Product

Discovery
Recipes
Jeff Patton, jeff@jpattonassociates, jpattonassociates.com/handouts
 1
Table of Contents:
Product Discovery .......................................................................................2
Product Discovery Team ........................................................................ 3
Validated Learning Cycle ....................................................................... 4
Create an Evidence Board......................................................................5
Create Simple User Models ................................................................... 7
Map User Experience ............................................................................. 10
Ideate With Your Team .......................................................................... 11
Identify Questions & Tests ................................................................... 13
Structure & Plan Your Interview .................................................. 15
Note-Taking During the Interview............................................... 16
Synthesize Notes After the Interview ....................................... 17
Distill Learnings in a Delta-Next .................................................. 18
 2
Product Discovery
If you’ve been involved with software development for any length of
time, you probably know that there are always way too many ideas,
and way too much to build. But, that’s not the worst part. The really
tough part is that:

Most of the things you could build are a waste

of time
That’s not hyperbole. At least I don’t think it is. The failure rate of
technology startups is 85-90% depending on the study you read. That
data gets a little sketchy when thinking about features or capabilities
we add to software. But experts like Marty Cagan who lead product
at eBay for over a decade explains that “Typically about 50% to 80%
Use discovery
of all software we ship fails to accomplish its objectives.” I’m going to
trust his experience on this.
work to test before
But, there’s something we can do about it. you fully invest.
We use product discovery work to get evidence that the problems
we’re solving really exist, that our customers want and can use our
solutions, and that we can build them predictably.

Use discovery work to test before you fully

invest.
We can’t always be right, but at minimum we can validate that we’re
putting our time and money behind ideas that stand a reasonable
chance of succeeding.
 3

Product Discovery Team

Discovery checklist:
Discovery work goes best when you work together with a team that
combines a mixture of knowledge, experience, and skills. We’re There’s a paradox here where focusing on the mechanics
yields better results. But focusing on the results without
seeking a solution that’s: good mechanics causes teams to struggle to make progress.
§ Valuable to your Your week of focused discovery will go better if you do these
organization and your things during your week:
customers 1. Create and maintain an evidence board
§ Usable by its customers
2. Talk with at least 10 users, customers, or
§ Feasible to build given
other subject matter experts this week
the time and skills your
team has 3. Try a type of experiment you’ve never
To do this you’ll need a tried before
discovery team that includes 4. Conduct at least one ideation/design
people with business, user, and technology experience. They must be sketching session with your team
able to:
5. Express your current beliefs as a
§ Work with stakeholders and subject matter experts in your hypothesis statement
organization
§ Identify candidate customers and users 6. Create a learning backlog: an explicit
§ Plan and facilitate interviews and tests list of risk, assumptions, and questions
§ Design user experience 7. Track your hypothesis changes (your
§ Create simple prototypes deltas)
§ Create functional or live-data prototypes
8. Kill off at least one idea, or part of one
You’ll be working together quickly and collaboratively. At times
you’ll break up work so you can work independently or in pairs, then 9. Break into pairs to work on several
come back together to rebuild shared understanding of your problem discovery tasks concurrently
and solution so far.
 4

Validated Learning Cycle

Discovery work follows a validated learning cycle. As a
discovery team, be mindful of where you are in the validated
learning cycle at any given time.
1. Declare your hypotheses: Hypothesis are your
beliefs about:
§ problems you’re solving for specific customers and
users
§ solutions you believe will solve them
§ outcome – specifically how customers will react and
use your solution
§ impact – how customers engaging in the behavior
your expect ultimately benefits your organization
2. What do you need to learn? Identify your risks,
assumptions and questions
3. Create a test plan
4. Run your test
5. Synthesize your results and update your
hypotheses
 5
Create an Evidence Board
Tips & Tricks:
Every good team of detectives in every crime show builds an Keep it fresh: After every round of
evidence board. You should too. It’ll help you build shared experimentation add new information and remove
understanding, and move faster. the stale stuff
Find a large visual space in your team area. On it display any or all Compress it: Photograph flipchart paper and
of these things. But try to include something from every category.
reprint it letter sized
Stack it: Interview notes and other information
you don’t need frequently can be stacked
 6
Your Hypotheses: ¨ Metrics that could measure the success of your
solution
Distilled Hypothesis Statements
We believe that: ¨ Inspiration: other products that solve problems
similar to your customers’ in interesting ways
[these customers or users] have [this problem or need], and
If we build/make [this solution/feature/capability]
they will [observable and ideally measurable behavior] Your work:
resulting in [measurable business benefit] Learning Backlog
List and rank:
You may have one or more hypothesis statements. Your
statements may be incomplete. For example, you may only § Risks: what might go wrong
understand the problem, but not yet have a solution. § Assumptions: what must be true
Problem Hypotheses § Questions you’ll need answered to make decisions

