MEADOWRANK
Play free ▶▶

All niches · Creative & content · UX/UI designers · YouTube

UX/UI designers · YouTube

Someone looked at their signup flow and asked the comments whether it needs a designer.

Nobody goes looking for a UX designer. They watch somebody rearrange a clean example app, stop halfway, and describe their own mess in the comments instead. Meadowrank reads the comment sections, so you don’t have to, scores the ones that are real, and drafts the reply in your voice. You send three. That’s the whole goal for the day.

Free is a full game. $35/mo unlocks your own scans.

Hazel last counted this meadow on .

Reddit SOONYouTube 45/70Facebook groups SOON
The last 70 we looked at

Forty-five had real products behind them. Twenty-five were there to learn the tool.

Every cell is one comment thread, taken from design tutorials, app walkthroughs, and why-users-leave uploads on the newest pass. The lit ones held somebody worth writing back to.

Pain15Signups that stop on day two, a support inbox asking where a button is, and software that works and looks unfinished.
Buying intent5Asking the comments outright who reworks a live app, what a flow redo runs, and whether small projects get taken.
Discovery24Paused halfway through the tutorial, weighing another month of learning the tool against handing the screens over.
Rival1Done with a designer they hired once, usually after a beautiful file nobody could build from.
Visibility0No project in it, but it is the thread a founder reads the week before they start asking.
Not a signal25Plug-in links, which-typeface questions, the design-system argument, and several hundred comments saying clean.

Sample of 70 conversations · classified by the same model that scores the game · from our most recent crawl. Small numbers because we're showing you the real ones.

What a half-built product sounds like in a design tutorial’s comments

Nobody searches for a UX designer. They search for how to make an app look professional, how to lay out a dashboard nobody has to be taught, why people reach the second screen and stop. Then they follow along with somebody arranging a clean example in a tool they half know, pause it, and type their own product in underneath: the signup flow, the number of screens, the launch date that has moved twice. What they’ve typed is a scope of work, handed over for nothing, before a single call has been booked.

On the newest pass, twenty-four of the forty-five lit cells were people partway through deciding to do it themselves: a design file open beside a half-built product, the tutorial paused on a color style, and an honest question about how far somebody who isn’t a designer gets before it starts to show. Fifteen had stopped weighing anything and described the damage instead: people signing up and never returning on day two, a support inbox that keeps asking where a button is, and software that works perfectly and looks like a spreadsheet with ambitions. Five skipped the method entirely and asked for a person: who reworks an app that already ships, what a signup flow costs to redo, and whether anybody takes on a project this small. One had paid a designer once already and got back a beautiful file nobody on the team could build from. The three want different things from a reply. Somebody still learning needs to hear which part of this a tutorial will never reach. Somebody counting lost signups wants a number, and the weeks it takes to move it.

Nothing in a comment says whether there are paying customers on the other side of it. All 70 had to be read to find the ones that had them, and no client covers that hour.

Filling the space around them are plug-in recommendations, people asking which typeface that was, the standing argument about whether a design system matters at this size, and several hundred comments saying clean. At one-to-two, somebody shipping to paying customers and somebody building a first side project type the same two sentences, which is the entire reason all 70 get read.

Where to find UX design clients on YouTube

The uploads divide into four shapes, and only one of them is all designers. Tutorials aimed at people who are not designers are the richest seam, because everybody underneath has a product they’re trying to draw themselves and a growing suspicion about how that’s going. Videos about why an app loses users gather the founders who already know something is wrong and have a figure for it, which makes their comments the most answerable on the platform. Landing page and conversion walkthroughs pull in people who will pay for one screen long before they’ll pay for twenty, and one screen is how most of this work starts. Portfolio reviews come last, since nearly everybody under those is already in the trade. The marketing site out front, rather than the screens behind the login, belongs to the page a stranger sees before they sign up. A checkout leaking customers the way a signup does sits with the last three steps of a store. The logo argument that arrives halfway through every redesign is a separate job, over in the part of this that is about the name.

The comment that wins the project names the screen they are losing people on

Somebody who has typed their signup flow into a comment box has two questions underneath it, and a portfolio grid answers neither. The first is whether design is the problem at all, since half of what looks like a broken interface is a product asking for a decision too early. The second is what it takes to put right, in weeks and in money, because everything quoted to them so far came from an agency pricing a quarter. Both fit in a reply. Name the screen you would change first and what it’s costing them in the figures they already gave you, say how long that one change takes on its own, and say plainly which of their problems no redesign is ever going to solve. Giving the diagnosis away in public feels like running the kickoff meeting for nothing. It is also the only reply in that thread written by somebody who has shipped a product rather than a mockup.

What finding clients on YouTube actually costs you

Call it three comments, and the morning is still intact, which is less time than one call that ends in maybe next quarter. It slips anyway, because every other hour of this work is attached to a file with a date on it and this one is attached to nobody. A finished flow ends in something a developer can build. A handoff ends in a link somebody opens. Four weeks of good comments end in nothing anybody reviews, so it drops behind the sprint that started Monday. Meadowrank makes it countable. A comment with a product behind it is held under a short match-3 board. The cover clears, the card opens, and there is the upload it was left beneath, the numbers they typed in, and a reply written the way you talk in a first call about somebody else’s product. On a day given over to revisions, the feed lists the same signals for a little less XP, so an afternoon with no gap in it still counts. Points only ever come from replies you actually sent, so the total at the end of a month counts founders you wrote to, not videos left open in a background tab.

Do your three. That’s all the marketing today needs.

