What is evaluation research? (methods and examples)

Few things derail a project faster than a team relying completely on guesses and instinct. It doesn’t matter whether you’ve got a team doing market research for a new product or you’re a school or government group coming up with a new policy — if you don’t have the hard data, every decision you make could end up being more costly than you’d ever expected. 

Evaluation research is how organizations avoid making embarrassing and completely preventable mistakes. You use various methods to collect and assess information that can help you answer questions about a product, program, process, or user experience.

It sounds time-consuming, and it can be, but it can also be extremely valuable. According to Qualtrics, too many research teams are already trying to cut corners. Around 95 percent of researchers now use AI tools regularly or are experimenting with them. NORC also shared a 2026 industry presentation claiming that 40 percent of nonprobability interviews in 2025 were likely fraudulent. 

Deciding to collect feedback fast is always useful, but bad feedback just sends you in the wrong direction quicker. That’s where evaluation research comes in.

What is evaluation research?

So, what is the purpose of evaluation research?

Evaluation research methods give teams a way to collect proof about the effectiveness and value of programs, products, or policies. It helps determine if the solutions you’re introducing to the world are meeting the right goals or driving unintended consequences.

Although it sounds a lot like a scientific process mostly used by government groups or healthcare teams, it’s actually a pretty common data collection strategy for a lot of use cases. Organizations can use evaluation research for things like a product launch, a training program, a nonprofit project, a new internal process, an AI tool, or an onboarding flow. 

The work is direct by design, focusing on the questions that really matter, such as

• Did it work?
• Which group did it work for?
• What was the cost?
• What went wrong?
• Does something need to change?

Answering those questions early can save companies a lot of resources, particularly when they’re building out a product roadmap or designing a new process that impacts a large number of people. You find out quickly if whatever you’re working on is worth the time, money, and effort actually spent on it. And once you know that, you can make a more informed decision about whether you should keep going, fix the next step, change the audience, or even pull the plug.

Common real-world use cases for evaluation research

We’ve already mentioned the most obvious one: product development. No company should continue building products that no customer is going to pay for, or that never actually deliver what’s intended. However, evaluation research methods can also help with insights into

• User experience (UX) and usability: If a product team can actually see when customers struggle at checkout, or give up on an app, they can use those UX insights to fix the problem before scaling to a larger group.
• Employee training: With evaluation research, a company could compare pre- and post-training scores, review employee feedback, and make changes to training that actually improve behavior.
• Public health programs: A clinic could test whether reminder texts increase appointment attendance for the patients most likely to miss care.
• Social programs: A nonprofit could check whether a job-readiness program led to interviews, placements, or better retention, rather than stopping at attendance numbers.
• AI workflows: A support team could measure whether an AI triage tool reduced backlog without sending urgent issues to the wrong queue.

When should you carry out evaluation research?

Evaluation research is helpful for any product, program, or strategy that’s about to move from the initial idea stage towards something that people actually use. You use it through the entire project or program’s life cycle. 

You start with formative research — a lot like the kind of market research you do before you build a product or a new business. Then, midway through, you evaluate the process to see if things are being implemented properly, and address challenges you hadn’t thought about. At the end of the project, you measure the final impact and see whether you really achieved whatever you were hoping to achieve. 

Evaluation research methods are ideal when you’re about to add a new feature to the product roadmap, make a change to UX, or deliver a training program. You could also use it when you’re trying to gather proof for an investor before they give you funding. 

A lot of companies are actually starting to use evaluation research for AI strategies, too. The UK Government even says that AI interventions should be measured against the status quo and judged through process, impact, and value-for-money questions. 

What are the types of evaluation research?

We touched on these very briefly, but the four main types of evaluation research are formative, summative, process, and outcome evaluation. 

Formative evaluation research is the early-stage investigation, and it’s the one that most companies are tempted to skip. This is done during the development of a program or product. The whole point is to identify areas for improvement and hopefully make positive changes to the program before it launches. 

Summative evaluation research takes place when the project ends, and it looks at the main questions, such as whether the initiative achieved its goal. For a company testing a product, for instance, that might mean seeing whether people are actually using the product, and whether it delivers the benefits the company promised. 

