{"id":8695,"date":"2026-07-28T06:53:56","date_gmt":"2026-07-28T06:53:56","guid":{"rendered":"https:\/\/nextagile.ai\/blogs\/?p=8695"},"modified":"2026-07-28T06:56:29","modified_gmt":"2026-07-28T06:56:29","slug":"kirkpatrick-model-training-evaluation","status":"publish","type":"post","link":"https:\/\/nextagile.ai\/blogs\/agile\/kirkpatrick-model-training-evaluation\/","title":{"rendered":"Kirkpatrick Model: Complete Guide to the 4 Levels of Training Evaluation (2026)"},"content":{"rendered":"<p><b>Quick Answer: <\/b><span style=\"font-weight: 400;\">The Kirkpatrick Model is a four-level framework for evaluating training: Level 1 Reaction measures whether participants found the training relevant and engaging, Level 2 Learning measures whether knowledge or skill actually increased, Level 3 Behavior measures whether people apply that skill on the job, and Level 4 Results measures whether an organizational outcome moved. Donald Kirkpatrick introduced the model in 1959, and it remains the most widely used training evaluation vocabulary in learning and development. Most organizations only measure Level 1, which is why leadership can quote satisfaction scores for years and still not know whether anyone works differently. The model is best planned backward, starting from the Level 4 business outcome you want and designing the training and Level 1 to 3 measures to support it, and it works best for major or high-stakes training investments rather than every short session.<\/span><\/p>\n<h2><b>Key Highlights of Kirkpatrick Model<\/b><\/h2>\n<ul>\n<li><span style=\"font-weight: 400;\">The Kirkpatrick Model has 4 levels: Reaction, Learning, Behavior, and Results, introduced by Donald Kirkpatrick in 1959 based on his 1954 University of Wisconsin dissertation.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Most organizations measure only Level 1, according to multiple 2026 industry sources, missing the behavior and results data that actually proves training ROI.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">In 2026, Vanessa Milara Alzate expanded the New World Kirkpatrick Model to explicitly include the performance environment, the systems and culture that determine whether training transfers to real behavior.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Kirkpatrick Partners recommends planning evaluation backward, from Level 4 Results to Level 1 Reaction, rather than the traditional sequential Level 1-to-4 order.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Common complementary frameworks include the Phillips ROI Methodology (adds a fifth ROI level), CIRO (adds front-end context), and Brinkerhoff&#8217;s Success Case Method.<\/span><\/li>\n<\/ul>\n<p><b>Introduction<\/b><\/p>\n<p><b>The Kirkpatrick Model<\/b><span style=\"font-weight: 400;\"> is a four-level framework for evaluating whether training programs actually work, built around Reaction, Learning, Behavior, and Results. Donald Kirkpatrick introduced it in 1959, and more than 65 years later it remains the dominant vocabulary in corporate learning and development. The model is simple to recite in a meeting and genuinely difficult to run in practice, because the record-keeping underneath each level is where most L&amp;D teams stall.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">This guide walks through all 4 levels with real sample questions, shows how to plan an evaluation backward from business outcomes, covers what changed in the 2026 update to the New World Kirkpatrick Model, and compares it honestly against alternatives like the Phillips ROI Methodology. It is built for L&amp;D professionals, HR leaders, and students studying instructional design who need more than a definition.<\/span><\/p>\n<h2><b>What Is the Kirkpatrick Model and Where Did It Come From<\/b><\/h2>\n<p><span style=\"font-weight: 400;\">Snippet answer: The Kirkpatrick Model is a training evaluation framework with 4 levels, Reaction, Learning, Behavior, and Results, created by Donald Kirkpatrick in 1959 and still the most widely used evaluation vocabulary in learning and development today.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Donald Kirkpatrick developed the framework as part of his 1954 PhD dissertation at the University of Wisconsin and published the four levels through a series of articles in 1959. In the 2010s, his son Jim Kirkpatrick and daughter-in-law Wendy Kirkpatrick launched the New World Kirkpatrick Model through Kirkpatrick Partners, keeping the original 4 levels but adding required drivers for on-the-job behavior, confidence and commitment as Level 2 indicators, and a strong emphasis on planning evaluation backward from Level 4.