HNHacker News
TopNewBestAskShowJobs

JoelMarsh

528 karma · joined September 9, 2012

Born in Canada, lives in Sweden.

Co-founder of Teller Email Browser — www.tellerapp.com

Blog: www.thehipperelement.com

Author of "The Composite Persuasion".

submissionscomments
JoelMarsh··on A/B Testing
From the article: "A few months ago we tested a new design for our map UI. The new UI converted a bit worse. But we preferred it, so we kept it anyway."

They A/B tested and then went with the LOSING version, because it felt better in their gut.

Go for something radically different if you want, no problems there, but don't test if you're going to choose your favorite even when it loses. That's just dumb.

JoelMarsh··on A/B Testing
A/B testing does not tell you what to do. It measures what you have done. HUGE difference, but a difference that very few people actually seem to understand.

The last paragraph of this article made me shake my head though. If you do A/B tests and then use the losing version anyway - because it feels better in your gut - you're an idiot.

Choose a different path if you want, but if done properly, your A/B results are fact (among the options you tested).

JoelMarsh··on Website Impounded
Typically - and I am purposely avoiding too fine a point on this - the copyright remains with the creator until payment occurs. Otherwise no transfer of rights has occurred.
JoelMarsh··on Show HN: Fundhawk - VC Analytics
This is great! Therefore, I have requests! :)

1) Sort the co-investor list by clicking the title of each column. Specifically, I would like to see the list in % order.

2) It would be fantastic to filter search results by location, or VCs that match certain criteria.

3) It would also be amazing to get the full list of VCs, ordered by the categories in the results. For example, show me the whole list, in order of who has done the most seed rounds.

4) And finally... a bigger request... if you could include exit deals, that would allow you to calculate their success rate and ROI, which would be REALLY interesting. But I realize that's harder than my other three thoughts. :)

Well done!

JoelMarsh··on Ben Horowitz: We Like Dropouts, Insane Ideas, Tiny Markets, No Way To Monetize
Hate being a buzzkill, but that is precisely what couchsurfers was before AirBnB.
JoelMarsh··on Wireframing Tools Suck
I love that competition idea.

For non-drawing features I definitely agree. Linking would save lots of work.

JoelMarsh··on Wireframing Tools Suck
As a 10-year UXer, I totally agree! I was nodding and saying "exactly" out loud throughout the whole article.

People that WANT shadows and details and pre-made stuff like that are the people who want to design the UI in stead of the functional specification. The more experienced you get, the less you want them.

Here is an exhaustive list of every feature I need to make wireframes that you would call beautiful:

- shapes: rectangle, rounded rectangle, circle, triangle, line, arrow

- rectangle with an X in it, representing images

- text in a couple different sizes and two weights (bold & regular)

- a few colors (maybe 5). Shades of gray, red, green, & blue.

That's it. Anything else and you're slowing me down. Seriously.

JoelMarsh··on Do you think the desktop is dead?
"Mobile first" is a design strategy, which ensures that you identify the primary content/features when making responsive designs.

It has nothing to do with which device is more important or more common.

JoelMarsh··on Ask HN: Feedback about HN itself. What do you think?
If an upvote is a save, that's a good idea (it's a good feedback loop) but that definitely should be communicated somewhere when you register. If nobody knows that it sort of kills the thought behind it.
JoelMarsh··on Desktop App for All Your Email
What has the response been like over the past day or two in terms of downloads and traffic, etc.?
JoelMarsh··on Ask HN: How should I spend a year?
Humbly, I would like to offer a suggestion that is more about the way you're doing this, rather than what you're doing.

You sound like you're thinking only short term, and only in terms of keeping yourself busy. That's not how motivation works, unfortunately.

With that kind of approach you never see how your tasks fit into the bigger picture, so everything that comes along is more attractive because it's new. Discipline demands a plan.

Choose a goal that will take you 5 years to complete. Something concrete. Something like "build my own app" or "be an early employee at a hot start-up" or something else that takes multiple steps, and something you feel extremely excited about achieving.

