All posts

Blog

Pre/Post Training Measurement: Templates and Real Examples

November 7, 2026 · By Zohaib Butt

Pre/Post Training Measurement: Templates and Real Examples

The CFO signs the invoice — a 30-person program costs about $26,220 at the national average of $874 per learner (Training magazine, 2023) — and eight weeks later wants one thing: evidence that people changed. A pre and post training assessment produces that evidence, because it measures the same people twice on the same items and reports the delta instead of the anecdote. It costs a spreadsheet, not a research project.

Most teams skip the "pre" half. They collect a satisfaction survey at the end, compare nothing, and then wonder why only 13% of employees are engaged at work globally (Gallup, 2023) while the training line keeps growing. The commonly cited Ebbinghaus forgetting curve adds the second problem: people lose about 50% of new material within 24 hours without reinforcement, so a single post-course test flatters whatever anyone retained.

This guide hands you the method in one pass: what to measure before and after, the four tracker columns, a worked example with numbers, and the traps that make before-and-after data lie.

Key Highlights

  • A pre and post training assessment measures the same people twice — knowledge, confidence, behavior and one business metric — and reports the delta.
  • The average US learner costs $874 a year (Training magazine, 2023); a 30-person program is roughly a $26,220 bet.
  • About 50% of new material fades within 24 hours without reinforcement (Ebbinghaus, commonly cited) — post-only tests flatter the truth.
  • 71% of organizations offer leadership development (Training magazine, 2023), yet far fewer measure behavior or results.
  • In our illustrative cohort example, knowledge rose from 58% to 84% while observed application rose from 30% to 65%.
  • First-contact resolution in the example moved from 61% to 74% — the business metric is the line the CFO reads.
  • You need four spreadsheet columns per learner and identical test items, not a data team.

Pre and Post Training Assessment: The Method in One Pass

The method is a paired design: capture the same measures before the program and again after, then report the change per person and per cohort. Four layers are worth capturing — knowledge (a short test), confidence (a rating scale), behavior (a manager rating against a rubric) and one business metric the skill touches. The pairing is the point; a post-only number has nothing to argue against.

Kirkpatrick's four levels give the capture plan its shape — reaction, learning, behavior, results (the level definitions live in our Kirkpatrick model guide). Reaction is the only level that lives entirely after the program; the other three get stronger when you baseline them. The table below is the version worth printing.

Kirkpatrick level What you capture before What you capture after
1. Reaction Nothing — it is post-only Survey within 48 hours
2. Learning Same test you will re-run Identical items, within 48 hours
3. Behavior Manager baseline rating on a rubric Manager rating at 30 and 60 days
4. Results The business metric, dated The metric re-pulled at 60–90 days

Key takeaway: measure the same people on the same items twice — the delta is the evidence, and the anecdotes are decoration.

The Stakes: What an Unmeasured Learner Costs

Anchor the effort in money and the discipline stops feeling optional. The national average is $874 per learner (Training magazine, 2023), and typical 2026 US ranges we see quoted (illustrative) run $50–$150 per person per year for off-the-shelf e-learning, $200–$400 a head for a one-day custom workshop, $800–$1,600 per person for a multi-week cohort, and $3,000–$6,000 for a three-month coaching engagement. The midpoints sit below; every one of those dollars is unprovable without a before-and-after pair.

Hbar chart of typical US training cost per learner by format, using range midpoints: e-learning $100, one-day workshop $300, cohort program $1,200 and three-month coaching $4,500. Illustrative 2026 US market ranges.

The stakes scale with format. An e-learning seat at the midpoint is a small bet you can afford to leave unmeasured; a 25-person cohort at $1,200 a head is $30,000 that deserves a delta. The larger the per-learner number, the harder the CFO will look for the before-and-after evidence behind it.

Key takeaway: an unmeasured learner is an unauditable purchase — the before-and-after pair is what turns spend into evidence.

Pre and Post Training Assessment Beats Post-Only Surveys

A post-only test answers the wrong question: it shows what people knew on exit day, not what they retained or whether they knew it before you taught them. Without a pre-test you cannot tell a genuine gain from prior knowledge, and without a second metric pull you cannot tell training impact from a good quarter. Decay makes the flattering version worse — the commonly cited Ebbinghaus curve holds that people lose about 50% of new material within 24 hours without reinforcement.