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">In 2026, Kirkpatrick Partners credits Vanessa Milara Alzate with expanding the model further to name the <\/span><b>performance environment<\/b><span style=\"font-weight: 400;\"> explicitly, the systems, culture, and reinforcement structures that determine whether training transfers into real behavior change, extending the framework&#8217;s use beyond L&amp;D into enterprise performance intelligence more broadly, according to<\/span><a href=\"https:\/\/www.kirkpatrickpartners.com\/the-kirkpatrick-model\/\" rel=\"nofollow noopener\" target=\"_blank\"> <span style=\"font-weight: 400;\">Kirkpatrick Partners<\/span><\/a><span style=\"font-weight: 400;\">.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Teams building a training plan around this framework can pair it with NextAgile&#8217;s guide on<\/span><a href=\"https:\/\/nextagile.ai\/blogs\/agile-transformation\/how-to-develop-agile-training-plan\/\"> <span style=\"font-weight: 400;\">how to develop an agile training plan<\/span><\/a><span style=\"font-weight: 400;\"> to structure the design phase before evaluation even begins.<\/span><\/p>\n<h2><b>The 4 Levels of the Kirkpatrick Model, Explained<\/b><\/h2>\n<p><span style=\"font-weight: 400;\">Snippet answer: The 4 levels of the Kirkpatrick Model are Level 1 Reaction (did learners find it engaging and relevant), Level 2 Learning (did knowledge or skill increase), Level 3 Behavior (do people apply it on the job), and Level 4 Results (did a business outcome move).<\/span><\/p>\n<h3><b>Level 1: Reaction<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Level 1 measures whether participants found the training engaging, relevant, and worth their time. It is most commonly captured through a post-training survey, often called a smile sheet, delivered immediately after the session ends. The focus should stay on the learner&#8217;s takeaway, not the trainer&#8217;s performance, since a well-liked facilitator does not guarantee any actual skill transfer.<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Sample question: How relevant was this training to your day-to-day role?<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Sample question: How confident do you feel applying what you learned within the next 2 weeks?<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">Level 1 is the cheapest and fastest level to measure, which is exactly why most organizations stop here. Research cited across multiple 2026 L&amp;D sources notes that learner reaction is a poor standalone indicator of whether real learning or behavior change occurred, so treat Level 1 as an early warning signal, not a success metric.<\/span><\/p>\n<h3><b>Level 2: Learning<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Level 2 measures whether knowledge, skill, or confidence actually increased, typically through pre- and post-training assessments, skills demonstrations, or certification exams. The New World update added confidence and commitment as explicit Level 2 indicators, since a learner can pass a knowledge test and still lack the confidence to apply it on the job.<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Sample question: Pre\/post assessment score comparison on the core skill taught<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Sample question: On a scale of 1 to 5, how confident are you in applying this skill without support?<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">Teams running <a href=\"https:\/\/nextagile.ai\/blogs\/agile\/kanban-vs-scrum\/\">Scrum<\/a> or SAFe certification programs can benchmark Level 2 assessment design against NextAgile&#8217;s<\/span><a href=\"https:\/\/nextagile.ai\/blogs\/career\/corporate-training-programs\/\"> <span style=\"font-weight: 400;\">corporate training programs<\/span><\/a><span style=\"font-weight: 400;\"> guide, which covers assessment structure for enterprise learning tracks.<\/span><\/p>\n<h3><b>Level 3: Behavior<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Level 3 measures whether people actually apply the training on the job, typically measured 30 to 90 days after the session through manager observation, 360-degree feedback, or on-the-job performance data. This is the level most organizations skip, because it requires tracking someone&#8217;s behavior weeks after the training ended rather than a single survey at the exit door.<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Sample question: In the last month, how often did you apply [specific skill] in a real work situation?<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Sample question: What barriers, if any, prevented you from applying this skill?