Grow your store with our weekly podcasts, videos, and letters. JOIN NOW
Ethercycle logo

Within 24 hours of Apple announcing iOS7, we downloaded the developer preview to our own phones.

It is the biggest refresh of the OS ever. It's a welcome change as iOS was starting to feel stale, especially in comparison to the animation love-fest of Windows Phone. Every single thing in the OS has been redesigned, so there's too much for us to go over in detail, but let's talk about a few features we've noticed and loved:

  • Accelerator based animations: When the phone is tilted, the icons and wallpaper shift slightly to give the phone a subtle depth. It's both superfluous and amazing.
  • Multitasking: The new task switcher works like the old WebOS from the the HP TouchPad. Swiping an app's "card" up off the screen closes it. Much handier than the old method. Supposedly multitasking has been rewritten to improve battery life, but we haven't noticed anything yet.
  • Camera: The new camera includes a square crop for the Instagram fans, a nifty new interface, and even filters with live preview.
  • Control Center: From the lock screen, swiping bottom-up reveals a new control panel with toggles for wifi, bluetooth, and brightness. It even includes shortcut keys and a flashlight. Finally, a flashlight.
  • Photos: New albums can be generated on the fly according to date or location. It's like iPhoto.
  • App Store: Apps now auto-update in the background (like Android.)

Aside from the new and beautiful interface, there's dozens of new features. It's exciting to see Apple update our favorite smartphone, especially with a design language we hope to see continue in to OSX. Overall, I like iOS7. Admittedly, it's in part because it feels so new and different. It's bright and tries to present itself as fun. It screams, "Computers are fun!" and I'm okay with that.

Dedicating 20% of our time to personal projects and research is the secret sauce that makes our client work successful.

We call it Ethercycle Labs. It’s shaped our culture in to one of open exploration, let us take risks that clients couldn’t, and given us insight on everything from the landslide shift toward mobile to the dynamics of crowd-funding.

To explore projects outside of clients’ demands, we began a Labs initiative in March 2012. The Labs program allows us to catalyze new thinking while getting into the consumer mindset. This research has provided us guidance on what to support for our clients, helped us reach new customers, and just generally been a great learning experience for the whole team.

Sharing everything we do publicly for free has been our best advertising. Our labs projects are almost universally things we built for ourselves that we shared. Collectively, those releases generate up to 1 million page views monthly, resulting in a lot of referrals to Ethercycle, and subsequently new leads or newsletter subscribers.

Explore what you can do with no restrictions, share it, and see what it can do for your business. At worst, you'll have learned something. At best, you'll have made something wonderful.

In 2009, I quit my job to launch Ethercycle. Back then being a consultancy hadn't crossed my mind.

Quite the opposite, I wanted to build an ecommerce SaaS app for bike shops. A little over a year after we started, we realized that our business model was flawed. We had made ourselves too dependent on other people. At the same time, we had met a lot of people who misunderstood what we were trying to do. These new friends did, however, understand that we knew the Web, and they would often ask: "Could you maybe help with our website?" Looking at the problem rationally, it became obvious that we needed to pivot.

Startups typically need to pivot and evolve their business model over time, especially as customers start to use the product or service. In order to pivot and keep your business moving forward, set limits. If a business model isn’t working, you have to set a limit on saving it.

Decide in advance, before your emotions get the best of you, how much time, money, and effort you’re willing to put in to iterating a business model before accepting that it isn’t working. Failure doesn’t have to be scary if you embrace it early, learn from it, and move on to the next thing. The last thing you want to do is rearrange deck chairs on the Titanic.

During the month of March 2013, Calming Manatee received 613,792 total page views. Of them, 26.87% were from mobile devices.

Unique Visitors by Mobile Device

  1. iPhone - 59.45%
  2. iPad - 14.06%
  3. iPod - 3.03%
  4. Galaxy S III - 2.10%
  5. Other - 21.36%

Unique Visitors by Mobile Operating System

  1. iOS - 75.74%
  2. Android - 22.60%
  3. Windows Phone - 0.93%
  4. BlackBerry - 0.62%
  5. SymbianOS - 0.05%
  6. Other - 0.06%

Unique Visitors by Viewport

  1. 320x480 - 45.06%
  2. 320x568 - 16.15%
  3. 768x1024 - 13.77%
  4. 720x1280 - 5.61%
  5. 480x800 - 3.01%
  6. Other - 16.40%

Audience Demographics
Compared with internet averages, the site appeals more to users who are under the age of 35; its visitors also tend to consist of childless women browsing from school and work who have incomes between $30,000 and $60,000. Roughly 90% of visits to the site are referred from social media.

Insights

  1. The Galaxy S3 is clearly the dominant Android device. Consider adding one to your testing suite. Don't spend too much though as it only accounts for 1 in 50 mobile visits. Consider a Bad ESN unit from eBay.
  2. These stats are largely unchanged compared to January. There was no market upheaval this quarter.
  3. New BlackBerry devices became available worldwide, yet their market share has not yet changed. We'll keep an eye on it.