Playing costs nothing, and free here means the finished game rather than a locked-down slice of one: the board, the feed under it, the streak, and five cards dealt each morning out of the pool everybody works from. $35 a month, or $350 across a year, swaps that shared pool for a Quarry nobody else digs in, and the scan then follows the kinds of product you’re quickest on, the stage of company that can decide inside a week, and the uploads their founders actually watch.

Three steps

How to find UX design clients on YouTube, in three steps

  1. 1

    Start with the design tutorials made for non-designers, then the why-users-leave uploads

    Take one channel’s newest tutorial aimed at people outside the trade and one upload about why an app loses users, sort both comment sections by recent, and read for ten minutes before the first block of real work. Skip the portfolio reviews, where the audience is other designers. Of the 70 comment threads gathered under design uploads on the newest pass, forty-five came from people building a product that has outgrown what they can draw themselves.

  2. 2

    Seed the thread with the screen to fix first, what it is worth, and what design will not solve

    Put their own figures back in front of them, the signups and the drop-off, so it is obvious somebody read the comment rather than skimmed it. Name the single screen you would change first and what that change is worth against the number they gave. Then say how many weeks it takes, and say which part of the trouble is the product rather than the pixels, because that is the sentence nobody else in the thread will write. Three sentences, left under the video they paused to ask the question.

  3. 3

    Capture the demand by keeping count

    A ticket gets written, sent, and closed out, and these are no different: three comments before the first block of real work. Points only ever come from replies you actually sent, so an evening spent reading comment sections moves the number not at all. The watcher further down takes an address and reports on Mondays, a count of the week’s design asks and the three that deserved an answer first. A week when three founders at once decide the screens are the problem shows up there before it shows up anywhere else.

Try the board

One of them is under the cover. Clear it.

This is the game, at its smallest. A real UX/UI designers signal from YouTube sits under the cover, the same kind the honeycomb counts. Match three on the cover to strip a layer. The Monday email sends three of these; the board shows you one.

Short on time, or not in the mood to play today? The game also has a feed. It lists the same signals with the same drafted replies, for a little less XP. The board is the fun way in, not the only one.

One real UX/UI designers signal from YouTube is under the cover. Match three on it to strip a layer.

75
intent
Free · no signup

Paste a YouTube comment, get a score

The same scorer runs on a page of its own, the YouTube Comment Signal Score Checker. Feed it the comment you keep rereading, the one where a business and a weekend project look identical.

Paste a post, get a score

No signup. The same five-type classifier that scores every signal in Meadowrank.

Or just watch

Email me the weekly movement for UX/UI designers

One email every Monday with this niche’s real counts across all three platforms.

One email a week, Monday. Real counts, no pitch. Stop any time.

UX/UI designers on YouTube: what gets asked first

Is replying to comments on someone else’s video allowed?

It is, and a creator teaching people to lay out an app would far rather have somebody answering properly underneath than another row of applause. Meadowrank writes replies that take the viewer’s own screens seriously and say what to change first, leaving your rates out of it until somebody asks. They come away knowing what is wrong with the flow, and the visit to your profile follows from that.

How many signals should I expect per day?

Two or three on an ordinary day, more in the week after any big upload about a redesign. Out of 70 comment threads the newest pass turned up 45 that deserved a reply. Three readings a day is the whole habit, and thirty days of that leaves dozens of founders having watched you name the screen costing them customers.

Is this scraping YouTube?

No. The searches are the ones somebody with an ugly app would run, and what gets read afterward is whatever public comment threads come back under them. Nothing needs a login and nothing is kept that the page would not have put in front of you. Hazel only saves you the scrolling.

What does it cost?

It costs nothing, and the paid tier holds none of the game back: five cards are dealt out of the pool everybody shares before the day starts. $35 a month, or $350 for the year in one go, opens a private Quarry, which aims the scan at the products and the company sizes you want, and lifts the daily ceiling on cards.

How do UX designers find clients on YouTube?

By fixing one screen in public. The founder who hires you has already described a signup flow, a drop-off, or an app that looks unfinished under somebody else’s tutorial, and the project goes to whoever named the screen to change first and what it was worth. Nothing in the sample suggests a comment advertising design work ever turned into a project. Three readings a day, under the uploads founders watch rather than the ones designers watch.

Which YouTube videos have founders looking for a designer?

Design tutorials made for people outside the trade, first, because everybody under one is trying to draw a product they cannot quite draw. Why-users-leave and conversion uploads come next and convert fastest, since anybody who typed a drop-off figure underneath has already measured what leaving it alone costs. Landing page walkthroughs bring people who will buy one screen before they buy twenty. Portfolio reviews come last, where the audience is the trade itself.

Is YouTube worth the time for finding UX design clients?

Ten minutes, yes, particularly for founders who already believe the screens are the problem. An evening in the comments, no, and leaving that half out is why most designers stop doing it. Across 70 comment threads on the newest pass, forty-five came from people with a product that has outgrown what they can draw. That is three replies of real work in a day, spread under uploads nobody has time to watch. The ten minutes earns its place only when the sorting has happened first. Like Hazel already does inside Meadowrank.

How this page is made
  • Counted on September 21, 2026, from 70 YouTube conversations found by a fixed query set for UX/UI designers, scored by the classifier that runs the game. Nothing here is typed by hand: every number is computed from that sample.
  • Written for this niche and platform by Meadowrank's own model, from the rows behind this page. Communities named on it appear in its own sample.
  • Signed off by the founder, who stands behind what it says and pulls a page that gets it wrong.
  • Never stored. The posts themselves, or links to them. We classify, then discard. The full method.