Line chart of why a pre and post training assessment beats a post-only survey: retention without reinforcement falls from 100% on day zero to 50% by day one, 30% at week one and 20% at month one. Based on the commonly cited Ebbinghaus forgetting curve.

The pre/post pair fixes both failures. It controls for prior knowledge, it measures retention on the same schedule for everyone, and it gives the decay curve nowhere to hide. In our work with a 95-person property-management company, the maintenance team's post-only scores looked strong for two cohorts before the pre-tests revealed half the group already knew the material — the program was re-scoped, not repeated.

Key takeaway: run the pre-test even when it feels obvious — half of what a post-only score credits to training may have been there all along.

Build the Tracker: Four Columns and Three Rules

The tracker is a spreadsheet, one row per learner, and four columns do the real work: pre-score, post-score, behavior rating and metric delta. Add dates to every entry, because undated numbers cannot survive a finance review. Cohort totals roll up from learner rows; never let the cohort average replace the distribution, because averages hide the people who did not move.

Column What to enter Example entry
Pre-score (dated) Test score and confidence rating before session one 58%, 3/5 on 2026-03-02
Post-score (dated) Same items within 48 hours of the last session 84%, 4/5 on 2026-03-18
Behavior rating Manager rubric score at day 30 and day 60 2 → 4 of 5
Metric delta The business metric, cohort level, before and after FCR 61% → 74%

Three rules keep the data honest. Use identical items and scales both times, or the delta measures the instrument instead of the learner. Freeze the rubric before the first session so managers rate against the standard, not against memory. Keep the business metric pulled by whoever owns it operationally — a self-reported metric is an anecdote with a column header (the before-side design lives in our pre-training assessment guide).

Key takeaway: identical items, frozen rubrics, dated entries — three rules that keep a two-column delta from becoming an argument.

A Worked Example: One Cohort, Measured Twice

Here is an illustrative 24-person customer-service cohort measured with the tracker above. Knowledge rose from 58% to 84% on identical items, confidence from 62% to 78% on the same five-point scale, and manager-observed use of the new call flow from 30% to 65% at day 60. The business metric — first-contact resolution — moved from 61% to 74% between the baseline month and the day-60 month. Every number is illustrative, but the shape is what a real tracker produces.

Grouped bar chart of a pre and post training assessment for one illustrative 24-person customer-service cohort: knowledge test scores rise from 58% to 84%, confidence ratings from 62% to 78% and manager-observed application from 30% to 65%. Illustrative example, not survey data.

Measure Before After Change
Knowledge test (identical items) 58% 84% +26 points
Confidence rating (5-point) 62% 78% +16 points
Observed application (day 60) 30% 65% +35 points
First-contact resolution 61% 74% +13 points
Escalations per week 18 11 −7 per week (−39%)

Notice the order of usefulness to a CFO: the knowledge gain is the least interesting number, the observed application is the credibility bridge, and the metric delta is the line that gets funded. A report that leads with the test score buries the payoff. In our work with a 70-person insurance brokerage, moving the metric line to the top of the one-pager was the change that turned measurement into budget renewal.

Key takeaway: lead the report with the metric delta and use knowledge scores as supporting evidence — CFOs fund outcomes, not quiz results.

Where Measurement Stalls: The Kirkpatrick Ladder

Most programs measure the easiest level and stop. The pattern we observe across client engagements (illustrative, not survey data) looks like this: roughly 90% of programs capture reaction data, about 60% re-test learning, 30% rate behavior, and 10% connect anything to a business result. Each step up the ladder is cheaper than teams expect — the pre/post pair is the same spreadsheet with one more column.

Hbar chart of where measurement stalls on the Kirkpatrick ladder, illustrative pattern from our client observations: about 90% of programs capture level-one reaction data, 60% capture learning, 30% track behavior and 10% reach business results. Illustrative, not survey data.