We curated this checklist to make our own website testing more efficient.

We run through it before we launch a client's site. We're sharing it with other designers and developers to make the web as a whole better.

  Businesses fall into a trap where when they have a lot of sales coming in, they don't put as much effort into advertising or marketing.

  It's a combination of two factors. It's the thought that "I have more customers than I can handle right now", and a lack of time to devote to marketing. The danger in doing this is that when you do run out of customers or sales, you have to scramble to resume your marketing efforts to get back where you were. The result for many businesses is an unending cycle of ups and downs, or empty and then overflowing sales funnels. The number one commandment of marketing is to never stop selling. It's easier than it sounds if you automate your marketing efforts.


Automating Your Social Media

  Let's start simple. The first thing that you stop paying attention to, is your social media efforts, right? That's probably the easiest thing to fix. There are two central approaches to social media that we can automate: content distribution and content curation.

  To distribute content, I use two tools: IFTTT and Buffer. Buffer lets me quickly schedule postings to Facebook, Twitter, and LinkedIn from my phone or my desktop. There are several platforms like it, but I prefer Buffer because it's incredibly easy to use. Once we've added my accounts and setup schedules, I add scheduled updates to it just as easily I would update Twitter, and I can push updates to it through their API. That's where IFTTT comes in to play.

  IFTTT stands for If This, Then That. It's a free service that lets you create "recipes" or simple triggers and actions that leverage the power of other services with APIs. Let's use Instagram as an example. Whenever I post a photo to Instagram, IFTTT sees it and looks for recipes that start with "If new instagram photo by ethercycle," and then it carries out any actions I've specified. I think photos are great content marketing, so I have IFTTT post my new photos Tumblr, Facebook, and Twitter. I even have it backup those photos to my Dropbox.

  To get you started with IFTTT, I've shared some of my recipes:

  1. Instagrams to Tumblr Photo posts
  2. New Flickr photos to Tumblr photo posts
  3. New Instagram photo upload a new Photo on Facebook Page Album
  4. Instagram to Twitter
  5. Send interesting business articles to LinkedIn
  6. Buffer my YouTube videos
  7. Download my Instagrams to Dropbox


Automated Content Curation

  We've come up with an easier way to distribute our content, but we need to create content first. To find new and interesting topics and inspiration, subscribe to a newsletter in your industry, and some general newsletters. At the very least, sign up for Dave Pell's NextDraft which correctly claims to have the day's most fascinating news. I've taken content curation a step further, and use IFTTT to send RSS feeds of industry sources directly in to my Buffer account. Once a day, I go through and a delete all but the most interesting or relevant articles. If I have time, I'll try to add commentary to each tweet.


Content Creation: No Shortcuts

  Now comes the hard part. The best content marketing is the creation of original blog posts or articles. Your website has given you homework: you need add 3,000 words of original content to your site every month if you want to stay relevant. As the principal at your business, you're best suited to create the most compelling content. It is hard work, but there are ways to make it easier. The hardest part of writing, for me anyway, is editing. I hate it. I have a freelance copy-editor I pay by the hour to edit case studies and web content. It's easier to get motivated writing when you know you don't have to proofread your work yourself if you don't want to.

 If you're like me, you should now be able to get started on writing. But, hold on, I need to check my email. And Twitter. And feed my cat. You get the point, distractions are the real reason blogging doesn't happen. I have a solution. I sit in my parked car and dictate a blog to Siri. That gives me a rough draft easily, and it's even written in my natural tone. Siri may make some pretty atrocious errors but she's a good listener and doesn't interrupt me.

  If you want to keep your sales funnel full, you need to stop procrastinating, and start marketing smarter through automation.

The trap that many creatives fall in to when marketing themselves is not differentiating themselves.

Saying "Hello, my name is Chris, and I make beautiful websites" doesn't help to sell anything. A tagline like that is only slightly more productive than writing "Welcome to my website!" like it's 1998 again and we're on Geocities. And saying you make "beautiful" anything is like a restaurant saying "Hi, we're McDonald's and we make good cheeseburgers." Yes, it's factually true, but it's also incredibly generic. There's no one claiming they make a bad product in their tagline. It'd be better to say what you make for who you make it and why you make it. For example, "I make mobile websites for non-profits that increase donations." The critical difference is that it's no longer about the creative, it's about the client.

Slow is annoying.

People like things fast. Fast cars, fast good, and fast downloads. Who wouldn't? There's nothing more annoying than sitting on your phone and waiting for something to load. And if something takes too long to load, we move on. That's why performance of a website is so important.


We know from surveys that people expect pages to load in two seconds or less. That's our definition of fast because it's what our users expect. We also know that nearly half of people say they'll abandon a page that takes longer than four seconds to load. In actually testing we know they're a little more patient than that, but if a page takes longer than five seconds, we can expect about half of people to abandon the page.