¨ Simple user profiles or personas Your next most important thing to learn
¨ Users journeys as they are today What’s at the top of your learning backlog?

¨ Sketches or examples that help you describe the Your ideas for next best test
challenges users face today Ideas for next experiments
¨ Examples of other tools your customers use to Discovery Task Board
solve their problems today Tasks your team has identified to do, are doing, and
Solution Hypotheses have done

¨ Solution sketches or solution storyboards Notes from past interviews and tests
¨ Solution journey maps or journey sketches What You’ve Learned
¨ Architectural or technical sketches for the Deltas: specific learnings that changed your hypotheses
technology you’ll need for your solution
 7
Create Simple User Models
Simple user profiles or personas collect what you believe about your
Start with assumptions. Use
users. It’s OK to start with your guesses or assumptions, so long as discovery work to test them.
you use discovery work to validate those beliefs. Replace them with facts over time.

Identify user types Gather general Personify to create

Just because users use the same information in a Profile examples
functionality, doesn’t mean they’re Start by collecting information about Select specific profile details to
the same type. For example: someone your users: create an example user your team
might use a simple planning can understand and empathize with.
§ General information
application like Trello to plan their
§ Behavioral information Use personas as a design
next vacation. But that’s a different
§ Motivations, pains, obstacles, targets. Focus on delighting this
need than a team member at a
technology company using the same and rewards type of person, knowing that similar
tool to manage work done with their users will also be happy.
Focus on relevant information. Ask
team. yourselves: “would knowing this
Identify different user types based change the way we build a solution
on differences that make a for this user?” If the answer is “not
difference in how they’d use likely” then discard it.
your solution: Information should pass the
§ Goals, motivations, relevancy test: Would
responsibilities
knowing this change the way
§ Place, time
§ Collaborators
we design or build this
solution?
 8
User characteristics to Consider
General information:
§ Age, gender, education, economic background
§ Subject matter expertise
§ Technology skills

Behavior:
§ Activities or work which users engage in routinely where
your solution is relevant
§ Frequency: how often might they use your solution?
§ Collaborators: do they work with others to do the work? To
share information?
§ Locations: where are they when doing things that might use
your solution?
§ Other tools they use in conjunction with your solution,
other tools they like using to get things done
§ Alternatives and work-arounds – the things they
use instead of your solution

Motivations:
§ Pains: what pains are your users trying to avoid?
§ Obstacles: what problems get in the way of users
accomplishing their work?
§ Rewards & joys: what makes your users happy, delighted,
pleasantly surprised?
9
 10
Map User Experience
Use simple journey maps or story maps created on sticky
notes to describe how your users do things today. Later, use
the same map to describe how you envision them doing things
with your solution.

1. Start by identifying users and activities

Identify the types of users you’d like to focus on
Identify the activities your users will be engaged in
Identify a starting point and an end point for your
story

2. Tell users’ story using steps

Organize steps into a narrative: remember your
narrative is never perfect for all users, start with a typical
user’s journey
Get to the end of the story, then go back and add alternatives
and variations
4. Add details and context:
Activities Other details and context may help you better understand
Activities help categorize lots of steps that help users reach a your user’s story:
bigger goal § Frequency: how often do users do this step? Many
3. Add pains, and rewards times daily? Daily? Weekly? Monthly? Hardly every?
§ Duration: how long does a step take? A series of
Where in the flow do users encounter big challenges?
steps? An activity?
Problems? Where’s the fun? The payoff for doing the work?
§ Location: where are users when they do this?
§ Tools: what technology or tools do they use today?
§ Other people: who else is there?
 11