Process evaluation looks at the delivery of the project. For instance, if a clinic decided to automate reminder texts for patients to reduce no-shows, they’d need to see whether the texts were sent at the right time, whether they hit the inbox, whether they were compliant with healthcare standards, and whether the phone numbers were correct.

Finally, outcome evaluation looks at what happened as a result of the project or process working. For an HR team, that could mean retention improves. For a clinic, fewer people miss appointments. For a training team, day-to-day performance gets better. Whatever the goal, outcome evaluation tells you if the work was worth doing.

The main methods of evaluation research

The two main types of evaluation research are the same types you’ll see with virtually any form of research: quantitative and qualitative. The first looks at numbers, and the second looks at the meaning and thoughts behind those numbers.

Quantitative evaluation research

Quantitative evaluation research is the number-focused option. You ask questions designed to get actual figures back from respondents. Depending on what you’re analyzing, you may be looking for costs, completion rates, training scores, satisfaction ratings, time saved in hours, adoption rates, or error counts. What you get is something very easy to display in a dashboard.

At a glance, quantitative research can tell you very quickly whether the new onboarding flow actually improved activation, whether the support AI reduced ticket backlog, or whether a training program improved test scores. Still, it’s only reliable when the data is clean. If the sample is bad or the questions are leading, you end up with the wrong figures.

Qualitative evaluation research

Qualitative evaluation research looks deeper, beyond the numbers. It takes more work to gather, although there are plenty of survey tools for research that can help.

For qualitative insights, you’re running interviews, sending out forms with open-ended survey questions, managing focus groups, and gathering observations or user comments. Instead of just asking “how many people used this,” you’re asking questions like, “what did they feel when they were using it?” or “where did they get stuck, and why?”

Focused product survey questions can tell companies a lot more about what their customers really need than a standard satisfaction score. Again, though, you need to be careful that you’re not accidentally leading respondents towards a specific answer.

What are the steps of evaluation research?

With evaluation research, the best path is usually to start with the decision you need to make and work backward. 

1. Define the objective: Decide exactly what you’re judging, based on a decision you need to make. For instance, you might be figuring out whether onboarding for a new tool has to improve before it goes public.
2. Decide who needs the answer: Work out who needs to use the findings. It might be a founder, a product lead, an entire board, a department head, or a funding partner. Different people will read the information with a specific focus in mind.
3. Choose the evaluation type: Pick formative, summative, process, or outcome evaluation research based on the timing. Early draft? Formative. Finished pilot? Summative. Messy rollout? Process. Real-world change? Outcome. 
4. Pick the method: Use quantitative questions when you need counts, rates, scores, costs, or percentages. Use qualitative questions when you need the story behind the number. Most research studies end up combining both.
5. Build the research tools: Collect survey question examples and design your surveys, interview guide, assessment, or checklist, too. Make sure every question will actually help you get closer to making a decision.
6. Collect the data: Start gathering the information. If you’re collecting data with surveys, make sure you send them to the right audience, at an appropriate time, and give enough time to actually respond. 
7. Read the findings: Analyzing the survey data can sometimes be the most difficult part. What you’re looking for are problematic patterns. Did the tool help new users but annoy power users? Did training improve scores without changing behavior? Did the program reach the easiest audience while missing the intended one? 
8. Make the call: Finally, send the information to the right people, in the right format (slideshow, PowerPoint, visual graph). Then use that information to make your decision.

How can you collect data for evaluation research?

There are plenty of ways to collect data for evaluation research; some people are even using AI to help them speed up the process. The evaluation research methods you use really depend on the data you need to end up with. 

