What Is the Phillips ROI Methodology?
Sooner or later, someone with a budget asks the question every training team dreads: "What are we getting for this money?" Completion rates and smile sheets won't satisfy them. They want to know whether the program paid for itself. You might have heard that the Phillips ROI Methodology can answer that question, but it can sound intimidating, full of formulas and finance terms. It's more approachable than it looks. What is the Phillips ROI Methodology, and how does it actually work?
The Phillips ROI Methodology is a training evaluation framework that measures results across five levels, from participant reaction up to return on investment, and calculates ROI by comparing a program's monetary benefits with its fully loaded costs. Developed by Jack Phillips, it builds on the Kirkpatrick model with a fifth level for ROI and a required step to isolate training's effects from other factors. The formula is simple: net program benefits divided by program costs, times 100. Its conservative rules, like counting only benefits you can attribute to the program and including every cost, are what make it credible. It's best saved for a few expensive or strategic programs.
What Are the Five Levels of the Phillips ROI Methodology?
The five levels form a chain. Each level depends on the one below it, so a program that people disliked and didn't learn from is unlikely to produce business results.
| Level | What it measures | Example question |
|---|---|---|
| 1. Reaction and planned action | How participants felt and what they intend to do | Was it relevant? What will you apply? |
| 2. Learning | Knowledge and skills gained | Can they do it in a test or simulation? |
| 3. Application and implementation | Use on the job | Are they doing it at work? |
| 4. Impact | Business measures that changed | Did errors, sales, time or safety change? |
| 5. ROI | Monetary benefits compared with costs | Did the benefits exceed the costs, and by how much? |
If you know the Kirkpatrick model, the first four levels will look familiar. The Phillips approach adds two important pieces: isolating the effects of training at Level 4, and converting impact into money so you can calculate Level 5. The ROI Institute, which Jack and Patti Phillips lead, maintains detailed guidance on the full process.
How Is ROI Calculated?
ROI is calculated with two simple formulas once you have your benefits and costs:
- ROI (%) = (Program benefits − Program costs) ÷ Program costs × 100
- Benefit-cost ratio (BCR) = Program benefits ÷ Program costs
An ROI of 0% means the program broke even. An ROI of 100% means it returned the cost plus that much again. A BCR of 2:1 means two dollars of benefit for every dollar spent. Try your own numbers in the calculator below.
The math is the easy part. The hard, valuable work is getting honest numbers to put into it, which is where the methodology's steps come in.
The Key Steps: Isolate, Convert and Count Every Cost
Isolate the effects of the program
Business results change for many reasons: new managers, a new product, a seasonal swing. Isolation separates the part of the improvement caused by training from everything else. Common approaches include comparing a trained group with a similar untrained group, analyzing trends from before the program, and asking participants or managers to estimate how much of the change came from the training, then adjusting that estimate by their confidence.
Convert impact to money
Next, each impact measure gets a monetary value. Some are easy, like the cost of an error your finance team already tracks or the profit on an extra sale. Others need expert input or historical data. Measures you can't credibly convert, like teamwork or job satisfaction, are reported as intangible benefits, which still count in the story even though they stay out of the formula.
Fully load the costs
Costs include everything: needs analysis, design and development, delivery, materials, facilities, travel, the evaluation itself and, often the biggest item, participants' time away from work. Leaving costs out makes ROI look better and makes the whole study less believable.
A Worked Phillips ROI Example
Here's a simple example with assumed figures. Say a customer service team takes a program on resolving complaints on the first call. After six months, the reported annual value of fewer repeat calls is $150,000. Here's how the conservative rules shrink that number before it reaches the formula:
Worked example with assumed figures
Model: worked example with assumed inputs (50% attributed to training, 80% average estimate confidence, $40,000 fully loaded cost). Not research data.
View the data
| Dollars | |
|---|---|
| Reported improvement | $150,000 |
| After isolation (50%) | $75,000 |
| After confidence (80%) | $60,000 |
| Program cost | $40,000 |
The chart above shows each step. Managers estimate that half of the improvement came from the training, so $75,000 remains. Their average confidence in that estimate is 80%, so the counted benefit is $60,000. With fully loaded costs of $40,000, the ROI is ($60,000 − $40,000) ÷ $40,000 × 100 = 50%, and the BCR is 1.5:1. That's a smaller number than the first figure suggested, and that's exactly why a finance leader can trust it.
What Rules Keep Phillips ROI Results Credible?
The methodology includes a set of guiding principles, and several of them protect you from overclaiming. Among the most useful:
- Use the most conservative option when you have a choice.
- Use at least one method to isolate the effects of the program.
- If you have no improvement data for a group, assume little or no improvement.
- Adjust estimates for potential error, as in the confidence step above.
- For short programs, count only the first year of benefits.
- Include all costs, fully loaded.
These rules can feel strict, but they make your results defensible. A modest ROI that holds up under questions is worth far more than a dramatic one that falls apart in the first meeting.
Common Mistakes in Training ROI Studies
Most ROI studies that lose credibility make one of a few avoidable mistakes:
- Skipping isolation and claiming the whole improvement for training.
- Leaving out participant time, which is often the largest cost of all.
- Counting several years of benefits for a short program.
- Choosing business measures after the program, once you can see which ones improved.
- Hiding the intangible benefits, which leaves out part of the real story.
Each of these is easy to fix when you plan the study before launch. They're nearly impossible to fix once the results meeting is on the calendar.
When Should You Use the Phillips ROI Methodology?
Use it selectively. A full ROI study takes real effort, so save it for programs that are expensive, strategic, aimed at a clear business problem or under close scrutiny from leaders. For most courses, measuring reaction, learning and on-the-job application is enough.
The biggest secret is to plan for ROI before the program starts. Agree with stakeholders on the business measures, the baseline and the isolation method up front. That design-first mindset is the heart of instructional engineering, which treats training as a system built to produce measurable results rather than content built and hoped for the best.
If you only remember one thing, remember this: decide how you'll measure impact before you build the course. It's very hard to prove ROI after the fact.
The Phillips ROI Methodology Turns "Was It Worth It?" into a Real Answer
The Phillips ROI Methodology measures training across five levels and ends with a clear financial answer: what the program returned compared with what it cost. Its real strength is discipline. Isolating the effects of training, converting impact to money with care and counting every cost make the result something leaders can believe.
Pick one important program this quarter and plan its evaluation now. Define the business measure, collect a baseline and choose your isolation method. When the budget question comes, you'll have an answer ready.