Then outline those steps and get into the first one now. Then when a new opportunity comes along you can evaluate it by saying "is that a better 5 year goal?" If it isn't, "does that get me closer to my 5 year goal?" If not, ignore it and keep on keepin' on.

Voila: discipline through motivated planning.

JoelMarsh··on What's Happening, Facebook? You're turning me into an asshole, that's what
I love that Facebook is blamed for making one's internal dialogue external. That's not Facebook MAKING you an asshole. That's Facebook REVEALING you're an asshole.

I think Facebook is guilty of a lot of things, but this isn't one of them.

JoelMarsh··on Ask HN: Do QR Codes work?
If you don't remember it later, you probably weren't motivated enough to do the QR code in the first place.
JoelMarsh··on Ask HN: Do QR Codes work?
QR codes are probably the greatest example of monkey-see-monkey-do ad spending in the last decade.

I have had large, international clients (you would recognize the brands) ask us to "invent something that uses QR codes" because their competitors use them.

I have seen the results of a campaign that was (and is!) huge in Sweden, involving many regional celebrities, and the total number of QR scans was in the double digits.

QR codes are a link. Plain and simple. Except they're not plain and simple, they're weird and complicated. They take you from one piece of content, which happens to be in real life, to another piece of content, which happens to be online.

And you need to include your URL and instruction text anyway, so it's not really providing a big advantage.

Some advertisers even use them ONLINE! That just baffles me completely.

If all smartphone cameras included QR readers by default, there MIGHT be a chance for them, after a lot of public education. But as it stands now, the incentive has to be huge to get over that initial barrier.

If QR codes really want to succeed they will have to become better than the alternatives: Google + memory, URL + keyboard, etc.

JoelMarsh··on The Appification Of Everything Will Turn The Web Into An App-o-verse
Is it doomsday again for the web already?

http://www.wired.com/magazine/2010/08/ff_webrip/

JoelMarsh··on Meet the World’s Cheapest Venture Capitalist
I feel like this is a joke that I almost get, but don't.
JoelMarsh··on Why founders shouldn't be the developers
What I get from this article and from the Git-commit list (upvoted!) is that founders shouldn't REMAIN developers. Having a coder-founder seems almost fundamental to a tech start-up, no?
JoelMarsh··on Ask HN: What would make you excited about email?
Thanks for the feedback! Surveys have their limitations, of course, and after years of experience I have found that (depending on the goals) it is sometimes better to allow the respondent to "interpret" a little for themselves rather than restricting them too much. Thanks again (and I have noted your encryption comments too).
JoelMarsh··on Ask HN: What would make you excited about email?
Thanks! If that's what floats your boat then I definitely suggest you sign up at getremark.com. ;)
JoelMarsh··on Ask HN: What would make you excited about email?
Thanks! Can you elaborate a little more on the signing suggestion?
JoelMarsh··on Ask HN: What would make you excited about email?
Thanks! (and noted!)
JoelMarsh··on Ask HN: What would make you excited about email?
I agree. We have been watching them closely. We have an entirely different approach than them though. If you're interested, there is a little sign-up form for updates at www.getremark.com. We'll be releasing more info in a few weeks.
JoelMarsh··on Ask HN: What would make you excited about email?
Easier link: http://bit.ly/T088MN
JoelMarsh··on Show HN: A community for designers to hone their skills, critique and share work
Anything else I noticed seems insignificant compared to those ones.

One note though, in response to a comment above... be careful not to "force" people to be honest; they might leave instead. Motivate them to be honest by rewarding honesty and making it clear how they are rewarded. Feedback loops are your friend. ;)

JoelMarsh··on Show HN: A community for designers to hone their skills, critique and share work
Fantastic.

For #1... maybe the real feedback is that I didn't understand that. But fair enough. I don't disagree with your goal. :)

For #2... you're right, and my point was that there is nothing for those lurkers (or "pre-contributors") to do. The community is very singular in its focus, and that focus is on the minority. A "browse" page or a footer full of browsable suggestions would be worth a lot.

