The Skill Gap Diagnostic: Find Out What You Actually Need to Learn Before the Next Role
Why this prompt matters
Professionals routinely spend months on generic upskilling — a Python course here, a stats refresher there — without checking whether any of it maps to what a specific target role actually requires. A LinkedIn Learning survey found the average employee spends 24 minutes a week on self-directed learning, but without a gap analysis tied to a real target, much of that time goes toward skills that don't move the needle on the next promotion or job application, and critical gaps go unaddressed until they surface in an interview.
What we use it for
A data analyst with 4 years of SQL experience is applying for a senior analyst role that requires Python ETL pipelines and A/B testing design — two skills the job posting lists as core requirements but that aren't in their current toolkit. They have 5 hours a week and 90 days before they plan to apply.
Prompt
Role: You are a skills-development coach who has reviewed thousands of job postings and helped professionals close specific, measurable skill gaps rather than chase vague self-improvement. Context: I currently work as [YOUR CURRENT ROLE] with [X YEARS] of experience. My current skill set includes: [LIST YOUR KEY SKILLS, TOOLS, AND EXPERIENCE]. I am targeting this specific role or opportunity: [PASTE JOB POSTING, PROMOTION CRITERIA, OR TARGET ROLE DESCRIPTION]. I have [AVAILABLE TIME PER WEEK, e.g., 5 hours] to invest in closing gaps over the next [TIMEFRAME, e.g., 90 days]. Task: Compare my current skill set against the target role's actual requirements. Produce a gap analysis that separates what I already have from what I'm missing, then build a time-boxed learning plan to close the highest-priority gaps within my stated timeframe and weekly time budget. Constraints: - Do not list generic advice like "improve communication skills" — every gap must be specific and tied to something in the target role description. - Rank gaps by how much they would actually block me from getting or succeeding in the role, not by how interesting they are to learn. - Respect my stated weekly time budget. If the gaps can't realistically close in the timeframe given that budget, say so explicitly and tell me which gaps to defer. - Distinguish between gaps I can close through self-study versus gaps that require hands-on practice, a mentor, or real project experience — self-study alone won't fix all of them. - Do not recommend paid courses or certifications by brand name unless I specifically ask; prioritize free or already-accessible resources first. Output format: 1. **Strengths That Transfer** — skills I already have that map directly to the target role, with the specific requirement each one satisfies. 2. **Critical Gaps** (ranked) — for each: what's missing, why it matters for this specific role, and whether it's closable via self-study or requires practice/mentorship. 3. **Nice-to-Haves** — lower-priority gaps worth noting but not blocking. 4. **[TIMEFRAME] Learning Plan** — week-by-week or milestone-by-milestone breakdown that fits my stated weekly time budget, with a concrete, checkable deliverable at each milestone (not just "study X"). 5. **Reality Check** — one paragraph on whether this timeline is realistic given the gaps found, and what to deprioritize if time runs short.
Result
**Strengths That Transfer** - 4 years of SQL and dashboarding (Looker) directly satisfies the posting's "proficiency in BI tooling" requirement — no gap here. - Experience presenting metrics to non-technical stakeholders maps to "communicate insights to leadership," listed as a core responsibility.
**Critical Gaps (ranked)** 1. **Python for data pipelines (blocking).** The posting requires "building and maintaining ETL pipelines in Python," and your experience is SQL-only. This is self-study closable — Python's pandas/SQL overlap is substantial, so this isn't starting from zero, but it needs deliberate practice, not just reading. 2. **A/B testing statistical design (blocking).** The role owns experimentation strategy; you've only consumed test results, not designed them. This requires some self-study (power analysis, sample size calculation) but the sequencing and pitfalls are best learned by running a real test — pure reading won't be convincing in an interview. 3. **Stakeholder-level experiment readouts (moderate).** You present dashboards well, but translating a null or ambiguous A/B result into a confident recommendation is a distinct skill. Closable through deliberate practice: write 3 mock experiment readouts this month using public datasets.
**Nice-to-Haves** - Familiarity with Airflow (mentioned once in the posting, not emphasized) — defer unless time allows after the above three.
**90-Day Learning Plan** - Weeks 1–3 (5 hrs/week): Python fundamentals focused on pandas, not general syntax tutorials. Deliverable: rebuild one of your existing SQL reports as a Python script. - Weeks 4–6: A/B testing design — power analysis, minimum detectable effect, common pitfalls. Deliverable: design (not run) a full test plan for a hypothetical feature change at your current company. - Weeks 7–9: Run an actual small experiment if your current role permits, or use a public dataset to simulate one end-to-end. Deliverable: one complete experiment readout, written as if presenting to leadership. - Weeks 10–12: Buffer and portfolio polish. Combine the Python script and experiment readout into two concrete work samples you can reference in interviews.
**Reality Check** Given a 5-hour weekly budget, this plan is realistic but tight — the A/B testing gap is the riskiest because it depends on having something real to test, which you don't fully control. If you can't access a live experiment by week 7, substitute a simulated analysis using a public A/B testing dataset (several exist on Kaggle) rather than letting the plan stall. Defer Airflow entirely; it's mentioned once and isn't worth the time against two blocking gaps.
Most upskilling fails for a boring reason: it isn't tied to anything specific. People take a Python course, a leadership workshop, a stats refresher — all reasonable skills in the abstract, but with no guarantee any of it maps to what a real target role requires. This prompt fixes that by forcing a direct comparison between what you have and what a specific posting or promotion criteria actually asks for.
Why each section of the prompt exists
The Role frames the AI as a coach who's seen thousands of job postings — this matters because it steers the model away from motivational-poster advice and toward requirement-matching, which is a more mechanical, more useful task.
The Context section forces three specific inputs: your current skills, the target role's actual text (not a vague description of it), and your real time budget. Pasting the actual job posting text, rather than summarizing it yourself, matters — summarizing introduces your own blind spots about what's actually required.
The Constraints do the most work in this prompt. Banning generic advice (“improve communication”) forces every gap to trace back to specific text in the target role. Ranking by blocking severity rather than interest keeps the plan honest — the gap you find most interesting to learn is not necessarily the one standing between you and the role. And forcing a distinction between self-study-closable gaps and practice-requiring gaps prevents the common mistake of thinking that reading about A/B testing design is the same as having designed one.
What makes the output format work
The five-section structure — strengths, critical gaps, nice-to-haves, timeline, reality check — mirrors how a good manager or mentor would actually structure feedback: lead with what's working, rank what's missing by how much it matters, and close with an honest assessment of whether the plan is achievable. The Reality Check section is the one most prompts skip, and it's the one that keeps this from being generic motivational output — a model that always says “yes, you can do this in 90 days” isn't useful. One that says “this specific gap is the risky one, here's your fallback” is.
How to reuse this well
Re-run this prompt every time you're evaluating a new role, not just once a year. The target role field is meant to change constantly — paste a different job posting, promotion rubric, or even a performance review's stated expectations, and get a fresh, specific gap analysis each time rather than a generic learning-and-development plan that doesn't map to any particular opportunity.