Altay Suna

Product designer since 2008, ex-head of product design at eBay Türkiye. I work with startup founders and product teams, close to whoever makes the calls.

[Altay Suna @rotudam]

TuckerFox was already a working food delivery product. I had four weeks, I was the only designer and I worked directly with the founder.

I wrote the full case study here, including what changed, what shipped and what I would still change today. https://t.co/lD6QNC7MFk

[Altay Suna @rotudam]

@staycoolxiao Can't share the details because of an NDA but it was a live football feature in a smart TV app for a global brand :)

Can't wait your DAD TEST results if you'd like to share here :)

[Altay Suna @rotudam]

@kylekane Exactly 👌

[Altay Suna @rotudam]

Really liking how Claude Code uses these subtle color indicators. Such a nice little touch.

[Altay Suna @rotudam]

Great idea! I've been working on a project and wasn't happy with the direction AI was taking.

Your tweet gave me an idea, so I just gave three different AI coding agents the same client brief and asked each to come up with its own solution from scratch.

I didn't predefine any layouts, interactions or visual direction. Really curious to see what they come up with and how their solutions compare to my original approach.

[Altay Suna @rotudam]

Here's a fun experiment. Generate a design with your favorite AI tool. Then ask the same tool to brutally critique its own work. You might be surprised by how many problems it finds.

That's why you still need someone who can make design decisions, explain and defend them, see how they perform in the real world and change their mind when the evidence calls for it.

[Altay Suna @rotudam]

Not directly related but there's something I always tell clients when we start working together: "Please don't soften your feedback just to be polite or avoid hurting my feelings. If you hate something, say you hate it. If you think it's terrible, tell me. You don't need to find a nicer way to put it."

I usually bring this up in our very first meeting and remind them from time to time. I'd much rather hear what they actually think than spend time trying to figure out what they really meant.

[Altay Suna @rotudam]

As someone who makes a living designing products, I hate to admit it, but I think I'd love to see this become the norm. Especially if users could build exactly what they need without waiting for the product team to prioritize it.

PS: Haven't experience those on my own but it's worth to check Canva Code, Notion's Custom Agents and Salesforce's Agentforce Vibes for a few examples.

[Altay Suna @rotudam]

@hyeokStory Exactly. They just leave and analytics logs another bounce.

[Altay Suna @rotudam]

@dijedont Exactly! We used to do this all the time when I was at eBay Türkiye, both with people from other teams and employees at neighboring companies in the same building. Dad test rocks :)

[Altay Suna @rotudam]

I have this little thing I call the "dad test" Not sure if there's an actual name for it in UX :)

Whenever I'm designing something for a broad audience, I'll casually hand my dad a prototype and ask him to do something. Half the time he thinks it's a real product. He's not a power user, he doesn't care about impressing me and he'll happily point out anything that makes no sense to him.

I've done this once with smart TV apps while he's sitting on the couch eating tangerines. Sometimes I'll even run downstairs and ask a neighbor to try something.

Photo-to-video isn't really my area, so I wouldn't pretend to know what to remove. But I'd probably start with someone at home, a sibling or even a regular at my local café. Give them a task, watch what happens and resist the urge to explain anything.

[Altay Suna @rotudam]

Slightly unrelated but one thing I hear some designers complain about lately is actually making my job easier.

I used to ask product teams to sketch their ideas on paper (no matter how bad their drawings were) because so much could get lost in translation.

Now they can just show me a working prototype. I can understand what they have in mind immediately, keep the good ideas, make the design decisions and we still have something tangible to discuss with developers.

I genuinely like this change.

[Altay Suna @rotudam]

Instagram and Flickr pretty much answered this question years ago :) Instagram (originally Burbn) started as a check-in app but people were more interested in sharing photos. Flickr came out of an online game. Both ended up building products around something quite different from the original plan.

[Altay Suna @rotudam]

I think this overlooks a bit of consumer psychology. Curiosity and perceived scarcity can make people want something they wouldn't normally consider buying.

Restaurants and nightclubs have used waiting lists for years, sometimes deliberately. You can have a perfectly ready product and still benefit from making people wait.

Of course, if your actual users and conversion data tell you otherwise, that's a different story. But I wouldn't dismiss waitlists just because building software got faster.

[Altay Suna @rotudam]

@matt_teeixeira Exactly. If we go down that road, you could argue that half the products we use today are just "you can do that in Excel" with a nicer interface :)

[Altay Suna @rotudam]

A couple of things come to mind. I actually wrote about this on LinkedIn last week.

Years ago, I was designing a courier app for an on-demand delivery company. I thought I'd done a great job. Followed the best practices, kept up with the design trends, everything looked pretty good on my screen.

Then I spent a day making deliveries with the couriers.

Turns out they were using cheap Android phones I'd never even heard of, mounted on motorcycles inside thick protective cases. Some had cracked screens. They were trying to use the app with motorcycle gloves, under direct sunlight, sometimes in the rain.

That day, I pretty much threw out my fancy UI decisions. Tiny fonts, subtle borders, all the things that looked great on Dribbble suddenly made no sense.

Another time, I was designing a dashboard for a mobile game publishing startup. Lots of whitespace, clean layouts, nice visual hierarchy. I was pretty happy with it.

Then I talked to 4 or 5 actual users. They wanted almost the opposite. Dense tables, smaller text, more information visible at once. Things I would've considered bad design at the time.

They were spending hours in it every day and wanted to get through their work without constantly scrolling or switching views.

I still care about visual quality, probably just as much as before. But I guess, like with most things in life, I've become less interested in comparing my work to what everyone else is doing and more interested in figuring out what the work actually needs.

[Altay Suna @rotudam]

@caleb_vorynza That was the point of saying it at component level. A page-only fix fought the next rebuild from the specs, so the tweak had to live on the shared component or it came back.

[Altay Suna @rotudam]

I took one client website from brief to working pages in AI.

By the end of this article you won't be vibe designing or vibe coding anymore, you'll be designing with AI the same way you design in Figma or Photoshop. https://t.co/SWK4IFC4N1

[Altay Suna @rotudam]

@EdwardCassas I've been meaning to gather a few of my own failure stories into one post. They make me laugh when I remember them and they weren't funny while I was living them :) Also remembered this https://t.co/CA3QrxWlny

[Altay Suna @rotudam]

Read the success stories you see here (including the "you can do this with AI" ones) knowing that 90% of them are exaggerated for engagement and clicks or simply made up.

A lot of us get the dopamine hit from reading them and stop there. Some want to try but have no idea where to begin.

The best way is still to learn by doing. Find the smallest useful step you can take, take it and build from there.

That's how you find out what actually works.

[Altay Suna @rotudam]

@Ozzy_design Same bet from the other side. I kept components and tokens in shared markdown so Cursor, Claude and Codex all read one place. An MCP that handed the real components would cut a lot of the glue I still do by hand 👌

[Altay Suna @rotudam]

@Vivian1937869 I wrote it into the page specs as a hard rule. Missing client copy stays blank. First builds kept inventing headlines from the brief tone, so placeholders only until they send the real lines 👌