Design is the Business

Why Good Design Matters, Where to Focus, and What Makes a Great Process for Startups

This past October at Celebrate 2015 I delivered a talk to the startup entrepreneurs in attendance (non-designers) about how they should be thinking about using design within their companies. In summary, design is the business. I hope you will pick up one or more ways of thinking about your business that maybe you haven’t before, or arguments to advance design thinking within your company.

The video from my talk at the Celebrate conference

I divided my points into three areas:

  1. Why Good Design Matters
  2. Where to Focus
  3. What Makes a Great Process

Why Good Design Matters

In short: money. You’ll make more of it.

The typical Apple store does 17x more revenue per square foot than the average retail store.

My first example: Apple retail stores. They generate 17x more revenue than the average retail store. $300 per square foot is considered pretty good, Apple makes over $6,000. Every part of the retail experience is thoughtfully considered, including even filling the gaps with intent. When you buy an Apple product, the best way to get help is to bring it back in the store and work with a Genius. While you wait? Have a look around at more Apple products.

“It’s sad and frustrating that we are surrounded by products that seem to testify to a complete lack of care. That’s an interesting thing about an object. One object speaks volumes about the company that produced it and its values and priorities.” –Jony Ive

As for the products themselves, volumes have been written, but it’s safe to say that the relentless attention to detail of Steve Jobs and now Jony Ive has added up. In 1997, Michael Dell said he’d “shut [Apple] down and give the money back to the shareholders.” Almost 20 years later, Dell had to take a $2B loan from Microsoft and delist itself, while Apple has over $200B of cash on hand.

Another one of my favorite examples is the customer expectation setting of Zappos. Shoe retail is a commodity business, which means you can race to the bottom and simply sell cheaper than anyone, which is what happened with and all the similar victims of the crash, or you can create a better experience for customers so they come to you first and keep coming back. Nordstrom has mastered this at retail and Zappos has done it online: considering the experience of returns, shipping, and customer communications all thoughfully and with intention. When you order a pair of shoes, even if you select the lowest priced shipping option, you’re often randomly upgraded to 2-day, which just makes you feel great. Consider that versus just having free 2-day shipping for everything. Exceeding expectations is much more memorable., circa 2009

Of course I have to bring up Mint as well. Yodlee, the backend infrastructure that powered all of Mint’s bank connectivity, actually had their own consumer-facing product. It’s just that it looked like this:

Yodlee Moneycenter, circa 2009

Yodlee wasn’t particularly intentional about creating a customer-facing experience, they focused on infrastructure as a company, and it was reflected in their product. That gave Mint the opening to build something much simpler for consumers, focusing on that instead of infrastructure. Starting with college students (a perfect combination of early adopters of technology and simple financial lives) both in our messaging and product offering, we built a product with strong word of mouth, helped by PR, SEO, and our own content. It sold just two years after launch for $170M.

In summary:

“Over the last year, we’ve started explaining design as ‘the rendering of intent.’ The designer imagines an outcome and puts forth activities to make that outcome real.” –Jared Spool

What is your intended outcome? You have a product vision—a culmination of your experiences with people, and observations of the market conditions. Now you need a process to get yourself there.

Your product is the intersection of these two.
“…design research activities are the core of what we use to make sure our team has a unified set of intentions. We discover what’s happening with our users that we didn’t intend. We look for places where we could do something that would improve the users’ experience. We look at what the competition is doing and ask what we could do that’s even better. All this research helps us hone in on what our intentions should be. The best teams engage in research like this frequently and often discuss intentions explicitly. They explore different possibilities and arrive at something that everyone agrees matches their overall intention.
We need to look at our design process as a way to come to a single intention as much as it is to make that intention real in the world.”
Jared Spool

Where to Focus

You get it already, design produces a lot of value, but there’s 66,000 people at Apple and only a handful of people at your startup. In my talk I brought up the market maturity framework as a way to help you decide where you should focus your efforts, and a few examples of how companies have navigated this framework over their histories.

