Skills Have a Half-Life

Keep what’s tested and true—refresh the steps that carry it.

I work with a lot of sharp people. They can take a tangle of facts and turn it into something you can act on. That craft still amazes me.

But I’ve also sat in rooms where those same pros hit a wall with today’s rails—platforms, creator culture, and now AI. Someone tries a giant, one-shot prompt—“do it all”—and walks away thinking the model isn’t any good. From my seat (middle-aged Gen Xer, lifelong tech nerd who still enjoys the evening weather report), I see something else: the work wasn’t broken into steps the machine can handle.

Here’s the simple backbone:

Knowledge builds; skills wither if you don’t tend them.

Keep chasing knowledge and judgment—they stack. Keep skills current—update, upgrade, or swap them when they stop pulling their weight.

Why “half-life” belongs in your career playbook

Skills don’t die, but they do lose value if you leave them alone—especially the ones tied to fast-moving tools and platforms. Think of them like a battery: strong at first, then slower, then not enough to power the same jobs. The fix isn’t to ditch what’s tried-and-true—it’s to carry it into new places with updated steps.

Knowledge vs. skills (plain and fast)

Knowledge (the “why” that lasts): principles, mental models, ethics, story sense.
Skills (the “how” that needs upkeep): the steps you run—usually tied to a tool, format, or channel.

Quick test:

If it depends on where you click or what a platform allows, it’s a skill. If it explains why something works no matter the tool, it’s knowledge.

What “skills wither” looks like

You still know the why, but you can’t ship the how—the steps changed.
Old muscle memory keeps choosing the slower route.
Risk shows up quietly: missed speed, missed reach, missed chances.

The meta skill that saves your skills

When AI disappoints, the request usually hides six or seven smaller jobs. Use this five-card pattern (it works for reports, emails, code, SOPs—almost anything):

Inputs — files, rules, examples, constraints.
Steps — 5–8 small moves, in order.
Output format — exact fields/table/bullets.
Quality checks — what “good” looks like; traps to avoid.
Next action — what you’ll do with the result.

Run the cards one at a time. You’re not hoping for magic; you’re showing the tool how to help.

Adapt & repurpose: tested truths, new rails (quick snapshots)

Healthcare: keep clear patient education; update to short-form explainers and portal messages; draft with AI → review in plain language.
Manufacturing & ops: keep quality and safety basics; add sensor checks, AI shift summaries, and one-page SOPs the model can read back.
Retail & services: keep offer, proof, follow-through; add local search hygiene, creator collabs, and inventory sync with marketplaces.
Education & training: keep learning objectives; add micro-lessons, open-book vs. closed-book designs that assume AI exists.
Small business: keep cash flow, customer care, simple KPIs; add bookkeeping automations, one-click pay, and SMS follow-ups.

A 30-minute setup that pays back every week

List your 10 real skills. Next to each: last updated and next update due.
Sort by decay speed.

Fast (≈2–3 yrs): ad platforms, SEO/ASO, social algorithms, genAI tools, niche code packages.
Slower (≈5–7 yrs): safety/quality basics, finance controls, stakeholder management, story craft.

Put tiny reps on the calendar.

Weekly (15 min): skim one tool’s “What’s new.”
Monthly (45–60 min): hands-on test of one new feature/workflow.
Quarterly (2 hrs): adapt one old step; retire one stale step.

Save one “before/after.” Six bullets for the old way; six for the new way. That folder becomes your living playbook.

Where generations actually help each other

Boomers / older Gen X: strongest on judgment and story; benefit most from short, regular tool refresh.
Millennials: social-mobile-cloud translators—great bridge builders.
Gen Z: fast with tooling and distribution; levels up fastest with mentoring on standards and stakeholder politics.

BSAS sidebar: Portfolio Life + Half-Life

Short skill cycles reward stackable income streams. Keep one main lane, then maintain a small set of “side skills” you can rent out: editing short videos, building simple automations, setting up email flows, proofreading with AI. Date those skills like you date food. Rotate them before they go stale.

Bottom line: keep what’s tested and true—refresh the steps that carry it.