Ideate With Your Team

Your first idea is rarely your best. One way to identify better Tips and tricks:
solutions is to identify more. Do this collaboratively with your team.
Crazy 8s: Fold tabloid paper in half 3
Combine and refine everyone’s solutions to come up with something
times. Unfold and you’ll have 8 squares.
really great.
Sketch 8 ideas in 8 minutes
Who: Find a comfortable corner: Step
Involve your whole product team, subject matter experts, or others away from your desk and find a comfy chair
that are interested. or place to sit
Someone from your discovery team should lead. Start with magical thinking: Start
Allow 60-90 minutes depending on the size of your group. with a first round of sketching where all
ideas must use magic or impossible science.
Supplies: These ideas loosen participants up and keep
§ Paper to sketch on: sketching templates or tabloid-sized paper things fun. Follow with a next round of
§ Pens for sketching feasible ideas. You might find that a feasible
§ Tape for posting sketches on the wall idea is a small adjustment to a magical idea!
§ Highlighter pens or sticky dots for marking important ideas
 12
Discuss users and their Sketch Converge & design
challenges § Set a timebox – 5-15 minutes Use sticky dots or a highlighter pen to
Discuss the problem-side of your § Everyone take sketching paper mark favorite ideas: try giving
hypotheses: and quietly sketch their ideas everyone 3 or 4 votes to mark their
§ Users or personas § Try to sketch as many ideas as personal favorites
§ Journey maps that explain how possible You could:
they do their work today § Repeat: sketch for another
Share out & discuss
§ Challenges that make round including and building on
accomplishing their goals § Post everyone’s sketches on the
ideas you got from others
difficult wall
§ Form small groups of 2
§ Motivations that make them § Set a timebox – 15-30 minutes
or 3 to combine and
want to use your solution Person by person: refine ideas in a more detailed
Discuss inspiration § When it’s your turn, explain drawing.
§ Do users have creative ways of your favorite idea – try
Ultimately your best designer(s) will
solving their problems today? telling a story about how your
take all this information and converge
§ How have other products solved user would use your solution
the best ideas into a design to review
this problem? § As a team, discuss the later with the team
idea: Ask clarifying questions.
Write a focusing statement Express opinions from the
Try this template or invent your own: perspective of your target user:
“Clair, our user, would love this
How might we help [customer/user]
To [reach a goal, perform some activity] because…” or “How would Clair
When [What makes it hard? rewarding?] use this to handle this problem…”
§ Repeat with every person
explaining their best, then again
with their second best and so on
until the time-box is done.
 13
Identify Questions & Tests
Work together to plan your next interview or test
Use flipchart paper or a whiteboard
Use a 2 column approach:
§ What we want to learn
§ What we’ll ask
Brainstorm ideas for each column
Connect what you’d like to learn to what you’ll ask:
some experiences you ask for answer many questions, or get at
questions in different ways
Distill and prioritize: distill and simplify questions you’ll ask.
Decide on the most important questions to ask.
 14
What you want to learn: What you’ll ask:
About them: Ask for experiences and examples: Instead of
asking what they like about a product, try eliciting a story
§ Are there characteristics or skills you’d
using prompts like “can you tell me about the last time you
expect them to have?
used it?” or “tell me about a time you really struggled using
§ What are challenges and frustrations you it?” or “a time when using it really helped?” From their
believe your users have relevant to your product? stories, you’ll learn what they do and don’t like about your
What are motivations or joys they seek relevant product.
to your product? Can you show me?: For your interview subject,
About what they do: showing and telling will help them explain things faster,
and you’ll learn more. Ask if they can show you in the
§ How do they accomplish things today using or not
product, show you an example, or even draw you a picture.
using your products?
§ Would they use your product? How? Observe to test
Orchestrate a way for your customers to
use your product. Or even just see your product, a
prototype, or example. Their body language, facial
expressions, and the involuntary sounds we all make when
we like or dislike something will tell you what you need to
learn.. So, pay close attention.
 15
Structure & Plan Your Interview
With your team discuss and write down an outline of your interview.

