Skills Gap Analysis: A Step-by-Step Method With a Free Template

A skills gap analysis compares the skills each role needs with the skills your people have today, scores the difference, and ranks it so you know what to fix first. The method below uses five steps and a 0-4 rating scale, and the table you can copy into a spreadsheet is the template.

The reason leadership keeps asking for one is simple. In the World Economic Forum’s Future of Jobs Report 2025, 63% of employers named skills gaps as a major barrier to business transformation through 2030, and employers expect 39% of existing skill sets to be transformed or become outdated.

This guide is written for an L&D manager or HR business partner who has to put a defensible analysis in front of leadership this quarter. It ends with a worked example for a mid-size company’s L&D and operations team. The numbers in that example are illustrative, not real data.

What is a skills gap analysis, and what is it not?

A skills gap analysis measures the distance between the proficiency a role requires and the proficiency people currently show, skill by skill. It does not explain why a gap exists or whether training is the right fix. It produces a ranked list of gaps that you can act on.

That last point separates it from a needs analysis. A training needs analysis asks why a performance problem exists and whether training will solve it. A skills gap analysis supplies the measured gaps that a needs analysis then investigates. If the difference between the two terms still blurs for you, this comparison of training needs analysis vs training needs assessment sorts it out.

It also differs from a labor market skills shortage. The OECD’s October 2026 report Mapping Skill Shortages in OECD Countries uses its Skills for Jobs database to measure skill imbalances across 38 countries, with shortages most acute in high-skilled occupations such as ICT and health. Your analysis is the company-level version of that picture: it covers the people you already employ, not the hiring market.

What do you need in place before you start?

You need three things before you rate a single skill: a named sponsor, a short list of roles, and a decision date. Without a sponsor, nobody owns the required levels. Without a decision date, the results become a report that sits in a folder.

The sponsor is usually the business leader whose team is being analyzed, not HR. They set the required proficiency, because only they know what the work demands. In LinkedIn’s 2025 Workplace Learning Report, which surveyed 937 L&D and HR professionals, 49% said their executives worry employees lack the skills to execute business strategy. Your sponsor probably already shares that worry, so frame the work as an answer to it.

Limit the first pass to three to five roles. A narrow, finished analysis beats a company-wide one that stalls at step three.

How do you run a skills gap analysis in five steps?

Run the analysis in this order: define the target roles, build the skills list from a taxonomy, rate current against required proficiency, score each gap by business criticality, then decide whether to build, buy or borrow. Each step feeds the next, so do not skip ahead.

Step 1: Which target roles do you define first?

Pick roles where a skills shortfall would hurt a business outcome within the next 12 months. Revenue roles, compliance-critical roles and roles tied to a planned change such as a new system are the usual starting points.

For each role, write one sentence on what a fully competent person delivers. This sentence keeps the skills list honest. If a skill does not help that person deliver the outcome, it does not belong on the list.

Step 2: How do you build a skills list from a taxonomy?

Start from an existing taxonomy rather than a blank page, then trim to six to twelve skills per role. Fewer than six usually misses something. More than twelve makes the rating exercise tedious, and the quality of ratings drops.

Public sources save weeks. O*NET, the US occupational database, lists 25 transferable skills across five categories: complex problem solving, resource management, social, systems and technical skills. Use these for broad capabilities, then add the technical skills specific to your tools and processes. Our guide to designing a skills taxonomy covers granularity and maintenance, including the test of whether you could assign a learning activity to a skill. If you cannot, the skill is too vague to close.

Write each skill as something observable. “LMS reporting configuration” can be rated. “Technical aptitude” cannot.

Step 3: How do you rate current against required proficiency on a 0-4 scale?

Rate every skill twice: the level the role requires and the level each person shows today. Use the same five anchors for both ratings so the numbers compare cleanly. The anchors below describe behavior, not seniority.

Collect ratings from the person and their manager, then reconcile any difference of two or more levels in a short calibration conversation. Where possible, anchor the rating to evidence: a work sample, an assessment result, a certification or a documented project.

Calculate the gap as required level minus current level, and floor it at zero. A person at level 4 in a skill that needs level 2 does not cancel out a colleague at level 1. Surplus skill in one person does not close a gap in another.

Step 4: How do you score each gap by business criticality?

Multiply the average shortfall for each skill by a criticality weight from 1 to 3. The weight is the sponsor’s call and reflects consequence, not difficulty. This step is what turns a list of gaps into a ranking leadership can accept.