In the market maturity framework, a consumer product market moves through the following stages:

  1. Technology is when being possible is enough. Your product is 10x better than the alternative and you’re the first to market / first in your category.
  2. Features is where you see the beginning of a competitive marketplace. There’s now a lot of products like yours, and adding features is the way to differentiate and position your product.
  3. Experience is often the reaction to the feature wars. Use cases are now being better defined and features removed to focus on more simply meeting user needs well.
  4. Integration occurs when a product is integrated into another. For example, the MP3 or digital audio player:
The MP3 player is an example of the progression from just being possible to integration into other devices.

Google Ventures applied this market maturity framework well to startups and software products:

  • If you find yourself in the technology or feature market, product design is where you make your stand: solve the problem, satisfy a desire.
  • If you’re in experience territory, interaction design (ease of use and understanding) and visual design (beautiful brand and product) are needed differentiators.

An argument I hear against “good” design is typically Craigslist. It was designed in 1995 and the competition was a newspaper. Both the buying and selling experiences were a 10x improvement. Searching in a newspaper is impossible for a buyer. Selling and listing an item was cumbersome and expensive. However, you can’t launch a web application in 2015 and point to it as a blanket example of why you don’t need “design” unless whatever you’re doing is a 10x improvement over the alternative. Craigslist is still a huge site in terms of traffic, as it’s famous after being in existence for 20 years, but an arguably stubborn refusal to improve the experience has resulted in huge opportunities for other companies, while it remains a lifestyle business with about 30 employees. It has not adapted to the market.

Companies that are picking off marketplaces on Craigslist, as of 2010

Stubhub sold to eBay for $310 million, which represents just a small fraction of the marketplaces on Craigslist. The ticket buying and selling experience is unquestionably superior. Airbnb last raised money at a $20B valuation. The key drivers have been superior user experiences and reputation—critical for a peer-to-peer marketplace.

Uber’s UI, circa 2010. Visual doesn’t matter when you’re competing with the pain of hailing a taxi cab in San Francisco.

Uber is a modern-day Craigslist in a way, its initial offering was not particularly pretty, but its competition was the experience of flagging a taxi in San Francisco, or calling a dispatcher. If there’s a magic button I can press to make a black car show up, it doesn’t matter what it looks like, that’s a 10x improvement in experience. Unlike Craigslist, Uber didn’t stay frozen in 2010. It’s continued to improve in the face of competitors, with additional offerings, a better designed product experience, and an aspirational brand.

Square faced a different set of challenges, and does to this day. Their initial offering was arguably a 10x improvement in technology and experience for a small or single person business looking to accept credit cards. From the application process to the transaction, Square’s offering was a massive improvement to a painful legacy process. However, the difference here from Uber was the amount of fast-moving competition in the payments space. Square needed to raise a lot of capital, and deploy design resources to the product design, as well as invest heavily in interaction and visual design of their software and brand to stand out in an incredibly crowded field. Square was truly Apple-inspired in their attention to detail across all customer experiences. However Paypal, Verifone, Intuit, and the others weren’t just going to throw their hands up and get disrupted out of business. They copied Square’s reader, and worked to copy their software and processes, so this remains an interesting market to watch. I’d still bet on innovation over imitation, as long as Square continues to invest in design. They can’t stop. (See Yahoo! for what happens when you do).

What Makes A Great Process

For me, the first step in a great design process is to have a great story. It’s the first step in a great business too.

The story must explain at a fundamental level why you exist. Why does the world need your company? Why do we need to be doing what we’re doing and why is it important?
You can have a great product, but a compelling story puts the company into motion. If you don’t have a great story it’s hard to get people motivated to join you, to work on the product, and to get people to invest in the product.
The mistake people make is thinking the story is just about marketing. No, the story is the strategy. If you make your story better you make the strategy better.
–Ben Horowitz

How do you have a great story? I like to refer to the five W’s of journalism: who, what, when, where, and most importantly, why.

The who in this case is your customer. Do you really know her? Find out.