<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">The New World Kirkpatrick Model calls the systems that support this transfer required drivers, meaning reinforcement, manager coaching, and accountability structures put in place after training ends. Training with no reinforcement plan reliably fails at Level 3 regardless of how good the original session was.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">This is the exact gap NextAgile&#8217;s<\/span><a href=\"https:\/\/nextagile.ai\/agile-coaching-and-training\/\"> <span style=\"font-weight: 400;\">Agile Coaching and Training<\/span><\/a><span style=\"font-weight: 400;\"> model is built to close, since ongoing coaching after a workshop is what typically converts Level 2 learning into Level 3 behavior change.<\/span><\/p>\n<h3><b>Level 4: Results<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Level 4 measures whether training moved an actual organizational outcome, such as revenue, defect rate, retention, or delivery velocity. It is the most expensive and time-consuming level to measure, since isolating training&#8217;s contribution from every other variable affecting a business metric is genuinely hard, but it is the level that actually proves ROI to a board or budget owner.<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Sample metric: change in sprint velocity or delivery throughput after an agile training rollout<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Sample metric: change in employee retention or internal promotion rate following a leadership program<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">Organizations connecting Level 4 to team performance often track it alongside NextAgile&#8217;s<\/span><a href=\"https:\/\/nextagile.ai\/blogs\/okr\/agile-metrics-and-kpis\/\"> <span style=\"font-weight: 400;\">agile metrics and KPIs<\/span><\/a><span style=\"font-weight: 400;\"> and<\/span><\/p>\n<p><a href=\"https:\/\/nextagile.ai\/blogs\/agile-transformation\/agile-transformation-metrics\/\"><span style=\"font-weight: 400;\">agile transformation metrics<\/span><\/a><span style=\"font-weight: 400;\">, which give Level 4 a concrete, pre-existing measurement baseline instead of inventing new metrics from scratch.<\/span><\/p>\n<h2><b>Kirkpatrick Model Levels at a Glance<\/b><\/h2>\n<table>\n<tbody>\n<tr>\n<td><b>Level<\/b><\/td>\n<td><b>Question It Answers<\/b><\/td>\n<td><b>How It&#8217;s Measured<\/b><\/td>\n<td><b>Typical Timing<\/b><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">1. Reaction<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Was the training engaging and relevant?<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Post-session survey (smile sheet)<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Immediately after training<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">2. Learning<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Did knowledge or skill increase?<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Pre\/post assessment, skills demo<\/span><\/td>\n<td><span style=\"font-weight: 400;\">End of training<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">3. Behavior<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Is it applied on the job?<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Manager observation, 360 feedback<\/span><\/td>\n<td><span style=\"font-weight: 400;\">30 to 90 days after<\/span><\/td>\n<\/tr>\n<tr>\n<td><span style=\"font-weight: 400;\">4. Results<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Did a business outcome move?<\/span><\/td>\n<td><span style=\"font-weight: 400;\">Business KPI tracking (revenue, velocity, retention)<\/span><\/td>\n<td><span style=\"font-weight: 400;\">3 to 12 months after<\/span><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n<h2><b>How to Implement the Kirkpatrick Model, Step by Step<\/b><\/h2>\n<p><span style=\"font-weight: 400;\">Snippet answer: Implement the Kirkpatrick Model by planning backward from Level 4, starting with the business outcome you want, then defining the Level 3 behaviors, Level 2 skills, and Level 1 experience needed to produce it, rather than measuring the levels sequentially after training ends.<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Step 1: Name the Level 4 business outcome first, such as a target reduction in defect rate or improvement in delivery velocity.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Step 2: Define the Level 3 on-the-job behaviors that would realistically drive that outcome, and identify who observes them.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Step 3: Design the Level 2 skills and knowledge checks that prove someone can perform those behaviors.