Product testing
Learn how your product performs in real use
Put your product in the hands of the people you make it for. Learn what works, what needs fixing and what makes them prefer it, before launch or after a change.
Blind · random order · no branding
Recipe 728 · 2 of 2
How did it taste?
Saltiness
Overall liking
Would you buy it?
Probably wouldWhy test it in use
A product can pass on paper and fail in use
A product can look right in the brief and still fall short at home: hard to open, the wrong taste, a feature nobody finds. Finding that after launch means returns, poor reviews and costly fixes. Product testing combines what users tell you with what you can measure. It shows the gaps in how the product works, how easy it is to use and how it is positioned, while there is still time to act.
What can go on the bench
Blind, side by side
Two recipes, tasted blind
- 01The product goes in their hands
- 02They use it, then answer
- 03Just about right builds up
- 04A short list of what to fix
Step 1 / 4
The product goes in their hands
Testers use the real thing. Here: two recipes, no branding, three-digit codes, a random order.
Study set-up
Product testing
- ProductThe current marinara and a new one with less salt
- WhoRegular buyers of the sauce
- DesignBoth recipes, unbranded, tasted in a random order
- QuestionsLiking, saltiness on a just-about-right scale, would buy, preference
- ThenWhat worked and what did not, in their own words or on video
Set up with our team, or yourself in CatalyseAI
Step 2 / 4
They use it, then answer
Overall liking, single features, just about right, would buy. Then what worked and what did not, in their own words.
Recipe 2 of 2
How was the saltiness of recipe 728?
Saltiness
Herb flavor
Thickness
"Could not tell them apart, to be honest."
"The second one tasted flat next to the first."
"Rich tomato. I would buy either."
On-camera answer
AI follow-up: What would you change about it?
"It is fine, but flat. I would put more basil in it."
- Live person
- Light
- Length
- Relevance 0.86
- Attention 0.91
Transcribed, checked and searchable. Up to three follow-ups per answer.
Step 3 / 4
Just about right builds up
Too little, just right or too much, for each recipe and each feature. Every result can be split by how often people buy.
Live tally, as answers arrive
Saltiness
Herb flavor
Thickness
Step 4 / 4
A short list of what to fix
The answers turn into priorities: what to keep, what to change and what to test again.
Recipe against recipe
What we measure (it depends on the product)
Each row is one measure, named the way the report names it.
Base: Regular buyers of the sauce; everyone tasted both recipes, unbranded and in a random order.
Top-two-box shares. Just about right shows the whole scale: too little, just right, too much.
The read-out
Example: a marinara sauce with less salt
A US pasta-sauce brand cuts the salt in its marinara and wants to know if regular buyers will notice. Buyers taste the current and the new recipe without branding, in a random order, and rate each one for liking, saltiness and whether they would buy it. Most like the two equally, but buyers who use the sauce several times a week find the new one flat. The brand keeps the new recipe, adds more herbs, and tests again before the switch.
Less salt (728)
Keep, add herbs
Most people like the two equally: liking 70 against 72.
Heavy users find it flat, so test the herb version before the switch.
Current recipe (314)
Reference
Today's recipe sets the bar every change is read against.
Stays on shelf until the new recipe passes.
Ask the studyAnswers from your own data
Do heavy users notice the salt cut?
Yes. People who use the sauce several times a week rate the new recipe lower on liking, 59 against 74, and more of them call it flat. Light users rate the two the same.
It checks its numbers against your data and draws the chart.
The process
How we run a product test
From the right testers to a short list of fixes.
- 01
Match the test to the stage
Early ideas are checked with a structured survey. A prototype or a finished product is tested in use. We choose the format with you, based on the product and the decision you need to make.
- 02
Find the right users
A short screener finds people who fit your target, such as current buyers of the category or users of a rival brand.
- 03
Let them use it, then ask
Testers use the product, then rate it on what matters for it: overall liking, single features, whether it met their expectations and whether they would buy it. They also say in their own words what worked and what did not.
- 04
Watch where it matters
Usability tests show how people actually handle the product, and tests in a central location let us see reactions as they happen. Focus groups and in-depth interviews add the context that numbers alone cannot give.
- 05
Turn the answers into priorities
Feature ranking shows what users value most, and driver analysis shows which parts of the experience move overall satisfaction the most. The result is a short list of what to fix or build first.
Outputs
What you gain
Output 01
Clear priorities
Which fixes and features matter most to users, so development time goes where it counts.
Output 02
Feedback from real use
Answers from people who used the product, not from people imagining it.
Output 03
How users compare and choose
How they weigh one feature against another, and against what they use now.
Output 04
Less launch risk
Problems found before launch cost less to fix than problems found in reviews.
Output 05
Their own words, on video if you like
Open answers grouped into themes, and short video answers with transcripts, so you can show the team the problem rather than describe it.
FAQ
Questions, answered
What is the difference between a concept test and a product test?
A concept test judges an idea before the product exists: people react to a description, a picture or a video. A product test judges the real thing after people have used it.
Which kinds of product test do you run?
We use structured surveys, usability tests, tests in a central location, and focus groups or in-depth interviews, and pick the mix for the product and the decision. Tell us about your product and we will suggest a design.
Can testers show us the problem on video?
Yes. Video questions let testers answer on camera. Answers are transcribed, grouped into themes and checked for attention, relevance and lighting, and an AI interviewer can ask up to three follow-up questions.
When is a product test the wrong choice?
When the product does not exist yet, test the idea first with a concept test. When the product is on sale and you want to track how customers feel over time, a satisfaction and NPS study fits better.
Related
Goes well with
Have a product ready for real users?
Tell us what you have made, who it is for and what you need to decide. We will suggest the test and the users to recruit.
Nearest office: Orlando, FL (corporate office)