Your bounce rate isn't going to reflect people who leave due to excessive load times. If the user gave up mid-download, your Google Analytics snippet probably never even had time to load. (Your javascripts are in your footer, right?) Using tools like Pingdom's Full Page Test, Webpagetest, or YSlow.


For that reason we need to consider performance as part of design when we create anything for the web. The reasoning behind it is simple: the faster our page loads, the more people that view our site, and the happier they'll be. More happy users means more conversions.


There are three approaches one can take to reducing the page load on the site. The most direct is to reduce the file size which we can do by stripping out content, or by compressing existing content via minification or gzip. First ask, "Do we really need this?" and then ask "How do we make what's left smaller?" Editing and then compressing content is fairly straight forward because we're just making it smaller.


We can also reduce load time by reducing overhead. In addition to considering the page size we also need to consider the total number of requests made by a page. Each request has its own overhead in that the browser has to handshake with the server, and the server can only handle so many concurrent requests. That overhead gets worse when the request is to a separate server because now it also includes an additional DNS lookup. Solving this is easier than it sounds. We can combine style sheets in to a single file, combine javascripts in to a single file, and even combine images. By creating "sprites," we can load a group of related images in to a single filmstrip and then "crop" it to just the relevant image using CSS. For site's that use a lot of SVG icons (you're using SVGs, right?) we can load them all at once in an icon font. We generate ours using IcoMoon.


Now that we've gotten our page down as small as it will go, and we've reduced our total number of requests, we can take our third and final approach which is to make our page render faster. By altering the order in which things load we'll give our site the appearance that it's loading faster. Browsers render things in the order they appear on the page. We can exploit that by loading our stylesheet first in the site's head which will render the site faster. If we move our JavaScript to to bottom of the page after the body tag, we'll add our JS functionality only after the site is rendered. Doing this doesn't change the size of the page but it will have the effect of making the page render faster. In some browsers we can even take this a step further and use the "async" and "defer" attributes on javascript. Async will specify that a file can be loaded asynchronously, and defer will specify that a file should be loaded only after the rest of the page is loaded. The final tool in our belt is to use good old progressive JPEGs instead of optimized JPEGs. Sometimes they'll be a little larger, but they render faster so they'll appear to load faster.


These are all things an experienced developer should be doing to begin with. Websites that load fast aren't just best practice, they're fun. Tweaking a page to reduce its load time is like playing golf. You want get your score as low as possible through careful optimization of techniques.

How do frameworks and grid systems work with the Mobile First approach?

Dorit Willenbrock asked:


"I'm just wondering - and maybe I have missed something, sorry for that in advance - but if I want to do a mobile first approach to redesigning my website, then I don't see how a default framework or grid system can work out. Sounds like the search for the perfect framework or grid system to match my requirements, all that analyzing and testing, wouldn't that be more work than just starting building mobile first...?"

Dorit,


If you don't have to use a framework, then don't. :) Sure, a framework has (someone's idea of) best practices baked in to it, but it also comes with a lot of overhead that you may not need for your project. If you're comfortable not using one, then good on you!


Mobile first just means that the base style of your website is for a 320px wide screen, and anything bigger than that is done in a media query. The only concession this requires is conditionally loading respond.js to add media query support to IE8. If you don't care about IE8, then skip this, but IE will always use your 320px base default.


Remember, the "magic" of responsive web design is committing to doing your CSS in percentages. The rest is just 5th grade math.

Did Responsive Design Kill the use of Photoshop for Web Design?

We still use Photoshop, and we definitely still create high-level visual design comps. However, we do it much later in the process now, and we only mock up key pages. Our workflow now looks like:

  1. Strategy, research, meetings, planning, etc. Until we know the business, we can't do any effective design. (Remember, this is design, not art.)
  2. We write a design brief to frame the design in the context of a business problem or objective.
  3. We write a 2-4 page content inventory and basic strategy that outlines the content we need from the client or that we need to create. Design does not proceed until we have this content.
  4. We create wireframes of key pages that show how the content will be used.
  5. Our strategy milestone is complete! Celebrate.
  6. Design begins. We create and revise moodboards (or style tiles, whichever you prefer) to define the visual without confusing it with content.
  7. Once the moodboard is approved, we can then apply it to the wireframes to create the traditional visual comps of key pages that Photoshop was traditionally used for.
  8. After feedback, revisions, and approval, the design phase is complete, and we celebrate once again.
  9. Development begins!

Though it's worth nothing that prior to this, we may have created development prototypes to test concepts. For example, when working on the Crain's Top 40 under 40, we created a prototype of the sliding tile interface before we had content or design just to test the concept. This is especially helpful when working on short deadlines where revisions aren't an option.

Responsive design has in no way killed Photoshop, but it has changed workflows and certainly made us question our processes as an agency.