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Step 4: Build the Level 1 survey last, focused on relevance and confidence rather than trainer likability.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Step 5: Put required drivers in place before training ends, meaning manager reinforcement, coaching check-ins, and accountability structures for Level 3.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Step 6: Set a review cadence at 30, 60, and 90 days post-training to capture Level 3 and Level 4 data rather than treating evaluation as a one-time event.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">Teams building this backward-design process for leadership programs specifically can reference NextAgile&#8217;s<\/span><a href=\"https:\/\/nextagile.ai\/blogs\/leadership\/leadership-coaching-program\/\"> <span style=\"font-weight: 400;\">Leadership Coaching Program<\/span><\/a><span style=\"font-weight: 400;\"> and<\/span><\/p>\n<p><a href=\"https:\/\/nextagile.ai\/corporate-leadership-training\/\"><span style=\"font-weight: 400;\">Corporate Leadership Training<\/span><\/a><span style=\"font-weight: 400;\">, both structured around measurable behavior change rather than a one-day session.<\/span><\/p>\n<h2><b>Limitations, Alternatives, and 2026 Updates<\/b><\/h2>\n<p><span style=\"font-weight: 400;\">Snippet answer: The Kirkpatrick Model&#8217;s main limitation is that Level 4 Results is hard to isolate from other variables affecting a business outcome, which is why organizations often pair it with the Phillips ROI Methodology, CIRO, or Brinkerhoff&#8217;s Success Case Method rather than replacing it outright.<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Phillips ROI Methodology: adds a fifth level calculating a monetary ROI ratio, useful when a board specifically needs a dollar figure.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">CIRO Model: adds Context and Input evaluation before training begins, useful for diagnosing whether a program was designed correctly in the first place.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Brinkerhoff&#8217;s Success Case Method: focuses on case-based insight from the highest and lowest performers post-training rather than aggregate survey data.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">Most L&amp;D teams in 2026 use Kirkpatrick as the spine of their evaluation program and borrow from these alternatives only when the situation calls for it, rather than picking one framework exclusively. The 2026 performance-environment update also reflects a broader shift toward treating evaluation as a continuous loop tied to one participant record, rather than a single end-of-program report.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">Organizations mapping training ROI to broader performance systems can connect Level 4 results directly to<\/span><a href=\"https:\/\/nextagile.ai\/okr-programs\/\"> <span style=\"font-weight: 400;\">OKR Programs<\/span><\/a><span style=\"font-weight: 400;\">, since a shared metrics language between training evaluation and organizational goal-setting makes Level 4 easier to defend to leadership.<\/span><\/p>\n<h2><b>Common Mistakes Organizations Make With the Kirkpatrick Model<\/b><\/h2>\n<p><span style=\"font-weight: 400;\">The most expensive mistake is designing the training first and bolting on evaluation afterward. When evaluation is an afterthought, Level 4 outcomes get invented after the fact to justify a program that was never built to move them, which is backward from how Kirkpatrick Partners recommends using the model.<\/span><\/p>\n<ul>\n<li><span style=\"font-weight: 400;\">Mistake 1: Measuring only Level 1 and reporting satisfaction scores as if they prove training effectiveness.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Mistake 2: Skipping required drivers, so no manager reinforcement or coaching exists to support Level 3 behavior change.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Mistake 3: Designing evaluation after the training is already built, instead of planning backward from Level 4.<\/span><\/li>\n<li><span style=\"font-weight: 400;\">Mistake 4: Applying full 4-level evaluation to every minor training session, which wastes budget better spent on high-stakes programs.<\/span><\/li>\n<\/ul>\n<p><span style=\"font-weight: 400;\">A short, low-stakes session genuinely only needs Levels 1 and 2. Save the full 4-level evaluation, including required drivers and a 90-day Level 3 check, for major or expensive training investments where the business actually needs proof of ROI.