You present a design. Everyone nods. “Looks great, let’s move forward.”
Two days later, the PM pings you. “Actually, can we try a different direction? Sarah from product had some thoughts.”
You redesign. Present again. More nodding, more approval. A week later, the VP sees it. “This isn’t what I had in mind at all.”
Back to the drawing board. Third version. You’re frustrated, behind schedule, and starting to wonder if anyone in this company knows what they want.
This happens constantly
Designers blame stakeholders for being indecisive.
Stakeholders blame designers for not “getting it.”
Most of the time, the team just skipped alignment and went straight to execution. When you show a design before everyone agrees on what problem you’re solving, you’re asking people to evaluate a solution without shared criteria. Each person fills in their own version of what “good” looks like, and of course they disagree. They were never on the same page to begin with.
What “changing their mind” actually sounds like
“Can we make it more modern?”
They don’t have specific feedback, they have a vague feeling.
“This isn’t what I expected.”
They had an image in their head they never shared with you.
“Let’s try something completely different.”
They’re reacting to the design instead of evaluating it against goals.
“I showed it to [someone not in the project] and they had concerns.”
New opinions entering late with no context.
“I like it, but...”
They approved it too early and are now catching up with their actual reaction.
Why this happens
Most design projects skip a step. They go from “we need this feature” directly to “here’s how it looks.” The gap between those two things is where all the disagreement lives.
That gap contains questions nobody answered:
Who is this for specifically?
What behavior are we trying to change?
What does success look like?
What are we willing to sacrifice?
What’s out of scope?
When those questions stay unanswered, every stakeholder fills in their own answers silently. The designer builds something based on one set of assumptions. The PM has a different set. The VP has another. Everyone thinks they agreed, but they only agreed on a vague sentence. The specifics were never discussed.
The design becomes the first time anyone sees those hidden disagreements made visible. And it looks like the stakeholder “changed their mind” when really they’re seeing their own assumptions challenged for the first time.
Psychology insight
🧠 False consensus effect
People tend to overestimate how much others share their beliefs, preferences, and expectations. In a meeting, when everyone says “yes, let’s build this,” each person pictures something slightly different but assumes everyone else sees the same thing.
This is why verbal agreement feels solid but breaks down the moment something visual appears. The design forces hidden disagreements into the open.
Research shows this effect is strongest when the topic is ambiguous, which describes most product decisions perfectly.
What works instead
Slow down before the design starts. Spend 30 minutes getting alignment on the problem before spending 30 hours on the solution.
Write down the goal in one sentence. “We’re reducing checkout abandonment for first-time mobile users.” Put it at the top of every design file. If someone gives feedback that doesn’t connect to that sentence, you have a filter.
Ask each stakeholder what success looks like to them. Do this separately, before the kickoff. You’ll be surprised how different their answers are. Better to find that out before you design anything than after three rounds of revisions.
Present the problem before the solution. When you show a design, don’t open with the screens. Open with: “Here’s the problem. Here’s what we learned. Here’s the approach I took. Now here’s the design.” By the time they see pixels, they’ve already agreed on the logic that led there.
Three techniques that reduce direction changes
1. Get alignment in writing, not in meetings
Meetings create the illusion of agreement. People nod because it’s easier than objecting. Send a one-page brief before the meeting: the problem, the audience, the constraints, the success metric. Ask each stakeholder to confirm or push back in writing. Now you have real alignment, not polite nodding.
2. Show low-fidelity first on purpose
When you present a polished design early, people react to the surface. Colors, fonts, layout. Show rough wireframes first and you force the conversation to stay on structure, flow, and logic. The visual polish can come after the direction is locked.
3. Limit the feedback window
“Let me know if you have any thoughts” is an open invitation to change direction indefinitely. Instead: “I need feedback on this specific question by Thursday. After that, we’re moving to build.” Give people a container. Open-ended timelines produce open-ended revisions.
Two rules that help
First: Every round of feedback should reference the original goal. “This doesn’t feel right” isn’t actionable. “This doesn’t help first-time users understand the checkout” is.
Second: New stakeholders don’t get to reset the project. If someone who wasn’t involved in alignment wants to weigh in late, they get context first. Nobody evaluates a design cold and gives useful feedback.
Where this matters most
Projects with more than two stakeholders, where the more people involved, the more hidden assumptions exist.
Features where “the vision” was a verbal conversation that nobody documented.
Any project where you’ve done more than two major direction changes. That’s usually a sign of alignment failure, not design failure.
Stakeholders don’t change their minds because they’re difficult.
They change their minds because they never fully made them up.
Help them figure out what they actually want before you open Figma. The three rounds of revisions you’ll skip are worth the 30 minutes of upfront conversation.
Thanks for reading :) If something in this issue resonated, please reply and tell me.
Even one line. These replies are my favorite part of writing this newsletter.
See you next Thursday 🙌
— Balint
What else I’m working on?
Thanks for reading Balint’s Newsletter! Subscribe for free to receive new posts and support my work.
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.