Your supervisor program ran well, the scores were strong, and then a board member asked the only question that matters: what changed in the business? If the answer is a shrug, you do not have a measurement problem — you have a document problem. A training case study is the artifact that answers that question with before numbers, after numbers and dollars, in one page a CFO will actually read.
The context is a market that spends heavily and proves little. US organizations spent $102.8 billion on employee training in 2023 — an average of $874 per learner (Training magazine, 2023 Training Industry Report) — yet most of that spend is reported back as completion rates. Meanwhile 94% of employees say they would stay longer at a company that invests in their learning (LinkedIn Learning, 2018), and managers account for 70% of the variance in team engagement (Gallup, 2015) — two levers a good case study can make visible.
This guide gives you the five-part training case study template we use, the metrics that survive finance scrutiny, the conversion math that turns results into dollars, and the honesty rules that keep the document believable.
Key Highlights
- A training case study is evidence with a plot: baseline, intervention, after numbers, dollars and one transferable lesson.
- US training spend hit $102.8 billion in 2023, an average of $874 per learner (Training magazine, 2023).
- 94% of employees would stay longer where employers invest in their learning (LinkedIn Learning, 2018).
- Managers drive 70% of the variance in team engagement (Gallup, 2015) — the behavior story behind most good cases.
- The worked example below converts a $14,500 program into $70,000 of measured annual benefits (illustrative).
- Raw ROI was 383%; we would defend 141% to a hostile CFO after a 50% attribution haircut (illustrative).
- People lose about 50% of new material within 24 hours without reinforcement (Ebbinghaus, commonly cited) — so results are measured weeks later, not on day one.
What a Training Case Study Must Prove
A training case study is a short, documented story with four working parts: a baseline taken before the program, an intervention described plainly, after numbers pulled from records, and a lesson the next team can use. It is not a testimonial, and it is not a brochure. The difference is that every claim in it can be traced to a system someone else can open.
Different readers want different things from the same document, which is why the strongest cases answer four questions at once. The table below is the audience map worth keeping next to your draft.
| Reader | The question they ask | The element that answers it |
|---|---|---|
| CFO | What did it return, and when? | Results converted to dollars, with a dated window |
| HR director | Can we repeat it elsewhere? | Intervention detail plus the context match |
| Front-line manager | Will it work for my team? | Baseline numbers that look like their team's |
| Skeptic | Is this real or marketing? | The isolation note and the assumption list |
The element most case studies skip is the baseline, and it is the one that carries the most weight. Without a before number, the after number is an anecdote. The chart below shows the shape a credible metric takes when the baseline was captured early enough to matter.
Key takeaway: write the case study for the skeptic, not the fan — every claim traced to a record is what separates evidence from marketing.
The Training Case Study Template: Five Parts That Do the Work
The template has five parts, and the order matters less than the discipline of each one. Part one is the title and context: company profile, industry, team size and the moment that triggered the program. Part two is the gap in numbers — two or three baseline metrics and what they were costing. Part three is the intervention: format, length, who attended and what it covered, in plain words rather than vendor language.
Part four is the results section: before and after numbers, the measurement window, and the isolation note. Part five is the lesson — one transferable rule stated in a single sentence. The table below shows the word budgets and the failure mode of each part.
| Part | Word budget | Must contain | Common mistake |
|---|---|---|---|
| Title and context | ~120 | Company profile, size, trigger moment | Vague "leading manufacturer" with no size |
| The gap in numbers | ~100 | Baseline metrics and their cost | Skipping straight to the program |
| The intervention | ~150 | Format, length, attendees, coverage | Feature-listing the vendor |
| The results | ~220 | Before and after numbers, window, isolation note | Day-one survey scores as results |
| The lesson | ~80 | One transferable rule | Sliding into sales copy |
Draft the results section first, because it tells you whether you have a case at all. If the records do not support a before-and-after pair, no amount of context prose will save the document — go collect the data and write the case next quarter instead.
Key takeaway: write results first and context second — a case without records is a story, and stories do not renew budgets.
Choosing Metrics That Survive Finance
Pick two to four metrics that already exist in a system, and reject anything that needs a new survey to defend. Quality and error rates live in operations records, output and productivity in production or CRM data, turnover in the HRIS, customer metrics in support logs, and behavior in manager observation checklists. Each metric also has its own honest re-measurement window, because some results take months to appear.
| Metric | Where the data lives | Re-measure window |
|---|---|---|
| Error or quality rates | Operations and QA records | 30–90 days |
| Output or productivity | Production, sales or CRM systems | 60–90 days |
| Regrettable turnover | HRIS exit records | 6–12 months |
| Customer satisfaction | Support tickets and surveys | 60–90 days |
| Behavior on the job | Manager observation checklists | 30–60 days |
The worked example that runs through this guide is a 120-person logistics firm that ran a $14,500 fully loaded program for twelve front-line leads (illustrative). Over six months, order accuracy moved from 97.8% to 99.1%, on-time delivery from 84% to 93%, and regrettable turnover from 18% to 11% (illustrative). Those are the kinds of ranges finance can check against its own reports.
Measure late, not early. Because the commonly cited Ebbinghaus finding puts about 50% of new material at risk within 24 hours without reinforcement, day-one quiz scores tell you nothing about what stuck — the metric windows in the table exist for exactly that reason. Our L&D metrics guide goes deeper on choosing and tracking the numbers.
Key takeaway: two to four metrics from existing systems beat ten metrics from new surveys — finance trusts what it can pull itself.