<\/span><\/p>\n<h2><b>Conclusion<\/b><\/h2>\n<p><span style=\"font-weight: 400;\">The Kirkpatrick Model still answers the question every L&amp;D leader eventually gets asked: did this training actually work? The 4 levels are easy to recite and genuinely hard to run, and the gap between organizations that stop at Level 1 and those that track all 4 levels is the gap between guessing and proving ROI. Plan backward from the Level 4 outcome you actually need, put required drivers in place before training ends, and treat Level 1 as an early signal rather than a finish line.<\/span><\/p>\n<p><span style=\"font-weight: 400;\">If your organization is measuring Level 1 satisfaction scores but still cannot tell leadership whether behavior actually changed,<\/span> <span style=\"font-weight: 400;\">NextAgile&#8217;s <\/span><a href=\"https:\/\/nextagile.ai\/nextagile-learning-programs\/\"><span style=\"font-weight: 400;\">corporate training and coaching programs<\/span><\/a><span style=\"font-weight: 400;\"> are built around Level 3 and Level 4 evidence, not just a post-session survey.<\/span><\/p>\n<h2><b>Frequently Asked Questions<\/b><\/h2>\n<h3><b>What are the 4 levels of the Kirkpatrick Model?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">The 4 levels are Reaction (did learners find it engaging), Learning (did knowledge or skill increase), Behavior (is it applied on the job), and Results (did a business outcome improve).<\/span><\/p>\n<h3><b>Who created the Kirkpatrick Model and when?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Donald Kirkpatrick created the model as part of his 1954 PhD dissertation at the University of Wisconsin and published the 4 levels through a series of articles in 1959.<\/span><\/p>\n<h3><b>Is Level 1 or Level 4 more important to measure?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Level 4 is more important for proving business ROI, but Level 1 is still useful as an early warning signal. The mistake most organizations make is stopping at Level 1 and never reaching Level 4.<\/span><\/p>\n<h3><b>How long after training should Level 3 behavior be measured?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Most L&amp;D practitioners measure Level 3 behavior 30 to 90 days after training ends, since on-the-job application takes time to show up consistently and needs manager observation rather than a single self-report survey.<\/span><\/p>\n<h3><b>What is the New World Kirkpatrick Model?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">The New World Kirkpatrick Model, developed by Jim and Wendy Kirkpatrick, keeps the original 4 levels but adds required drivers for on-the-job behavior, confidence and commitment as Level 2 indicators, and backward planning from Level 4.<\/span><\/p>\n<h3><b>How does the Kirkpatrick Model compare to the Phillips ROI Methodology?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">The Phillips ROI Methodology builds on Kirkpatrick&#8217;s 4 levels and adds a fifth level that calculates a specific monetary ROI ratio, useful when leadership needs a dollar-figure business case rather than the Kirkpatrick levels alone.<\/span><\/p>\n<h3><b>Can the Kirkpatrick Model be used for agile or leadership training specifically?<\/b><\/h3>\n<p><span style=\"font-weight: 400;\">Yes. Organizations commonly map Level 3 behaviors to agile ceremonies or leadership habits, and Level 4 results to metrics like delivery velocity, retention, or team maturity scores already tracked by the business.<br \/>\n<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Quick Answer: The Kirkpatrick Model is a four-level framework for evaluating training: Level 1 Reaction measures whether participants found the training relevant and engaging, Level 2 Learning measures whether knowledge or skill actually increased, Level 3 Behavior measures whether people apply that skill on the job, and Level 4 Results measures whether an organizational outcome&#8230;<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"content-type":"","footnotes":""},"categories":[2],"tags":[],"class_list":["post-8695","post","type-post","status-publish","format-standard","hentry","category-agile"],"_links":{"self":[{"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/posts\/8695","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/comments?post=8695"}],"version-history":[{"count":1,"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/posts\/8695\/revisions"}],"predecessor-version":[{"id":8696,"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/posts\/8695\/revisions\/8696"}],"wp:attachment":[{"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/media?parent=8695"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/categories?post=8695"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/nextagile.ai\/blogs\/wp-json\/wp\/v2\/tags?post=8695"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}