Goal of user discovery:

→ reduce the risk of your product bets

→ solve your ICP problems better than anybody

don’t do Discovery before delivery. do it before,

during and after.

Mix multiple types of discovery

- Quantitative (pure data)

- qualitative (e.g. user interviews)

- Observational (e.g. how users interact with

your product)

It can be easy to take bad decisions if you only

rely on quantitative data

Qualitative & observational discovery are the

the best types and should be your main method

sometimes, the best discovery is shipping fast

interview skills are underrated

Goal of user interviews:

- understand their motivations

- understand their goals

- understand their problems

- understand where you can help them the most

always look for fresh insights

make the interview feel like a conversation

Ask questions about specifics in the recent past

don’t ask about memories

Ask neutral questions

don’t ask for their ideas

don’t share your ideas

Let them speak, let them express their feelings

focus on the exact process they went through

leave blanks to make them speak more

go very deep in the details

you want to gather facts, not compliments

Make them tell you a story of their actions

Couple user discovery with observations

→ it’s common for people to act differently

Rephrase their answers and continue asking

find patterns between discovery

Don’t act on single pieces of feedback

know who your ICP is, you can’t build a good

product for everybody

If you don’t have consistent problems and goals

you don’t have a specific enough customer segment

Don’t 100% rely on discovery, take risky bets too

be good at reading between the lines

Don’t rely on survey, they are often overrated

survey open ended question will exclude all the

people that don’t feel like writing

repeat their last words to make them continue

make them show you their problems (screen share)

focus on peeling back the onion until you

get to its core

label their emotions

suggesting a certain angle for their replies

asking what feature you should add

asking what they think of your ideas

Asking about action they would take

Asking about things in the future

Asking something about what they think

What are the top pains when {{situation}} happens?

When last did you face {{problem}}?

what happened next?

how are you currently dealing with {{problem}}?

Why are you doing it this way?

What else have you already tried?

How important is this problem to you?

What happens if you don’t solve this problem?

What’s the worse that can happen?

What’s your goal?

What are you responsible for in your company?

What metric are you held accountable for?

What’s your main focus in your team?

What metrics do you report to?

What projects are you working on to try to

reach this result?

How did you feel when {{situation}} happened?

what results would outperform your goals?

how does this fit in your overall process?

Who in your team is also involved?

what results do you consider good?

why would you need this?

why do you need this now?

→ don’t prepare too many questions. focus on

asking questions based on the replies you get

segment your users based on:

- product retention (months)

- revenue generated (cash)

- results/usage (depends on your product)

- usage frequency (DAU, WAU, MAU)

- pricing plan they have

cold outreach them

leverage marketing assets, e.g.:

- community

- newsletter

- events

- content (e.g. help center)

add popups when they use your product

contact people behind reviews (e.g. g2, trustpilot)

contact mentions on social media

ask your team to put you in contact with users

They often have the best contacts

Talk to people that didn’t convert or churned

Try to automate setting meetings to save time