How to build products people actually want to use
Quick reset from last session
Portfolio review from the homework
Why product thinking matters more in the AI age
What makes a product useful, trusted, and memorable
Real product reviews
Real product development lessons
Practical framework for your next build
Final action items
You built something
We ask a harder question
AI makes building faster. That is good.
But it also makes it easier to build:
generic products
copycat products
over-engineered products
products nobody needs
what to build
what not to build
what users care about
what makes a product survive
Homework from the last lecture: Build your personal portfolio website
A portfolio site is not just a website. It is:
your product
your story
your positioning
your trust layer
your first interview before the interview
A good portfolio should answer these questions fast:
Who are you?
What do you build?
Why should someone trust you?
What have you done?
How can I contact you?
“Does it look cool?”
Is it clear?
Is it useful?
Is it trustworthy?
Is it memorable?
Does it create action?
Within 10 seconds, a visitor should understand:
your name
your role
your strength
what kind of work you do
what they should do next
Student
Developer
Tech enthusiast
Software engineer building reliable backend systems
Flutter developer focused on mobile products
Product-minded full stack engineer interested in fintech tools
Hero section
About
Projects
Skills
Honors & Awards
Contact
Social links
Photo gallery
Blog / notes
Visitor counter
Language switcher
Dark / light mode
English
Kinyarwanda
French
visit counter
contact form
social media links
day/night mode
projects, skills, honors & awards
photo gallery
navigation bar
Language support is not decoration. It shows:
inclusion
awareness of real users
local relevance
product empathy
A visit counter is simple, but useful. It can:
make the site feel alive
show momentum
create a sense of activity
give you a reason to track traffic
A contact form reduces friction. Not everyone wants to:
copy your email
open another app
search for your LinkedIn
Social links help with trust. They show:
you are real
you exist beyond one page
your work is connected to public signals
GitHub · LinkedIn · X / Twitter · Behance · Dribbble · Medium
Dark/light mode is not only aesthetic. It shows:
care for user preference
polish
attention to experience
Projects are the strongest part of a portfolio. They answer:
Can you build?
Can you finish?
Can you explain?
Can you solve real problems?
What is it?
What problem does it solve?
Who is it for?
What did you build?
What stack did you use?
What was your contribution?
Where can I see it?
Do not dump random tools.
Organize skills clearly:
Languages · Frameworks · Tools
Databases · Cloud / DevOps
Design · Product / collaboration
This section matters because it shows signal. Examples:
hackathon results
scholarships
certifications
academic achievements
speaking invitations
leadership positions
event speaking
hackathon participation
product demos
team collaboration
awards
random photos with no value
visual noise
too many similar images
A clean navbar helps the user move fast. It should usually include:
Home · About · Projects
Skills · Honors & Awards
Gallery · Contact
too much text
unclear value proposition
no project depth
bad spacing
weak contact path
too much animation
generic AI-generated copy
no proof of work
What is strong?
What is confusing?
What feels real?
What feels generic?
What creates trust?
What is missing?
What would make me contact this person?
Live review section
We will review selected homework submissions and discuss:
clarity
trust
structure
proof
differentiation
next-step action
A product eye is the ability to notice:
what matters
what is noise
where users struggle
what creates trust
what makes people come back
Can I build this?
Should this exist?
Who needs it?
Why now?
Why us?
What is the smallest useful version?
code faster
design faster
write faster
prototype faster
problem selection
prioritization
workflow design
trust design
user understanding
Most weak products do not fail because the code is bad. They fail because:
the problem is weak
the use case is unclear
the user is vague
the value is generic
the workflow is annoying
the trust is missing
What can I build with AI?
What problem hurts enough?
Who feels it?
How do they currently solve it?
Why is that bad?
What would make life easier?
Good products usually solve:
pain
friction
confusion
waiting
repetition
uncertainty
coordination failure
Examples of friction:
too many steps · unclear wording
slow loading · hard onboarding
too much choice · low trust
bad defaults · poor follow-up
speed
convenience
trust
coordination
discovery
status
cost savings
better outcomes
AI makes it easy to build a demo. It does not make it easy to build a product.
cool feature
no workflow
no retention
no reason to return
no trust
no real user habit
a clear user
a clear job to be done
a clear first win
a clear loop
a reason to return
a reason to trust
a reason to tell others
The first user experience matters a lot. Ask:
How fast does the user get value?
What is the first useful moment?
Is it obvious?
Is it satisfying?
Trust comes from:
clarity
consistency
accuracy
working as expected
visible proof
not wasting time
not misleading the user
A product is not just screens. A product is:
inputs
decisions
actions
feedback
repeat behavior
The best products often feel:
simple · natural · smooth · quick · clear
That feeling is rarely accidental.
We will review familiar products using a product lens:
Uber
Duolingo
Google Maps
What problem does it solve?
Why do people use it?
What is the first win?
What keeps users coming back?
What builds trust?
