Hypothesis builder
Write down one belief about your product. Then six plain questions, one at a time, turn it into a test, with a number that could prove you wrong and a date to decide.
- Free, no signup
- Saves in your browser
- A worked example on every question
What you'll end with
- The belief
- People forget to attach files, and send a second email to fix it.
- The change
- A check that asks "Did you forget the file?" when an email says "attached" but has none.
- We're wrong if
- Second emails with the file drop by less than a quarter.
- We decide on
- Four weeks after launch.
Build your hypothesis
Your hypothesis
Turn a belief into a test, one question at a time.
Question 1 of 7
What do you believe?
One belief your idea depends on, that nobody has checked yet.
How Gmail's attachment check answers this
People forget to attach files, and send a second email to fix it.
Question 2 of 7
What am I adding or changing?
Name the one thing you would build or change.
How Gmail's attachment check answers this
A check that asks "Did you forget the file?" when an email says "attached" but has no file.
Question 3 of 7
Who is it for?
Name the people whose behaviour you are testing, not "users".
How Gmail's attachment check answers this
Everyone who sends email.
Question 4 of 7
What should people do differently?
Something you could see them do. Not "be able to".
How Gmail's attachment check answers this
Add the file before they press Send, instead of sending a second email.
Question 5 of 7
Why would they?
The reason your change works.
How Gmail's attachment check answers this
The check catches the slip at the moment it happens.
Question 6 of 7
What result would prove it useless?
Count the thing that matters, and pick the number now.
How Gmail's attachment check answers this
Second emails with the file drop by less than a quarter.
Why that number: Below that, the check stops people more often than it helps them.
Question 7 of 7
On what date do I decide?
A real day. On that day you decide, whatever the numbers say.
How Gmail's attachment check answers this
Four weeks after launch.
Your hypothesis
This is what people mean by a hypothesis: every answer, together. You wrote each part of it.
Your saved hypotheses
What is a hypothesis, and why six questions?
A hypothesis is a belief your design depends on, written with the result that would prove it wrong. A belief like people want group ordering
can survive any test, because nothing you see will ever count against it. Add what people should do, a number, and a date, and a result can come out against you. That is the point: a test is only worth running if it could prove you wrong.
Most teams learn the one-sentence template with a gap for each part. It is hard to fill in, because it asks for six decisions at once. This builder asks them one at a time, in plain words, and puts the sentence together at the end. Two of the six are where most tests go wrong: the number that proves you wrong, which needs a reason, and the date, which needs to be a real day.
The trap it guards against is confirmation bias: trying only what you expect to work. A close cousin is the false-consensus effect, where a team assumes users think the way it does.
Questions
What is a UX hypothesis?
A belief your design depends on, written with the result that would prove it wrong. "People will like the new checkout" is a belief. "Fewer than 3 in 10 people who start the new checkout finish it, by 30 October" is a hypothesis: a result can come out against it.
Why six questions instead of a template?
The one-sentence template, with a gap for each part, is hard to fill in: it asks for six decisions at once, in research words. Asked one at a time, in plain words, each part is easy. The sentence is only the summary at the end.
What makes a good "we are wrong if" number?
It counts something people do, not something they say, and it comes with a reason. "Fewer than half the managers use the link twice" is stronger than "half the people do not open it", because opening is not using. The reason says what the number means for people, so the team can argue with it.
How long should a test run?
Until a date you set before it starts. On that date you decide, whatever the numbers say. "At least a month" or "until we have enough data" has no end, so a test can run for a year because the number never looks quite good enough.
What is the difference between an assumption and a hypothesis?
An assumption is a belief your design depends on that nobody has checked. A hypothesis is the same belief rewritten so that one result could prove it wrong, with a date to decide. Every hypothesis starts as an assumption.
Where are my hypotheses saved?
In this browser only. Nothing is sent anywhere, and there is no account. Copy a hypothesis to keep it somewhere else, or save it to the list on this page, which stays until you clear your browser data.
To see how people really behave once your test is running, the usability testing guide covers watching five people use it.