The ladder stalls because levels 3 and 4 need other people: a manager willing to rate behavior and an operations owner willing to pull a metric. Book both before the program runs, in writing, and the stall disappears. For the wider dashboard these numbers feed, our L&D metrics guide lists the KPIs worth tracking beyond completion.

Key takeaway: levels 3 and 4 stall on permission, not effort — book the manager rating and the metric owner before session one.

Turn the Delta into the CFO Conversation

A measured delta converts into the same four-part proof finance expects everywhere else: baseline, change, dollars, payback. Convert the metric movement at a rate finance approves — resolved tickets into labor hours, avoided errors into rework cost, retained staff into replacement cost — and the training line starts reading like every other investment the CFO reviews. The context numbers help too: 71% of organizations offer leadership development (Training magazine, 2023), 94% of employees would stay longer where employers invest in their learning (LinkedIn Learning, 2018), and just 13% are engaged at work globally (Gallup, 2023).

Bar chart of three statistics that justify measuring training formally: 71% of organizations offer leadership development, 94% of employees would stay longer where employers invest in learning and just 13% of employees are engaged globally. Sources: Training magazine 2023; LinkedIn Learning 2018; Gallup 2023.

Then repeat it. A pre and post training assessment run once is a report; run every quarter, it becomes the instrument that decides which programs get renewed and which get cut. In our experience with US mid-market teams, the measurement habit changes program design within two cycles, because nobody wants to sit through a baseline that predicts a flat delta.

Key takeaway: convert the delta at a finance-approved rate and repeat quarterly — the second cycle is what turns measurement into leverage.

What Your Business Should Do

Days 1–30: Set the baseline.

  • Choose one upcoming program and pick four measures: knowledge test, confidence rating, behavior rubric and one business metric.
  • Write the 8–12 item pre-test and the manager rubric, then freeze both versions.
  • Record every baseline number with dates before the first session runs.

Days 31–60: Deliver and re-test.

  • Run the post-test within 48 hours of the last session, using identical items.
  • Capture the end-of-program survey before the reaction window closes.
  • Book the 30-day manager observation while enthusiasm is still high.

Days 61–90: Convert results into dollars.

  • Pull the business metric again and compute the delta for the cohort.
  • Convert the delta into dollars at a rate finance approves.
  • Send the CFO a one-page before-and-after report, metric line first.

Key takeaway: freeze the instruments before session one — a baseline you can trust is the whole game.

The Bottom Line

Measurement is cheap; guessing is not. At an average of $874 per learner (Training magazine, 2023), an unmeasured 30-person program is a $26,220 bet placed on a hunch. A pre and post training assessment converts that bet into evidence: baseline the skill, re-test after, rate behavior at 30–60 days, and pull the business metric at 90; you do not need a data team, because a spreadsheet, four columns per learner and one honest baseline will outperform any satisfaction survey. 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)

Four columns per learner that turn your next program into before-and-after evidence.

Download the PDF pack

FAQ

What is a pre and post training assessment?

It is a paired measurement design: you capture the same knowledge test, confidence rating, manager behavior rating and one business metric before the program, then capture them again after on the same schedule for everyone. The delta between the two runs is the evidence of change. It is the cheapest credible way to cover Kirkpatrick levels 2 through 4 without a research team.

How long before training should I run the pre-assessment?

One to two weeks before session one is the sweet spot: close enough that baselines do not drift, early enough that results can shape the session. Include the knowledge items you will re-run verbatim, a confidence scale and a manager baseline rating. Record dates on every entry, because undated baselines cannot survive a finance review.

What is the difference between a pre-test and a baseline metric?

A pre-test measures the person: knowledge items and confidence, scored before the program. A baseline metric measures the work: the operational number the skill should move, such as first-contact resolution or error rate, pulled for the cohort before training starts. Strong programs run both, because the pre-test explains the delta and the baseline metric prices it.

How many questions should a pre and post training assessment include?

Keep the knowledge portion to 8–12 items you can re-run in under ten minutes, plus a five-point confidence scale and a short manager rubric. Length kills completion, and incomplete pairs destroy the delta. Fifteen minutes total per learner is the ceiling we see work in practice.