What can we learn from it?
clear pain: getting a ride is hard
fast first value
simple main action
real-time visibility
less uncertainty
payment built into workflow
show status clearly
reduce uncertainty
remove unnecessary steps
make the next action obvious
integrate trust into the flow
low barrier to start
progress feels visible
small daily actions
strong habit loop
clear feedback
playful motivation
make progress visible
reduce task size
give fast feedback
reward consistency
design for return, not only first use
very clear use case
low learning curve
very high utility
strong network effect
fast communication
minimal friction
simple beats fancy
reliability matters
utility beats decoration
products that become habits become hard to replace
high trust
immediate utility
location-based relevance
strong workflow support
live updates
problem solving under pressure
context matters
urgency matters
relevance matters
accuracy matters more than style
low-friction content consumption
clear reward loop
easy creation
identity and status signals
algorithmic relevance
distribution matters
identity matters
social proof matters
users often come for utility and stay for emotion
Now let’s look at how strong products usually evolve.
Good products often start smaller than people think. They usually begin with:
one problem
one user group
one strong workflow
one sharp promise
launch
observe
learn
remove
simplify
improve
build in isolation
add 50 features
hope it works
Often the best improvement is:
fewer steps
better copy
better defaults
faster loading
better onboarding
better trust signals
Even a good product can fail if:
nobody sees it
nobody understands it
nobody trusts it
nobody has a reason to try it
In the AI age, product builders need to do more than code. They need:
judgment
taste
prioritization
user empathy
communication
iteration discipline
problem selection
user understanding
workflow thinking
product writing
prioritization
technical execution
iteration discipline
Is this problem real?
Is it painful enough?
Is it frequent enough?
Is it clear who has it?
Would solving it matter?
observe real users
talk to people
study weak products
study why people switch tools
write problem statements before building
You do not need perfect research. But you do need to know:
who the user is
what they want
what frustrates them
what they fear
what makes them stay
interview users
watch users use products
read product reviews
study complaints
compare before and after flows
A feature alone is not enough. Ask:
What happens before this?
What happens after this?
What is the user trying to finish?
What is blocking them?
draw user journeys
reduce the number of steps
identify dead ends
identify uncertainty points
identify trust gaps
Words matter a lot. Products fail because of bad writing too:
unclear labels · confusing buttons
weak onboarding · vague value props
generic headlines
Practice writing:
headlines
CTA text
onboarding copy
empty states
error messages
trust explanations
You cannot build everything. Strong builders ask:
what matters most?
what creates value fastest?
what is truly necessary?
what can wait?
Useful filters:
user pain · business value · effort
trust impact · retention impact
frequency of use
Taste is not just visual design. Taste is knowing:
what feels clean
what feels useful
what feels excessive
what should be removed
what makes the product feel mature
review strong products
compare good vs bad flows
study simplicity
notice what feels heavy
ask why one product feels better than another
Do not fall in love with version 1.
test
learn
adjust
simplify
repeat
As a student, do not wait for a big company. Practice now by:
building small products
reviewing real products
writing case studies
improving your own portfolio
observing real users
explaining product decisions
review 1 real product
redesign 1 weak flow
talk to 1 user
build 1 small feature
rewrite 1 landing page
improve 1 part of your portfolio
Every time you use an app, ask:
why am I still here?
what annoyed me?
what felt smooth?
what felt trustworthy?
what would I improve?
A lot of students build too much and observe too little. Better:
fewer projects
better projects
clearer projects
more reflection
more iteration
prototype
summarize
compare options
generate variants
speed up execution
judgment
taste
empathy
ownership
a real problem
a clear target user
a useful first version
thoughtful decisions
visible iteration
clear explanation
Before writing code, answer:
Who is this for?
What pain are we solving?
Why does it matter?
What is the smallest useful version?
Why would users return?
While building, keep asking:
Is this clear?
Is this necessary?
Is this fast?
Is this trustworthy?
Is this better than the current alternative?
After launch, ask:
Did people understand it?
Did they get value fast?
Did they come back?
Where did they struggle?
What should be removed?
Before calling anything “done,” check:
clarity
speed
trust
usability
value
proof
return reason
Your assignment: Improve your portfolio using today’s lecture.
English · Kinyarwanda · French
visit counter · contact form · social links
day/night mode
projects · skills · honors & awards
photo gallery · navbar
Before submitting your revised portfolio, ask:
Is my value clear in 10 seconds?
Are my projects convincing?
Is my contact path easy?
Does the site feel trustworthy?
Does the experience feel intentional?
Choose one real product and write a short review:
What problem does it solve?
Why do people use it?
What makes it good?
What makes it weak?
What would you improve?
Build or improve one small product experience. Examples:
onboarding flow
project page
contact experience
search flow
dashboard screen
progress tracker
In the AI age, building is easier. That means your real edge is no longer just coding.
Your edge is:
judgment
clarity
product thinking
trust building
execution
A2SV · University of Rwanda · Emre Varol