  • Have open-ended conversations to build empathy and understanding
  • Find her needs, pain, and behaviors
  • What information will help you make a design decision?
  • Test your hypotheses

A favorite anecdote of mine is, of course, from Mint:

When Patzer had the idea for Mint, he formed a few messaging concepts to describe it. One said something like “Mint is your Money Ninja.” Another said “Mint is your Money Champion,” and another said “Mint is Free Personal Money Management Software.” These were all written out on a page which explained the idea for Mint and what it did.
Patzer then went to a train station, stopped people, handed them the concept, and asked them questions like “Does this sound like something you would use?” Patzer suspected that people probably would say “yes” more often than they actually would use it, so he would have to gauge their enthusiasm in their “yes” answer. If they were enthusiastic and genuine about their answer, Patzer would weigh that more heavily than if someone monotonically said “yes.”
How Mint Grew, Zach Bulygo

Don’t be intimidated by user research. Aaron’s an engineer but overcame his inner introvert to go to the Mountain View train station and talk to people about his idea. They had nowhere else to go! Mint being successful wasn’t an accident, it was by design. We had a combination of knowing what people wanted and needed, pain from existing solutions, as well as behaviors they currently exhibited. A story, product roadmap, and go-to-market strategy was constructed out of this understanding.

The what in our story is the problem you’re solving, which you’ll find through conversation about needs and pain. Is it what you thought originally? Have an open mind, listen more than talk. Ask general questions about your market and their problems before talking about your solution, see how they’d describe it first.

The why is twofold: first, for your customers and marketing position, why you? What separates you from the competition. Silicon Valley startup doctrine, or dogma, tells us that you want to be 10x better. Otherwise the benefit of switching from their product to yours wont outweigh the cost. For Mint it was automation: Quicken required manual entry of all transactions to get any value, Mint was automatic. 10x less work, at least, for the equivalent value. The second why is the inspirational why that Ben Horowitz talked about above: why do you exist?

When is timing. Periscope certainly wasn’t the first streaming video startup, but it launched at the right time, once LTE networks covered most of the country. Qik, SocialCam, and others weren’t so lucky. The how in our case will be how you’ll deliver on your product promises.

One tool I’ve used for every company I’ve worked with since Votizen is the positioning statement.

“For [target customers] who have [a problem], our product is a [new category] that provides [solution to that problem]. Unlike [the alternatives], we have a [key differentiator].”

Just fill it out like a mad lib. The other one I use is a value proposition statement. Value is in essence benefit minus cost that a particular market finds acceptable. If you want positive value, B > C. Both of these, along with mission (what do you want to do) and vision (what will the world look like if you do) statements, are great tools that will help you tell your story, and align and motivate your company. If you are in a many sided market, you’ll need more than one value proposition. LinkedIn for example has a very different value to a recruiter than a job-seeker, and they need to be messaged differently.

Know the entire journey, but choose where to focus

An important part of designing with intent is understanding all that your user is experiencing in your problem space, and deciding where you want to focus. A startup has limited time, limited resources, you need to do one thing or a few things well, not tackle everything at once.

Journey map by Kevin Shay, Brigade

This is a journey map I love, created by one of the product designers at Brigade. In just one day, he conducted a series of open-ended interviews / listening sessions about how people use group chat tools. He then synthesized all of those conversations, grouped their responses into activities (pink circles), and brought the most common responses into the core loop (dark grey) above. We then decided which of these problems our prototype was going to focus on.

Mental Model by Indi Young

Another tool for this type of exercise is the mental model. The one above encompasses activities associated with the morning before work or school. The large areas represent themes, the blocks above the line are activities, and the products needed for the activities are below the line. The darker blocks represent P&G product offerings. Now, you’re not P&G, and you’re not going to create this many products, but it’s important to know the context of your product’s use. It definitely will help you make better design decisions. An example I often cite is how we used to quickly log people off due to inactivity at The engineers that designed that very sensible security measure weren’t aware that fit into a set of evening activites for moms that included cooking, taking care of their children, and a number of other things all happening concurrently. We were logging out our moms since money management wasn’t an uninterrupted activity.

  • Understand when your product is not being used
  • What happens in the gaps? What changes?
  • Consider mobile context (short bursts, many interruptions) vs. desktop (longer sessions)
  • How are you handling interruptions?