Define the three weights before anyone rates anything:

• 3, business critical: a shortfall causes lost revenue, a compliance failure, a safety risk or a missed commitment.
• 2, important: a shortfall slows delivery or lowers quality, but work still ships.
• 1, useful: a shortfall is an inconvenience or a future preference.

Average shortfall is the mean of each person’s gap, counting only people below the requirement. Track a second number beside it: how many people sit below the required level. A skill with a small average gap that affects most of the team is a different problem from one concentrated in a single person.

The priority score is average shortfall times criticality. Set your own band cut-offs after you see the spread. In the example below, 2.5 or higher means act this quarter, 1.5 to 2.4 means plan this year, and below 1.5 means monitor.

Step 5: How do you decide to build, buy or borrow?

Choose build, buy or borrow for each high-priority gap using the decision rules in the section after the worked example. Build means developing current employees, buy means hiring, and borrow means using contractors, vendors or partners for a defined period.

What does a skills gap analysis template look like?

The template is one sheet with a row per skill. Copy the column headers below into a spreadsheet, add one row per skill, and add one rating column per person. The formulas calculate everything else.

Paste these formulas into row 2, then copy them down. They assume six people in columns C to H and work in both Excel and Google Sheets:

I2: =SUMPRODUCT((B2>C2:H2)*(B2-C2:H2))/COUNT(C2:H2)
J2: =COUNTIF(C2:H2,"<"&B2)
L2: =I2*K2
M2: =IF(L2>=2.5,"Act this quarter",IF(L2>=1.5,"Plan this year","Monitor"))

What does a worked skills gap analysis look like for a mid-size L&D team?

The example below analyzes a six-person learning and operations team at a mid-size company against one shared role profile. All ratings, weights and decisions are illustrative, not real data. In practice you would run one profile per role.

Two things stand out. First, weighting changes the order. AI-assisted content production has a smaller raw gap than evaluation design or taxonomy work, but its criticality of 3 lifts it to second place. In this fictional company, leadership wants AI-drafted content reviewed before it reaches learners, so a shortfall carries real consequence.

Second, the averages hide the shape of the problem. Instructional design has an average rating of 2.8, which looks fine, yet two people sit below the requirement. Always read the average shortfall and the people-below count together.

Now the decisions. Each row applies the rules in the next section.

No row says buy. That is deliberate: for every skill, at least one person on the team already holds the required level, so the buy rule below does not trigger.

How do you choose between building, buying and borrowing skills?

Build when someone on the team already holds the required level and can teach it. Buy when nobody does, or when the need is permanent and growing. Borrow when the need is short, urgent or uncertain. Check these three tests in order and take the first one that fits.

The World Economic Forum’s 2025 survey reports two employer responses that map to the first two options: 85% of employers plan to prioritize upskilling their workforce (build) and 70% plan to hire staff with new skills (buy). Borrowing is a third route that those two figures do not cover, yet it is often the fastest way to cover a gap while a build effort ramps up. For the build route, our guide to upskilling vs reskilling vs cross-skilling explains which type of development fits which gap.

How do you present a skills gap analysis to leadership?

Give leadership one page: the top three gaps, what each costs the business if left alone, the decision you recommend for each, and the date you will re-measure. Keep the full sheet as an appendix. Executives need decisions, not 60 rows.

Lead with the consequence in business terms, using the sponsor’s own criticality definitions. State the decision you need from them, such as approving a 90-day contractor or protecting mentor hours. Then show the date you will re-rate the same skills so they can see progress against the same scale.

Once leadership agrees, the next question is how the decisions become a program. Our guide to workforce development programs and L&D strategy picks up from there, so this analysis stops at the decision.

What mistakes make a skills gap analysis useless?

Most failed analyses share five mistakes. Avoid these and your results will survive scrutiny:

• Too many skills: more than twelve per role produces rushed, inflated ratings.
• Required levels set by HR alone: the business leader must own them, or the gaps are opinion.
• Rating without evidence: unchecked self-ratings drift upward and hide the gaps you need to see.
• Offsetting gaps with surplus: a strong colleague does not cover a weak one, so floor each gap at zero.
• No decision or date: a ranking with no owner or re-measure date is a report, not a plan.

What should you do next?

Pick one role family and one sponsor this week. Copy the template, load six to ten skills from your taxonomy, and book a single calibration session to agree the ratings. That is a realistic scope for one quarter.

Set the sponsor’s criticality weights before anyone sees the scores, and write the re-measure date on the first slide. When you re-rate in six months, use the same anchors so the change in average shortfall is a fair measure of whether your build, buy and borrow decisions worked.