The Money Page: Converting Results into Dollars
The money page is where a case study stops being a story and becomes a budget document. Each metric converts to a dollar line using rates finance already accepts: replacement cost for avoided exits, loaded hourly rates for time saved, overtime rates for schedule relief. The arithmetic is simple; the discipline is showing the rate you used next to every line.
In the worked example, three buckets emerged. Avoided regrettable turnover produced $40,000 — two supervisor exits prevented at a $20,000 replacement cost each (illustrative). Reduced rework produced $19,000 — fourteen hours a week saved across forty-eight weeks at a $28 loaded rate (illustrative). Overtime relief added $11,000, for a measured total of $70,000 against the $14,500 program cost.
The raw arithmetic gives ($70,000 − $14,500) ÷ $14,500 × 100, or roughly 383% — and you should not report that number. Apply an attribution haircut: if half the credit plausibly belongs to a new dispatch system, report $35,000 in benefits and an ROI near 141% (illustrative). The defensible number renews budgets; the inflated one gets you disinvited from the next budget conversation. The conversion mechanics live in our training ROI formula walkthrough.
Key takeaway: show the rate and the haircut next to every dollar line — a case study that survives a hostile read is the only kind worth writing.
Isolation, Honesty and Getting the Story Read
Isolation is the difference between "we improved" and "the program improved us," and the cleanest version is a comparison group. If one team ran the program and a similar team did not, plot both over the same window — the gap between the lines is the honest size of the effect. The illustrative chart below shows what that looks like for a quality audit score.
When a comparison group does not exist, use the honesty substitutes: the trend line before the program, a confidence-adjusted participant estimate, or a stated contributing-factors list. Print the measurement window, the sample size and the assumption list in the document itself. In our work with a 90-seat call center, the case that renewed the budget was one page with three numbers and a visible haircut — the full two-page version sat in the appendix as backup (our observation).
Honesty has a commercial payoff beyond finance. 94% of employees say they would stay longer where employers invest in their learning (LinkedIn Learning, 2018), and a credible internal case study is itself evidence of that investment — which is why the finished document belongs in onboarding, leadership meetings and recruiter conversations, not just the budget binder. Package the one-pager for finance, a three-line version for decks, and keep the full case as the audit trail; our guide on how to prove training value covers the packaging.
Key takeaway: a visible haircut and a named method make the case stronger, not weaker — trust is the product you are shipping.
What Your Business Should Do
Days 1–30: Rebuild the result you already have.
- List the last three programs you ran and the metrics each could plausibly have moved.
- Pull baseline numbers for the strongest candidate from records you already keep.
- Interview one manager for the context paragraph and one quote you can verify.
Days 31–60: Draft the case in the right order.
- Write the results section first, with before and after numbers and a dated window.
- Convert two benefit streams to dollars using loaded rates and replacement costs.
- Add the isolation note and the assumption list before anyone sees the headline number.
Days 61–90: Publish, trim and reuse.
- Cut the full case to a one-page version for finance and a three-line version for decks.
- Verify every number with the person who owns the source system.
- Schedule the next baseline pull now so the following case writes itself.
Key takeaway: treat the case study as a system, not a one-off — each baseline you pull today is the before number of next year's document.
The Bottom Line
A training case study earns its keep when it carries a baseline, an after number, a dollar figure and a visible haircut — the worked example above turned a $14,500 program into $70,000 of measured benefits, reported at a defensible 141% rather than a raw 383% (illustrative). In a market spending $102.8 billion a year (Training magazine, 2023), the documented result is the differentiator. Write results first, convert second, cut last. If you want this mapped to your own team, book a free strategy call or start with the free resource library.
Work With Zohaib Butt
1. Free Resource Library — Zohaibbutt.com/store
World-class training and coaching resources, all in one place. Download the assessments, templates and playbooks you need today.
2. Online Courses — learning.zohaibbutt.com
Self-paced leadership and coaching programs for managers and owners who want structure, not scattered tips.
3. Free Strategy Session — funnelsthatsell.com
Book your complimentary strategy call and we will map your biggest people-development bottleneck to a 90-day fix.
Free assessment pack (PDF)
Turn one program's results into a documented, finance-ready case study.
Download the PDF packFAQ
How long should a training case study be?
A working case runs 600–800 words, which is roughly one page: about 120 words of context, 100 for the gap, 150 for the intervention, 220 for results and 80 for the lesson (illustrative template). Cut a one-page version for finance and a three-line version for decks from the same draft. Length is not the point — traceable numbers are.
What is the difference between a case study and a testimonial?
A testimonial is an opinion with a name attached; a case study is a before number, an intervention, an after number and a stated method. Testimonials say the training was great, while case studies show order accuracy moving from 97.8% to 99.1% over six months (illustrative). If your document has no baseline and no isolation note, it is a testimonial wearing a case study's clothes.
Which metrics belong in a training case study?
Pick two to four metrics that already live in a system: error or quality rates, output, regrettable turnover, customer satisfaction or observed behavior. Each needs a baseline captured before the program and an honest re-measure window — quality at 30–90 days, turnover at 6–12 months. New surveys built after the fact are the weakest source, so avoid them when a record exists.
How many case studies do we need?
One per strategic program per year is enough for most US mid-market teams (illustrative, based on our experience). Depth beats volume: a single case with converted dollars, a comparison group and a visible haircut persuades more than five anecdote collections. Start with the program whose renewal depends on the answer, and let the baseline you pull this quarter become next year's case.