Validate assumptions, learn cheaply, lower your risk

[The design process] is about designing and prototyping and making. When you separate those, I think the final result suffers. –Jony Ive
Ballot guide journey by Kevin Shay

This is another great deliverable by the same product designer at Brigade. It’s another form of designing the journey, writing out the steps, and being able to gather the team to decide what’s working and what isn’t. One of the first things I do when working with a company at Bessemer is to see how well the teams are aligned: starting with the customer, then the problem statements and messaging, and only then do we start to look at actual solutions, and designs. These types of deliverables are key to lowering risk before starting more expensive kinds of design and engineering. They separate out the organizations that are using design properly, and the ones that aren’t. Both organizations may get to the same place eventually, but the one that doesn’t fully understand their customers will end up redesigning many times over. A/B testing is one tool, it’s not an excuse to skip the design process.

A paper prototype with a fake phone that you can move around by David Wright.

In this example, a designer created a bunch of screens on paper, and can simulate screen changes in a user research session by simply moving the phone around. This is much cheaper for most people than creating a prototype on a computer, and certainly building something. Can you learn everything? No, but you can learn quite a lot proportionally to the time invested, and that’s what we want in the early stages.

Wireframe by Galyn Bunnell

Once your flows and story is straight you can move on to some rough wireframes to get your interactions in order. I judge interactions this way in usability:

  • Obvious when a user is comfortable, moves through, meets expectations
  • Intuitive if user pauses, but can answer when you ask “what do you think it does?”
  • Poor if none of above or desirable by the user; consider onboarding, or scrapping

I always try existing patterns first, and only innovate where its necessary to do so. There’s no prizes for making a really gorgeous pattern that nobody can figure out, only make the tradeoff if it is much more powerful for an intermediate user.

Marvel prototype

Prototyping offers you another amazing tool to validate and lower risk. The tools are getting better so quickly it’s tough to keep track. Marvel, Pixate, Flinto, FramerJS, and many others are available, all with their own specialties, and differing support for animations, gestures, motion, and ease of integration with your UI design package.

Use visual to affect brand, build trust, connect emotionally

The thing that people don’t understand is that the only way you can be successful with your branding is if you have a great product to sell. If people go home and aren’t happy, that won’t work. Your product has to stand up for itself. –P. Diddy

The last weapon in your design arsenal is visual. Once you have everything else nailed, this is where you really hook your customer., 2015

Notice that nowhere on Uber’s homepage do you read anything about their product. They’ve moved into aspirational territory. The black and white imagery with a single color accent is similar to something you’d see from a luxury apparel brand. The headline “be the boss” is pure aspiration, and the subheadline reveals a benefit of letting someone else handle the driving without talking about a single feature. Their product design similarly had to evolve in the face of relentless competition., 2015

Square similarly has traded in the photo of their product, and gotten rid of anything having to do with their technology or offering here. This is a barber, a typical small business owner that Square wants to appeal to., 2015

Clearly targeted at younger women, the photos on Forever 21 look more Instagram than high fashion. Women have been shown to respond better to strong grids as well, Pinterest being another example.


Lastly, a great process always has a great feedback mechanism, and allots cycles for improvement. App store reviews, net promoter score, Intercom, customer service, user surveys, and social media monitoring all offer some measure of feedback.

When you do listen to customers, and get that feedback, don’t take them literally. Behind every feature request there’s a need, find out what it is. At we had tons of users asking for the ability to add cash transactions every day. Doing exactly what they asked for would have had the unintended consequence of creating more work for everyone as all our users would have to balance Mint like a checkbook, as the balances would quickly get out of sync. We realized the need was to categorize cash, so we allowed people to split ATM withdrawals instead to solve the problem. Remember that the people writing you in customer service forums and emails are almost always a vocal minority. Figure out what the silent majority thinks, and check with your vision, before you make any decisions. You can also may be able validate and test cheaply by launching a feature to a cohort of users.

Also remember to learn to love failure. Killing features and seeing if anyone complains, and making frequent small improvements are luxuries afforded us by the software business—take advantage.

Thanks to for the speaking invitation, and Jared Spool, Luke Wroblewski, Robert Hoekman, Jr., Indi Young, and Google Ventures for the research and hard work they share with the design community.