Common options include

• Surveys and questionnaires: Creating a survey is one of the easiest ways to collect both quantitative and qualitative data fast. You can send surveys out to people online, and mix up questions with rating scales, multiple-choice options, and a few open-ended prompts.
• Interviews: Interviews take more time, but if you can get someone valuable to agree to one, they can give you a lot more qualitative depth that you can use to make intelligent decisions. Whether you’re using a structured or semi-structured interview, decide the critical questions in advance — and stick to them.
• Document and records review: Existing records might be hiding more information than you know. Support tickets, refund notes, CRM fields, attendance logs, training scores, product analytics, complaint emails, call transcripts, and appointment data can all reveal patterns that would otherwise have gone unnoticed.
• Focus groups: Focus groups are the more social version of the interview. They’re helpful for community programs, product concepts, public health messaging, training materials, and service changes where language carries a lot of weight.
• Tests and assessments: Tests are right for skills, knowledge, or task performance. Use them for training programs, onboarding flows, prototype tasks, certification prep, or process changes where people need to prove they can do the thing.

Remember, tools are often extremely helpful here. Jotform is one of the best survey tools teams can use to collect data for evaluation research. With the no-code survey builder, anyone can create custom forms that combine all kinds of questions, giving you incredibly insightful feedback. 

Examples of effective evaluation research questions

Coming up with the right questions to ask is usually one of the tougher parts of conducting research. Before you start writing a research question, the most valuable thing to do is think carefully about the decision you’re trying to make and what you need to assess. 

Here’s what that could look like across a few common areas.

Product and user experience

Product teams need questions that expose friction. Don’t just ask if customers liked something — people like plenty of things they never use, never pay for, and never recommend. Instead, use questions like

• What percentage of first-time users completed setup without contacting support? 
• Which step caused the most drop-off during onboarding? 
• How long did it take users to complete the main task? 
• What part of the experience felt unclear or irritating? 
• Which feature felt useful enough to come back for? 
• What would make you stop using this product after the first week? 

Training and internal programs

For training, go beyond relying on certifications and completion rates. Ask questions like 

• What percentage of participants improved their assessment score after training? 
• How many participants used the new process within 30 days? 
• Which part of the training helped you most during real work? 
• What still feels hard to apply without help? 
• Did managers give you enough time or support to use what you learned? 
• Was the performance change worth the cost of the program? 

Public health and social programs

A public health campaign, community program, or nonprofit service can look successful while missing the people it was supposed to reach. Ask

• How many people from the target group participated? 
• What percentage completed the full program? 
• Which participant groups saw the greatest improvement? 
• What changed in behavior, knowledge, access, or confidence? 
• What made the program harder to use than expected? 
• Were there any negative side effects nobody planned for? 

Formative evaluation during development

Formative questions are the ones teams should ask before the launch party, the paid campaign, or the full rollout. Use questions like

• What percentage of testers understood the main value proposition? 
• How many users completed the prototype task without extra instructions? 
• Which version of the concept performed better in early testing? 
• What did users misunderstand on first contact? 
• What concern would stop you from trying this? 
• What needs to change before this is ready? 

Keep in mind that you don’t need to start with a blank form. Jotform’s ready-to-use evaluation form templates provide a useful base for your team. You can start with templates for customer satisfaction, product feedback, training evaluation, course evaluation, event feedback, implementation review, employee performance, and service quality. Then edit the questions until they match the decision you’re actually trying to make.

Conduct your evaluation research with Jotform

If you’re starting to think that an evaluation research session would be helpful for your company, getting the right technology in place will save you a lot of time and effort.

Jotform helps with the complicated parts of collecting information. Say a product team is testing a new onboarding flow, while an HR team is measuring training results, and a healthcare organization is collecting post-visit feedback. All three need different forms, but none of them should have to wait on a developer just to ask better questions. With Jotform’s drag-and-drop form builder, you can build surveys, polls, feedback forms, assessments, intake forms, and program evaluations without code.

The conditional logic is helpful, too. It means that if someone says they abandoned checkout, they can see a follow-up about what stopped them. If a training participant says they didn’t use the new process, they can get a question about the barrier.

You can also route responses to Jotform Tables, where your team can sort submissions, review patterns, and generate visual summary reports instead of piecing everything together manually. 

For sensitive evaluation research, Jotform also offers HIPAA-enabled forms and GDPR compliance support.

Try Jotform for free to build evaluation surveys, collect cleaner data, and turn responses into decisions people can actually defend.