JoelMarsh··on Show HN: A community for designers to hone their skills, critique and share work
Done. ;)
JoelMarsh··on Show HN: A community for designers to hone their skills, critique and share work
I am a UX guy who specializes in community design, and I see some fundamental areas you should work on:

1.) How is this better than Dribbble; a popular, well-respected site with precisely the same pitch? That's not a criticism so much as something to think about.

2.) You forgot half of your target audience. Community like this has two user types that support each other: designers and critiquers. Ideally the critiquers are designers too, but these are two separate behaviours, and your site is designed for one of them. You're never going to reach critical mass until you cater more to the majority (non-uploading, non-commenting lurkers).

3.) It's hard to browse. The start page has a few clickable images and once you choose one, you're doomed if it wasn't something you like. I bet the exit rate (Google Analytics) on that second page is through the roof... what should I do next if I didn't like it? Hit the back button? That's a no-no.

4.) The users are too downplayed. In a community like this, people discover the users through their content. Then they want to follow and see more, which motivates registrations. You should make it obvious who has posted each design, with a clear link to see more.

5.) You did a good job of selling the concept on the start page, and the path to registration is obvious and clear, but I don't want to sign up for something as time-consuming as a community without trying it and exploring a bit first. That's the part that isn't so great right now.

6.) I don't get the metaphor. Is it military or something?

7.) On a purely aesthetic level, you might consider simplifying your brand... there are a lot of colors, and mid-tones, and textures, and geometry, and it's forcing you to use lots of boxes and stuff. It's hard on the eyes. Pick a couple colors, a motif/pattern or two, and reduce, re-use, and recycle. ;)

Good Luck! Nice effort so far!

JoelMarsh··on Submittable (YC S12) raises $750K to replace email
I think we should pause before having a serious discussion about an interface replacing a protocol.

This is like Google's pitch being "Our search will dominate because we use a single text field, and we have a 'surprise me' button."

If you want to "replace email" you have to replace its infrastructural value to the internet too, not just the superficial day-to-day convenience of messaging. As long as every website requires email to register, messaging is a secondary consideration.

I think that 750k could have been used for a failure-less-obvious.

EDIT: for tone.

JoelMarsh··on Show HN : My weekend(s) project - Strapfolio
Also, good luck and good work! :)
JoelMarsh··on Show HN : My weekend(s) project - Strapfolio
Senior UX guy here... free advice:

1) Your demo button is killing your sales.

The drop between step 2 and 3 of your stats is the proof, although I immediately noticed the problem: People don't go "back" to make a sale, they go forward. When they click your demo button, they no longer have a call to action button to lead them to the sale. Recommendation: add a top bar that stays with them as they check out the demo. Regardless of how you solve this, it's worth losing sleep over it.

2) Changing the demo content (text) to be about the theme itself.

At least in the intro paragraph. I found it very disorienting after clicking the demo button.

3) Pick 3 selling points, not 9.

The first one should almost convince me by itself. Make them bigger and more visual. The rest are just extras. Make them look like extras.

4) Remember that you're selling design to designers.

There are beautiful Wordpress themes that are free. Yours looks ok, but not $30 ok. ;) TELL ME WHY I LOVE THIS, and tell me in the first sentence.

5) Screenshots suck. Use the image to show me or teach me something I need to understand about this product. Better yet, get a really nice background image and fill that baby blue area with it. Add some emotion, man!

6) By pitching it based on Underscores and Bootstrap you're immediately selling to a technical audience. In my experience, visual creatives are not particularly technical. Why is bootstrap good? What is Underscores? etc.

7) A/B test versions with and without the price in the buy button. You include the price when it makes the product more appealing. Yours is high. the A/B test will give you a very rough idea of how much effect that has on conversion.

8) Include the demo / buy links in the header of your documentation. Beside the logo, not in the top-right corner.

← PreviousPage 2 of 3Next →