What is the ideal relationship between design and engineering in a startup?
Joyce Park stashed this in Best PandaWhale Posts
As with most startup things, it depends.
I like this interview with Garry Tan of Posterous:
For startups on the web, design is often a fundamental differentiator. It’s important to have a design leader in the company as early as possible — and he should be given say over schedules, deadlines and product strategy.
Usually, the issue is that the founding team doesn’t have that capability and can’t hire founder-level design talent. But you do your best — freelancers and contractors can get you pretty far. A trusted design firm or good design freelancer can create the brand and set the visual language for a site with a contract at the beginning, and your team can basically use and remix those elements to great effect moving forward.
I do believe that great design doesn't just happen. It's a process.
It's certainly a complex issue, and the answer will vary from team to team and product to product. I think that a good general rule is to focus on building a core team of people that strive to be as broad as possible, from design to ops. You can bring in specialist designers, DBAs, ops people etc as necessary to round out the team, but team members that have real breadth are like a glue that holds the team and the product together.
Complex because of how people define "design". Some think its visual, making things pretty. Others associate with getting to holistic product fit. Here's where i think it becomes an issue:
What's wrong with this statement? "We need to find a rockstar designer to create a great user experience for our product!" [tbc...]
Answer: It misses the fundamental point the in software, your product IS the user experience. It's not something you add on later.
You launched a product & now you want to "give" it a user experience. It already has one. Every decision you made to that point created it.
If a company doesn't get that every product decision they make impacts user experience, no designer is going to be successful there.
Interesting to get the designer point of view guys! As LukeW points out, there are so many fewer designers than engineers, I think by default the user experience is going to be created by engineers at most tiny startups... how do we own that in a productive way without pissing off designers? :)
As with many questions "It depends" is the one answer everybody can agree on.
Partly it depends on what type of design you are talking about. For branding, visual design, logos etc I don't think most startups have enough work to keep a designer busy long term and therefore to attract a good one who specializes in those areas.
There is definitely plenty of work that requires somebody with some aesthetic sense who keeps the UI consistent, cares about presentation, writes some CSS and turns visual presentation into a working front end. Whether that person is a good enough visual designer to do that part themselves or whether they take material from a contractor and make it work I don't think matters.
It doesn't matter if the user experience is created by designers or engineers. Those are titles, not talents. Some engineers have great instincts for thinking like their users and addressing their needs with an elegant design. One of the best UX designers I have ever worked with is an ex-engineer from the early days at Apple. Now he's a CTO. He can't make something look pretty, but boy does he know how to get to the heart of a UI.
The trick, IMO is not so much about having a designer on your team from the start, but in having a visionary user-focused leader from the start. If a team doesn't have that talent when they build their first release, you are sure to find a pig that no amount of lipstick can beautify.
Design is important, but I think visual design is overrated.
For PandaWhale (at least my understanding of it) getting users and ironing out the friction points through actual usage and conversation seems more important at this point.
On that note, my expectation of where text will end up after I press the "comment" button and where it actually ends up don't quite match right now. (I expect the text ends up on the page right where I'm typing after I hit comment.)
And it happens that way *sometimes*
Also a lack of design is charming! Early eBay, Craigslist, and Google had more character.
Mad Dog! Regarding the comment location... yes we had a lot of discussion about how it works. And the topic has been opened again, since enough people have mentioned it as a confusion point. I'd love to get your feedback on a situation where UX design SHOULD be able to solve a concrete problem.
* We have two kinds of comments: the top-level comment, which is the form at the top of the page that confuses people when it posts to the bottom of the page; and the second-level comment, which has confused no one. However, they use the exact same visual design, which might not be ideal.
* My overarching design strategy for PandaWhale is to make everything as conversational as possible. Basically any question, I choose in favor of "which looks and acts more like a conversation?" This had two big implications here: 1) to treat the original topic as merely the first subthread in the conversation rather that something more like a blog post; and 2) to respect the order of replies within the context of which comment you thought you were replying to.
* Also I tried to design for Pandas aka readers rather than just Whales aka writers. Therefore I chose not to collapse anything, lay out parts of the conversation in reverse chronological, or have any forms that you can't see until you click/rollover. I want the lurker to be able to follow the entire conversation in context if they want.
Now all of this thinking could be logically consistent and still the page layout can be hard to understand. :) But you know... after a single moment of surprise, you certainly figured it out.
I need a whiteboard!
Yeah it may not be a problem at all, but here are my thoughts. Apologies if I'm using the wrong terms here.
OK, so there's 3 levels presented as 2 ("treat the original topic as merely the first subthread in the conversation") so posting a "top level comment" looks like posting a "comment reply" but has a different effect - starting a "subthread" rather than posting a comment, that shows up at the bottom of the page.
Some things to try to clarify what's going on (they aren't exclusive)
* Move the start subthread action out of the box at the top of the page to the bottom.
* Visually distinguish between the topic box at the top and subthreads below it
* Distinguish the call to action on the buttons -- maybe "start reply thread" vs "comment." Maybe something much better than that.
I'm a fan of the first option (though I think I see why you didn't do it that way) - in large part because I think having a comment box at the bottom of the page fosters way better discussions than at the top. Extreme cases being YouTube vs. Metafilter.
Huh, interesting! Thanks for your input, I always appreciate it. :)
It's a bit of a design rathole given the particular thread, but I think there are 2 issues here w/r/t the top level comment & the thread comment -- when I looked at my first thread, I actually couldn't figure out how to reply to the overall thread, so just made my best guess (which was right), but it caused some anxiety.
Anyway, the first problem is that the two actions look the same, but I think the actual problem is that when you've got a comment box & button that's contained by a thread, it looks like action on that will be related to the stuff in the box, not put it in a different box on the page.
Lots of different ways to address the issue, but I had the same anxiety as well, and was a little surprised by what happened.
Thank you for the feedback, John!
We'd prefer not to cause you or anyone else anxiety over participating in the convos.
This is one of the big items we know we need to refine, so we'll be iterating on it soon.
I'm not in the startup world so I am woefully underqualified to comment on specifics. Here's one thought though, backed up by some solid science:
Promoting happiness between designers and engineering is not a "nice to have", it would result in better productivity:
I'd like to think that, in general, promoting happiness between the different members of a team is conducive to higher productivity levels.
I wonder if a great mission also correlates to higher productivity levels.
If you have a user interface, you should have someone on the team who can make it a good one. Pandawhale has a lot lot lot of interface issues, none big enough to kill the power of the community, but enough to annoy people trying it. A frictionless intuitive interface can grow your numbers as well as any other hack. You are doing heavy lifting in design right now. You should have a member of a team who is good at it. Maybe it's a designers, or an interaction designer or a UI-skilled PM, but you will not regret it.Just because their title says design doesn't means they are good at it. Have them walk you through what they have designed, see if it's usable and pleasurable, and makes sense to you. Don't hire a designer just because you need design. Find someone who will make you greater. http://www.eleganthack.com/how-to-hire-a-designer/
I love that mantra: Find someone who will make you greater.