Structure Roles: Decide who will be the primary

Most interviews and tests go a little like this: interviewer, and who will be the
Intro Introduce yourself and your group. Explain why you’re seconds – the seconds are responsible for
talking with them today, and thank them for their help. most of the note taking while the primary
interviewer leads discussion.
Background Learn a little about them and their relationship with
your product today.
Understand Ask for stories that help you understand what they do Imagine if: When asking subjects to
with your product and about other ways they meet their evaluate a new solution or idea, ask an
needs. “imagine if” question like “Imagine
Test or Bridge: “we’ve been working on some ideas for the when you had the challenge you described
Imagine if before that you saw this..”, or “imagine if
challenges you discussed…” then: “imagine when you
you had this feature, how would you go
were in that situation you saw this…”
about using it?” Pay attention to how easy
Show a solution to get feedback on value. or hard it is for them to imagine and tell a
story about using the solution.
Test to get feedback on usability.
Wrap up Give note-takers a chance to ask clarifying questions.
Give them an opportunity to ask you questions.
Give them a thank you gift for their time. Ask if you
can contact them again if you have follow-up questions.
Let them know again how valuable their help has been.
 16
Note-Taking During the
Interview
Take Notes
Use a notebook or sticky notes, or place sticky notes directly in your notebook to write
on. Having ideas on separate sticky notes will speed up synthesizing with other interviewers later.
Don’t use a computer: unless your subject can’t see or hear you. The clicking of the keys can
be distracting. Your subject will quickly figure out that lots of key clicking means you think they
said something interesting. A tablet computer that lets your write using a pen may be a good
substitute if you’d like to stay electronic.

Watch for: Write down:

Body language: their body language will often reveal Facts about your subject
how they feel, even if they’re not saying. Hand-gestures
§ Name, gender, age (just your guess)
reveal they’re talking about something they’re recalling or
have strong feeling about. § Their job role (if it’s relevant)
§ General characteristics about them
What they actually do: when using your solution or
someone else’s, watch for things they do that they don’t What they do: Listen for verbs. Anything they do
explicitly tell you about. to reach a goal or solve a problem is a clue to what your
product could or should do.
How they feel: Any adjective that describes how
they felt is an important clue to find problems to solve,
or motivations your subject has.
Direct quotes: direct quotes in their language are
the best way to express how they feel and think. Write
down exactly what they said especially when
expressing emotion or a strong opinion.
 17
Synthesize Notes After the
Interview
Review and tidy your notes. You may be surprised at the
number of incomplete thoughts, and sentences, or confusing
writing.
Highlight your best or most important notes – those you’d
like to share.
Transfer them to sticky notes, if you didn’t write them
on sticky notes to start with.
Dump and discuss
Set up to catch notes Place a sticky note on the 2x2 and say what’s
For each interview, use this on your note (often we say more than what we
2x2 to collect notes wrote)
About: general information about If others have a similar note, now’s a good time
your subject such as name, age, job, for them to jump up and add theirs and say
other general information. what they heard
What they do: What did they say Each note-taker take turns placing and
they did? What did you observe them describing a few notes until all notes have been
doing? dumped into the 2x2
How they feel: Pains, needs, Sort and highlight
goals, joys and rewards… Discuss and highlight the most important
What they say: direct quotes notes
about anything they do or feel – “I Note insights: ah-ha’s or learning that comes
hate it when…” “I love it when..” from seeing a pattern across interviews and tests
 18
Distill Learnings in a Delta-Next
You’ll take lots of notes and learn something from every interview or
test. But stop and reflect on how what you’ve learned has changed
what you believe.
Use a sheet of flipchart paper to capture deltas and next steps.

Ö What did we gain confidence in? [We believed it before, but now we
have stronger evidence.]

D- What were we wrong about? [Busted myths]

D+ What is new information that changes our hypothesis?
? What questions do we have that we’d like to address in a next
round of tests?
What are ideas or opportunities we should discuss in another
round of discovery? These could change our current idea, or be
something else to focus on at a later time. Adjust hypotheses
Convert deltas to questions: As you discuss
Dump and discuss some of the deltas, you might decide that you don’t
Individually write down what you believe are deltas, and next steps. have enough evidence yet to really change your
Take turns placing and discussing your sticky notes. hypothesis. So, it may make sense to convert them into
questions to ask during a next round of testing.
Change your hypotheses: Discuss how this
changes your beliefs and then adjust your hypotheses
and supporting artifacts such as personas, journey
maps, and solution designs.