Joel Marsh

UX for business with Joel Marsh - UX Podcast Joel Marsh joins us to talk about how to design valuable digital companies and his book UX for business

Inspired by feedback from his booked UX for beginners, @joelmarsh.bsky.social has written UX for business. A comprehensive guide for designers navigating real-world challenges - covering over 300 different topics!

uxpodcast.com/336-ux-for-b...

#ux #uxpodcast #uxdesign

Typography, punctuation, phrasing, spacing, grouping, layout, alignment, repetition, and so on…
.
.
.
.
.
.
.
.
.
.
.
.
…creates timing.

If you ask ChatGPT to summarize my book, UX for Business, you will not learn anything from the book.

But you will learn which topics you would learn if you actually read the book.

Not the same.

You don’t become a master of anything by automating the work.

Had a fantastic chat today with someone in the later stages of a successful career.

We were talking about where you start versus where you go, in life.

She said:

“Things come along. You say yes or you say no. That’s it.”

Simple yet impossible:

- No decisions should be made with the goal of making stakeholders happy. (99% of companies fail here)

- All decisions should be the ones that are best supported by customer research.

- If you measure your decisions, and they work, all stakeholders should be happy.

God yes. This happens so often.

“Good design” doesn’t mean the design matches current trends.

Read that again, it’s important.

If you can't recognize a good solution until you see it, you can't recognize a good solution.

#product #design #management

Some of the best graphic design right now might be type specimens from smaller foundries.

This is Swiza by @atipostudio.bsky.social on @behance.bsky.social

It's amazing how often new tech comes out, and I am too busy to try it for a week, and by the end of that week the hype has passed.

It's like when you call the doctor and they say "if you're still sick in two weeks, book an appointment." Most people don't book an appointment.

Tactical patience.

“The smallest thing we can build in Q1” is not an MVP.

An MVP might still take a long time because it is the smallest thing that does the job, or the smallest thing the customer will enjoy.

Do not confuse “fastest” and “smallest”. Different trade-offs.

An MVP is a threshold, not a deadline.

There are two kinds of people in the world right now:

1) People who think AI is all-knowing and will be able to do everything better than people, very soon.

2) Everybody trying to build something with LLMs or trying to figure out what the hell their client actually wants.

A guy on LinkedIn said he had been working with UX for a long time and had never seen user research add value.

He also did not see the irony in being the one who didn’t add value.

These are the people killing UX’s credibility for the rest of us.

He now runs a venture studio. 🤦‍♂️

Some people think I am trying to be motivating when I say you can’t “lose” an A/B test, because you *learn* something new.

I’m not.

A/B’s are not win/lose things. They are scientific. You’re validating a solution. Losing is *interesting* and *valuable* information.

"Halo Effect": A known cognitive bias that makes people believe that a person who is good at one thing, is also good at other unrelated things.

"Competitor Halo Effect": Thinking that everything a competitor does is worth copy-pasting into your product, because they have a popular feature.

Developer: "Someday I am going to give someone a great answer to their question and they are going to love me and the answer. Today is not that day."

That's so real it brings a tear to my eye.

I.e. — “processing X data” is slow and the KR describes the measurable value that would mean we have eliminated that weakness.

I also prefer “pairs” of metrics or conditions that must be met when achieving the quant part of the KR, like “no quality loss”. Prevents bullshit tactics.

This has worked for me in practice, even if it might be OKR heresy:

O’s = core advantages/differentiators/strategy of our product. They rarely change. Something like “speed of delivery”.

KRs are defined as very concrete weaknesses in the O, like “process X data under 10s w/ no loss of quality”.

I was reading about strategies for running a 10k race, and then switched back to real work, which was documentation animated graphics files that were less than 10 kilobytes and would "run anywhere". All the same words and a totally different context.

I had to force quit my brain for a second.

Just saw an ad for an app “for people with ADHD, by people with ADHD.”

I’m skeptical.

If they really had ADHD it would never have launched.

Being a startup founder is all-consuming. A friend of mine just had a kid and I heard myself refer to that as "growing his family".

It's a problem.

Kids aren't scalable at all.

As a routine, I tend to subscribe to lots of newsletter and “test follow” lots of people—and then aggressively prune over time.

I have never done that all at once though.

Let the firehose begin!

Some designers think this WW2 chocolate bar is “ugly” or “badly designed”.

But designers still had to design the font, the layout, the box, consistently with other military items. It’s a brand!

Clear, functional, usable in the dirt. Perfect!

Design ≠ Style

upload.wikimedia.org/wikipedia/co...

At Peekerton we have designed the entire company to support a data strategy.

Product-wise it means there is a single process to optimize, for both UX and code; both internal and external experiences.

It’s very fun to have deep alignment with colleagues.

Hi Matt! Would also love to be added.

Hey Paweł! You can add me if you want. 😉

Hi Yana, would love to be added. 🤩

A sure sign that decisions are not strategic is that the mood of the stakeholders is a big factor in what is approved or not.

Ping. 😎