Skip to content

Instantly share code, notes, and snippets.

@alexbaldwin
Created July 19, 2012 17:42
Show Gist options
  • Save alexbaldwin/3145565 to your computer and use it in GitHub Desktop.
Save alexbaldwin/3145565 to your computer and use it in GitHub Desktop.
David Cole Notes
# Principles of product design
## David Cole, Quora
Applying principles is difficult, Dieter Rams and Paul Rand are seen as the standard, but principles are abstract and hard to apply.
History of the web has very much so evolved out of print and managing libraries worth of documents. Thus early web designers came out of this print mindset.
This changed with the introduction of AJAX, around the time of Google Maps and Gmail, interactivity and response time allowed websites to play on the level of native apps.
Now, iPhones they are crazy powerful and can morph into whatever tool is needed at the right time. What was traditionally called inventing, we now call designing.
Facebook is very influenced by graphic design, grids, type, etc, yet you wouldn't call it a graphic design.
Design is much more now about the why. Companies have the ability to be "design led".
If we can't look to graphic design for inspiration, who do we turn to. Ironically people that would have never called themselves designers, but inventors.
Products support business. It's a piece of a larger whole. In context of business they are drivers of income.
> Product: an outcome generator. Product design: improving the outcomes generated.
Business model should support the outcomes. This is what drives the vision. Vision is fixed, everything else is there to support it.
How to drive outcomes. Derivitives of why. What do we do to get there. How do we build it.
* Why?
* What?
* How?
Applies to software as well as businesses.
* Business
* Product
* Implementation
Software product business stack
* Business
* Product
* Interaction
* Visual
* Front-end
* Back-end
Product designer takes up larger bandwidth than traditional "designer". Extending from front-end to product, rather than interation to visual.
Visual / brand plays into interaction which makes them difficult to separate the work, however they typically work in different teams.
Magic happens when they're lines drawn and consistency between the layers.
Small teams are able to work so much faster because workers are forced to take more of the stack. One reason larger organizations are slower is the cost of translation between each piece of the stack. Also, everyone you add to the process is someone who can disagree or misunderstand. The integrated role has been shown to produce results in recent history. Usually this person is the *designer founder*.
Autonomy for designers to do whatever is necessary to achieve a goal will be the crux for design in the coming era.
### Scientest, teach us how to measure.
* Scientific method: creating hypothesis => drawing conclusions
* Form designs as hypotheses.
* By doing this you're setting yourself up for feedback.
* Because this context, with this solution, I propose this outcome will occur.
* Stay true to the implications of those assumptions.
* Most disagreements come up due to execution details. Which are derivitives of facts and assumptions. Argue about those.
* In design feedback, get to the why root, that's worth arguing about.
* You don't need to be confident in your hypothesis, just decide on one to see if it matched and judge from there.
* What should I expect to see if X is true.
* **Never compromise**. Don't do a half solution.
* Oh no-nos: "we can add a setting", "we'll decide later…"
* It's okay to go with someone else's gut, but design the solution congruent with their whole point of view.
* Hedgey is a bad term.
* Make sure there is something to learn and falsify.
* **Embrace being wrong**. The more wrong you are, the better.
* The worst redesigns only change interactions and visuals. More success comes from moving the entire product stack forward.
* Constantly be checking constraints, don't be stuck with something that was an arbitrary choice.
* Look for being wrong about being wrong. Don't assume anything is true, unless you have data in context.
* When stuck, just try to brute force it.
### Engineer, how to execute in a way that's efficient.
* **DRY**, you use it all the time if you write front-end
* Rather than designing everything as one-offs, think about patterns that can be re-used. Great example is @twitter bootstrap.
* Value you're getting from designing in context is lower than the ability to move fast.
* High level changes that can happen quickly will generate the most value.
* As things solidify, go back and focus on context and specifics.
* **Build progressively**.
* Make small pieces and build things on top of eachother.
* Design individual interactions and then zoom out and think how they inter-operate.
* Low-level => high-level
* Look for product wins in small ways.
* Learn what value you are creating at each step.
* Don't design an entire ecosystem.
* **Make it work and then make it look good**.
* Getting into depth of a project should be a choice.
* Good will from having something look good can guide you down the wrong path.
* If it's necessary and useful, don't hestitate to make it beautiful.
* Social products need real people, in real settings, require real use for learning to understand what you have.
* Not competing on prettiness, competing for outcomes.
Agreeing on outcomes, it's okay to have a gut-check "this just feels better" rather than data
Build up patterns over time. Abstract from projects that have been designed for multi-purpose use.
### Investor, not connected to craft but higher level where value comes from. How to work in a value-oriented way.
* **Identify a central metaphor for your work**.
* > Find a central metaphor that's so good that everything aligns to it. - Alan Kay
* Don't try to solve too many problems at once.
* Innovate in one area of the stack.
* Airbnb: hotel booking, but you are booking a spare room. Hotel.com for spare rooms. It designs itself once you've identified the change. Not changing visual or interactive elements. Makes something brand new feel familiar.
* Skeumorphism can be really powerful for inviting new users in.
* Strive to create familairity and increase in sophisitication over time.
* > When new founders pitch, they tend to focus on solutions over problems -DMC
* **Focus on problems, not solutions**.
* Problems are familiar and shared, creates common ground .
* Everyone can agree that somethings a problem, not what the exact solution is.
* Investors are giving resources over the long term.
* Long term, the product is expected to change.
* Focusing on a solution is dangerous because you tend to protect and scale only that solution.
* Be Netflix, not Blockbuster.
* Inverse of a problem isn't a solution, it's when the problem doesn't exist anymore.
* "If you have a hammer, every problem looks like a nail."
* Read all of Ryan Singer's personal blog.
* PM goes broad, designer goes deep. Roles can be one in the same, think about switching context.
* Collaborate on solutions with different departments, sales, support, etc.
* Think outside of design, pursue outcomes.
* Look at human labor, what happens when you do something all the time?
* Reddit: founders kickstarted a whole community they would want to participate in and then over time withdrew fake accounts.
* Lean on manual labor when you can, rather than a design solution.
* Unsexy solution to scaling a community, put real human beings there watching and moderating.
* **Measure your success in years**.
* Pinterest isn't successful… yet.
* Digg vs Reddit traffic, slow and steady won the race there.
* Short term success can blind you from long-term success.
* Most successful companies today look more like Reddit, they take time and grow slowly.
* > Companies won't show their value for 15 - 20 years. Even Paypal hasn't fully matured. - Peter Thiel
* > Most companies work on 2 year increments, Amazon works in 7 year increments. - Jeff Bezos
* Moving fast is the goal of being able to stay competitive on the 7 year scale.
* As a designer, everything you're doing is tossable.
* If you're not going to realize the value you are creating for 5 years, you have to be patient.
* Every one person that uses your service took the time out of their day and chose you over someone else. Treasure it.
* In a startup the path is super long, just remember it's only the beginning.
* Keep at it and eventually it'll be at a scale that no one ever imagined.
* "How we build" - video
[NOTE] Purpose-driven businesses
Getting ideas out of your head
Picking highest value targets
> Can I work on this for 5 years and throw it all away?
> How much value COULD this potentially create?
> With feedback, assume if they are right what is the why and then how to course correct.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment