Humanoid musings


humanoid musings

Book Review: Thinking Fast And Slow

Just finished Daniel Kahneman’s excellent cognitive psychology book, Thinking Fast and Slow, and I highly recommend it. I think I enjoyed it so much because knowing how people think is so crucial to what I do for a living: building web & mobile products. It’s also in the same genre as Predictably Irrational, another excellent book that tries to make sense of why people behave the way they do. Below are some quotes from the book that really stood out for me, along with how they’ve helped inform my thinking as a product manager. 

"We are prone to overestimate how much we understand about the world and to underestimate the role of chance in events. Overconfidence is fed by the illusory certainty of hindsight."

Overconfidence is a big running theme in the book. It’s something that I think can be an easy trap for product people to fall into - the idea that we understand what users want, and when we build something “successful”, we knew all along that it would play out that way. That’s usually not the case. A humbler approach - rolling out experiments and iterating along the way - is so much more vital to finding the right formula. 

"You are more likely to learn something by finding surprises in your own behavior than by hearing surprising facts about people in general. "

This is a good reminder that you are, above all else, the primary source of “user research” for your product. You can always uncover new and interesting things about your own behavioral patterns, even for features that you designed/engineered yourself and have already thought about for weeks. That doesn’t mean you should discard learnings you uncover from observing/talking to others; it’s more a reinforcement that you need to be the primary user you turn to.

"…you experience cognitive strain when you read instructions in a poor font, or in faint colors, or worded in complicated language, or when you are in a bad mood, and even when you frown."

The concept of “cognitive ease” takes up a full chapter of the book. It’s a nice affirmation that product designers should grease the wheels at every turn in order to improve usability of the product, because even a very small negative “strain” the user has to endure will add up and ultimately be the difference between whether the user returns. It might not even be something the user will notice consciously - it’s an example of the brain thinking “fast” and acting without putting much deep thought into it. But it influences behavior tremendously. 

"If you like the president’s politics, you probably like his voice and his appearance as well. The tendency to like (or dislike) everything about a person—including things you have not observed—is known as the halo effect."

I like this quote because it underscores that the “likeability” of a person can tip the scales and provide that small bit of leeway. The same goes for products & companies. Users tend to forgive missteps if they genuinely feel a small connection to the product due to this “halo effect”. One of the things that we consistently reinforce at Yelp is the idea of delighting users through really small, low-cost methods (one of the tenants of the Kano model). Leaving amusing release notes in our apps is one example of how we do this. 

"You see a person reading The New York Times on the New York subway. Which of the following is a better bet about the reading stranger? She has a PhD. She does not have a college degree. Representativeness would tell you to bet on the PhD, but this is not necessarily wise. You should seriously consider the second alternative, because many more nongraduates than PhDs ride in New York subways."

Kahneman exposes many instances where humans just flat-out overlook basic logic. I caught myself multiple times making assumptions based on limited information. It’s a good reminder of how important it is to be careful when analyzing your product. Reading too much into small sample sizes, overreaching on conclusions, estimating poorly - these are all pitfalls we face largely because we can sometimes feel too confident in our hypotheses to see the contradictory results staring us in the face. 

"Once you adopt a new view of the world (or of any part of it), you immediately lose much of your ability to recall what you used to believe before your mind changed."

This is something that anyone who works on a product can identify with. Being able to continually bring a fresh perspective to problems you’ve spent a good amount of time thinking about can be tiring, but it’s the only way to improve. Sometimes all it takes is a slight 1 degree shift in thinking, but that subtle change can be difference in overcoming a longstanding, engrained bias that previously blinded you from reaching a new Eureka moment.

Great [products] are played not made

My poor wife got quite sick this long 3-day weekend, so I spent a lot of time indoors watching movies, videos, etc. At one point I got swallowed into watching a series of “Making Of” videos on YouTube by the team at Bethesda Softworks, the makers of some really great video games like Fallout 3 and the Elder Scrolls series.

Anyway, one web page led to another and I stumbled upon a great quote by Todd Howard, lead game designer for the studio:

"Great games are played not made. ‘You can have the greatest design document ever made, and you’re going to change 90 percent of it as soon as you play the game.’"

This doesn’t just apply to games. It pains me to say it, but product design documents are meant to be bloodied up, torn apart, & flipped upside-down.

This is never the intention, of course, when designing a new product, and the goal is always to minimize this as much as possible, but it’s what inevitably happens. No matter how well you plan and think through the problem you’re trying to solve, a design will go through numerous iterations as it’s built and used (or “played with”). Everything’s a prototype right up until it’s, well, live and in use by users. And even then, it’s a work-in-progress.

It’s this mentality that I think makes building products both so difficult yet so rewarding - it’s a journey where sometimes only 10% of the original idea / design survives. 

Now, when’s Fallout 4 coming out anyway?

To listen or not to listen…

Today I came across this interesting Harvard Business case study (though it’s slightly dated). It describes a Spanish chef who has a very unorthodox restaurant business. Instead of selling just food, he sells “experiences”. How? Well, for one, it takes hours to get to his restaurant, and only 8,000 people a year actually get to be on the guestlist to even sample his meals.

Embedded within this study is the notion that listening to your customers is not always the correct approach. Certainly we in the tech industry are taught to listen to users and constantly gather feedback+iterate to make our products better. To me, the article is saying there’s a difference between listening+following and listening+leading.

For example, your user comes to you with the following:

Joe User: “Your web app need a way to report bugs. You should provide a form to do so with multiple fields, like they do on [insert competitor’s site]. Please build that!”

To listen+follow your user is to take his solutions and incorporate them as-is into your product. In this case, you’d spend X days building this submission form into your app exactly as specified.

To listen+lead is to take your users’ problems and lead them to the right solution based on your product’s core strengths and business strategy. In this case, instead of building a clunky submission form, you could build in a type of auto-detection that sends thrown errors to you. This could remove the need for users to worry about filing bugs. It also makes diagnosing these bugs easier as you’ll have all of the info about the error and not just what the user remembers to submit.

As a developer or product leader, knowing this difference is crucial to serving users in the best possible way. Everyone wins because the user is led to the best solution.

Money quote from the case study:

"Adrià’s idea is that if you listen to customers, what they tell you they want will be based on something they already know," Norton observes. "If I like a good steak, you can serve that to me, and I’ll enjoy it. But it will never be a once-in-a-lifetime experience. To create those experiences, you almost can’t listen to the customer."

Online site design tools

As I worked on a new homepage for Gifty, I was struck by how easy it has become to use free web-based tools to create graphics, designs, and basically everything you really need in order to develop a web front-end that’s at least half-decent. I thought I’d share some of the tools/sites I used:

Balsamiq Mockups - I learned about this from the Techzing! podcast. It’s a really cool web-based wireframe / mockup tool (though you get a nag message every 15 mins or so if you use it free). It has a bunch of pre-rendered graphics/buttons that you snap into a blank canvas, helping you visualize your site layout. My crappy rendition is shown here.

Picnik - Just bought by Google, this service lets you upload and manipulate graphics. It appears lightweight at first but it did everything I needed it to do, in a really easy way. Can’t wait to see what becomes of it with Google’s resources behind it.

Colour Lovers - A neat site that I used to generate a “palette” of colors to use. I already had my initial red, so finding a subsequent palette based on that was easy and really helped me figure out how to keep my colors consistent. Here’s my palette.

Da Button Factory - Yup, basically does what it says - button creation made easy. I still remember achingly creating my own .gifs for an old website I made way back in 2002 in Frontpage. No more!

Free Gradient Maker - Had to include some gradients to make it all web 2.0-ish.

jQuery - Not graphics/design related, but I found a library that powers an animated slideshow on the homepage.

Paintbrush - Ok, I snuck in a desktop app here at the end. But it’s free, and is basically the equivalent of Paint for Windows.

What does this all mean? On first glance it appears to bring us closer to a browser-centric OS world, as the forthcoming Chrome OS intends to do. But truthfully I needed to use that last desktop program to tie everything together. More likely, this all means that any walls that existed 5-10 years ago for amateur web designers to create decent-looking sites are coming down.

But most interesting (to me), it appears to signal an opening for a company/product that can tie all of these kind of services together in a neat and easy-to-use package, especially aimed at web developers who are decent with back-end work but are lost in setting up a front-end. Will be exciting to see where we are in another 5-10 years!

A Real Tomorrowland


I visited Disneyland last month and had a great time. However, the worst part of Disneyland is always the number of people waiting in lines for stuff I want to do. We even went on a Monday in the wintertime and it was jam-packed. It got me thinking about the incredible amount of data that Disneyland could harvest and how they could benefit their customers by opening this data up to them. For example, Disneyland could use sensors to track the movement of their customers (in aggregate, of course!). Give everyone a sticker or pin or something to wear  as they enter the park, explaining of course what it is. Then take that sensor data and do cool things like:

image Far-fetched? Probably, given the concerns people would likely raise about privacy, and given the primary customers that Disney wants to attract (worried parents). But maybe Disney would be more comfortable rolling something like this out at a smaller scale and iterating…they already have a perfect area of the park to test it out on - Tomorrowland !

[2nd photo courtesy of flickr/huffmans]

The Data Deluge & Business


Interesting series of articles in the Economist this week about data. While it mentions data-hungry companies like Google and talks about obvious implications such as advertising, I like how it also references how more traditional, older businesses are tapping into enormous amounts of data in order to make their products/services better:

"Older industries crunch data with just as much enthusiasm as new ones these days. Retailers, offline as well as online, are masters of data mining (or “business intelligence”, as it is now known). By analysing “basket data”, supermarkets can tailor promotions to particular customers’ preferences…"

It also talks about the dangers of living in an age when so much of our lives are recorded into bits and stored on servers that we as consumers never see or touch:

"But the data deluge also poses risks. Examples abound of databases being stolen: disks full of social-security data go missing, laptops loaded with tax records are left in taxis, credit-card numbers are stolen from online retailers. The result is privacy breaches, identity theft and fraud."

I’d also like to hear about the dangers that aren’t related to theft or government big-brotherness. For example, using data to deliver hyper-targeted news content to sites like Yahoo News (for example) could create silos that keep consumers from being exposed to any stories that don’t fall within their “profile”. Or, credit card companies that boast of “fraud detection” on every purchase could piss off consumers who suddenly can’t make any purchases while on vacation in Italy because those purchases are flagged as fraudulent expenditures (seeing as these purchases are far outside of said consumer’s typical purchase pattern).

While it’s important to focus on the security/fraud implications, I wonder how the actual *product* implications will play out as companies, old and new, feel more empowered to leverage the data at their exposure to grow their bottom lines.

Put the Music back in iTunes

As a music geek, I love all of iTunes’ options and features - smart playlists, tagging, even stuff like per-song volume adjustment and start/stop time. I use them all and they’re now second nature to me. It’s an extremely deep application, with a ton of nooks and crannies.

And yet…as iTunes matures, its focus has shifted from being the best music-listening experience to being a better store & hub for iPod & iPhone. Aside from Genius, a feature that tries to better group similar songs into mixes, iTunes simply hasn’t been innovating as fast as it could. I understand their focus here - after all, the overall strategy was likely always to make iTunes just good enough to help Apple push the sale of their hardware (like the iPod) and control the experience. But all that said, here are my suggestions for improving iTunes so that it’s once again on the leading edge of music applications:

Again, I recognize that Apple’s focus for iTunes is now centered around integrating its iPod, iPhone (and soon, iPad) products. I just hope they remember to keep innovating in ways that actually make the music experience better. Because if they don’t, they’re leaving themselves open to being flanked by more agile competitors like the aforementioned Songbird, or the Europe-only Spotify…and that’s when things get really interesting.

