. Each level shows where it came from, and your rating always wins.
The file is read in your browser, and your plans are saved in this browser, with a delete button and a backup file you can download. The text of your resume, and what you send to the AI, are saved to ClapAssist so our tools can fill them in for you next time; see or delete them at clapassist.com/my-data/. Only the AI buttons send text, and each one says what it sends.
The roadmap adds up hours per step: levels to climb times 15 hours for a tool, 20 for a work skill and 25 for a knowledge area. That is our starting guess, not a promise; change the hours on any step and the dates move. At 6 hours a week, 120 hours is about 20 weeks.
The links go to each site's own search for that skill: YouTube, Coursera and edX (most courses can be audited for free), Khan Academy, Microsoft Learn, freeCodeCamp, MIT OpenCourseWare, OpenStax and Alison; SWAYAM in India, TESDA in the Philippines and OpenLearn in the UK. We have no deal with any of them and earn nothing from the links.
Yes, as US data, on the About the job tab: the median yearly pay and the range most people earn, from the U.S. Bureau of Labor Statistics (May 2025), and how the number of jobs is expected to change from 2025 to 2035. We show the US figures to everyone and label them, because pay elsewhere is often very different and we have no official figures for other countries. The same tab links to open jobs for the title in our job finder.
ClapAssist is a Mac and Windows app that listens to the interview and shows you what to say, so your new skills come across when it counts.
Download with 10 free minutesThis page includes information from the O*NET 31.0 Database by the U.S. Department of Labor, Employment and Training Administration (USDOL/ETA). Used under the CC BY 4.0 license. ClapAssist has modified all or some of this information. USDOL/ETA has not approved, endorsed, or tested these modifications. Learning links go to each site’s own search; we have no deal with any of them.