The first screen was empty and silent
A new user landed on a dashboard with no data and no direction. An empty state is a design decision, and this one had been left to chance.
UNIBRIGHT SOLUTIONS · UI/UX · CASE STUDY
They Signed Up. Then the Empty Screen Sent Them Away.
New users landed on an empty dashboard and left before they saw what the product was for. Redesigning the first ten minutes lifted users reaching first value 58% and cut day-1 drop-off 37%, with first value in 4 steps instead of 12.
+58%
REACHED FIRST VALUE
−37%"
DAY-1 DROP-OFF
4
STEPS, NOT TWELVE
0
EXTRA TRIAL SIGNUPS NEEDED
01 / THE PROBLEM
The Problem
The product was good and the signups were fine. The problem sat in the hours right after someone joined. A new user logged in for the first time, landed on an empty dashboard, and had no idea what to do next. No data yet, no obvious first move, just a blank canvas and a menu of features that meant nothing without context.
A trial only converts once a user sees the thing the product is for — the first chart that tells them something they did not know. Most users never got there. They poked around, hit a setup step that asked for a data connection they were not ready to make, and closed the tab. The paid trial was quietly failing in its first ten minutes, long before anyone reached a pricing page.
The team had been adding features to win demos. The users were not losing on features. They were losing on the very first screen.
REVIEW
What Our Review Found
We watched first-session recordings, mapped every step from signup to first insight, and timed how long it took to reach something useful. Six problems.
Setup demanded the hardest step first
The onboarding asked for a live data connection before the user had seen a single reason to trust the product with it. The biggest ask came before any payoff.
Twelve steps to a result that needed four
The path to first value wandered through settings and options most users did not need yet. Every extra step was another chance to give up.
Features were shown before they meant anything
A full navigation of tools greeted someone who did not yet understand the core loop. Choice without context reads as confusion.
No sample of the payoff
Nothing let a user see what a finished dashboard looked like before doing the work to build one. They were asked to invest on faith.
Progress was invisible
No sense of how far along setup was or how close the payoff was. People abandon a task far sooner when they cannot see the end of it
BUILT
What We Built
Designed the empty state as a first screen, not an accident
The blank dashboard became a guided start: one clear next action, a line on what the user is about to get, and everything else out of the way until it matters.
Moved the payoff before the heavy lifting
New users now see the product working on sample data first — a real, populated dashboard — before being asked to connect their own. They understand the reward before they are asked to earn it.
Cut the path to first value from twelve steps to four
We stripped onboarding to the shortest route to one useful insight and deferred everything else to later, in-context prompts. Settings can wait; the payoff cannot.
Revealed features as they became relevant
The navigation now unfolds as the user progresses instead of confronting them with everything at once. The product teaches itself in the order it is used.
Added a visible progress path
A simple checklist shows what is done, what is next, and how close the first result is. People finish tasks they can see the end of.
Wrote guidance in plain language, in place
Short, human prompts at each step — what to do and why — instead of a help centre the user has to go find. The instruction arrives where the hesitation happens.
IMPACT
Where AI Helped — and Where It Did Not
AI did the pattern-finding. It grouped thousands of first sessions, timed where users stalled, and clustered the exact steps people quit on far faster than a team reviewing recordings by hand. That analysis is available to anyone now; the tools are not the advantage.
A designer decided what to do with it. That the fix was showing the payoff before the setup, not adding another tooltip to the setup. Which eight steps to cut and which four to keep. What “first value” even is for this product. AI can show you the cliff people fall off. It cannot tell you the answer is to move the reward, not to signpost the fall. That decision is the job.
05 RESULT
The Results
RESULT
DETAIL
Reached first value up 58%
Share of new users who hit their first useful insight, first month after launch [confirm window]
Day-1 drop-off down 37%
Fewer users leaving in the first session and not returning
First value in four steps
Down from a twelve-step path to the same result
Payoff shown before setup
Sample-data dashboard now precedes the data-connection ask
Guided empty state live
The blank first screen became a directed start, not a dead end
No extra signups required
Every gain came from converting the users already arriving
MEANS
What This Means
Activation is a design problem, not a feature problem
Users rarely churn on day one for lack of a feature. They churn because the first screen did not show them what to do. That is fixable in the interface.
Show the reward before the work
Asking for effort before showing value is backwards. Let people see the payoff on sample data, then they will happily do the setup.
Every onboarding step is a place to lose someone
The shortest honest path to first value wins. Steps that can wait should wait, in context, after the user is hooked.
An empty state is a first impression
The blank screen a new user meets is one of the most important in the whole product. Treating its design as an afterthought costs you the user before they have started.
Progress you can see is progress you finish
A visible path to the payoff keeps people going. Invisible progress gets abandoned early.
FIX
What This Does Not Fix
Worth being straight about:
A product without a clear core value
Onboarding can only lead people to the payoff if there is one. If the value itself is unclear, no flow will rescue it.
The wrong users signing up
A smoother start converts the right-fit trials. It cannot turn a poor-fit signup into a customer — that is a targeting question upstream.
Pricing or packaging that does not land
Getting more users to first value fills the top of the funnel. If the plan or price is the blocker, that is a separate fix.
Deeper retention past week one
Reaching first value is the start of the relationship, not the whole of it. Keeping users for month three is a different, ongoing job.
WORKS
Why This Works Elsewhere
Any product with a free trial or a self-serve signup has some version of this. The tell is a healthy signup number and a stubborn gap between joining and doing anything useful. People are arriving and then vanishing in the first session.
The fix is rarely another feature. It is designing the first ten minutes so a new user reaches the payoff fast, sees it before they have to work for it, and always knows what to do next.
NEXT STEP
The Next Step
Give us access to your product analytics and a session-recording tool. You get back the exact step where new users drop, a ranked list of what to fix first, and the activation number each fix is meant to move.
Services Provided:
UI/UX audit, Onboarding and activation design, Empty-state design, First-run experience, Product analytics and session analysis, In-product guidance and microcopy, A/B testing, Reporting