I just want to wave and go

I love the intentions behind Square, Payware Mobile, and iCharge - 3 services that are pushing to take personal money transactions to the next level. However, the way they’re all approaching the problem still falls short of the ease, simplicity, and innovation that I wish were upon us by now.

To back up: the idea behind all of these services, more or less, is to make it easy for individuals and small businesses to process money transactions. Their proposed solutions all involve hooking a small attachment to a mobile phone, which allows you to scan your credit card and register the transaction. Here’s a quick video on how Square’s implementation works, for example: Various tech blogs are excited about this. While it’s interesting, this model is still not ideal…basically, I’d rather credit cards be cut from the equation altogether. I just want to use my phone to transmit a payment to the merchant or friend in question. Maybe have my credit card info embedded into my SIM card and utilize near-field communication or something…and smooth the transaction out with a standard mobile app that runs on both phones. And just email my receipt to me and don’t require me to sign anything. Wouldn’t that be nice!

In fact, companies in Asia like Docomo are already doing this. And here in the states, it’s been written about and anticipated for *years* now (check the date on that article!). Unfortunately, I suspect it’d take cooperation from the telecom industry, among others, to make this a reality here. Until then, we’re stuck with services that still require me to reach into my wallet, fish out my credit card, and sign something.

So in conclusion, I just want to be able to “wave and go” - i.e., wave my phone next to my fiance’s phone to reimburse her for dinner. Let’s hope these companies are working behind the scenes on pushing the necessary players toward this, even if publicly they’re releasing solutions that only get us partially there.

[photo courtesy of flickr/sharpshutter]

James Cotton - No Cuttin’ Loose

James Cotton - High Compression

In the mood for some serious blues harp?  If so, check out James Cotton. He’s one of my favorite blues musicians, period. This is probably because he’s a bit less traditional than others - his classic songs incorporate funk drum rhythms that are energetic and driving. He couples this with one of the most wicked harmonicas out there.

This song showcases some serious harp solos - I highly recommend his entire High Compression album even if you don’t like traditional blues music.  Play it loud !

Vegas, Facebook, & Google

I was in Vegas the other week and a funny thing occurred to me. The amazing hotel/casinos that litter the Vegas Strip - the ones that are publicly open as gracious hosts - are very similar to the virtual “hosts” that can be found on the web. I’m talking about sites like Facebook, Google, Yahoo, and others.

Take the Bellagio hotel & casino, for example. Out in front, a free fountain show takes place every 15 minutes, set to music. Inside, huge, impressive statues and art exhibits (there was a Chinese New Year one when we went) are littered throughout. You’ll find similar free entertainment in every casino - the Venetian’s got a really impressive indoor river complete with working gondolas and live shows…Caesar’s Palace has got a crazy Disney-esque animatronic show called “The Fall of Atlantis”, along with massive sculptures and statues.

These all exist to attract people to the casinos and drive “conversions” - or, to get people gambling! Now take Facebook. They’ve built an amazing web application with numerous attractions - fan pages, games, walls - all in the hopes that users will eventually spend time clicking ads or purchasing virtual goods. The draw is the social interaction - all polished to perfection in order to drive conversions in the form of ad clicks and virtual goods purchases. Same goes for Google’s services and any other site that relies so heavily on free goods.

With no upfront-paying customers to lean on, both the casinos and these web services have to rely on constantly providing quality, new, free stuff to draw their customers in. The moment they stop innovating in this regard, it’s over. It’s why things move so quickly in both industries - and why following their progress is so much fun. Now if only Facebook & Google also offered cheap all-day buffets for their users…