“We\u2019ve always felt that Intercom has been part of driving this movement, and it\u2019s exciting to see many others joining in”<\/p><\/blockquote>\n
What\u2019s more, this straight-up design innovation is happening in a somewhat-unlikely corner of the industry: business software. While most of us may have looked down our noses at the design of \u201centerprise software\u201d a few years ago, some of the most interesting new products and design ideas are coming out of companies who are building tools for work. We\u2019ve always felt that Intercom has been part of driving this movement, and it\u2019s exciting to see many others joining in.<\/p>\n
Smaller upstarts are building exceptional products by out-innovating lumbering incumbents. SaaS products like Airtable<\/a> (better Google Sheets), Coda<\/a> (better Google Docs), Slack<\/a> (better IRC), Notion<\/a> (better Evernote), Superhuman<\/a> (better Gmail), Zoom<\/a> (better Skype), Linear<\/a> (better Jira), and yes, Intercom<\/a> (I couldn\u2019t possibly!) are all aimed at carving off a healthy slice of a large existing market.<\/p>\n People used to say that you could come up with startup ideas by replacing old Unix commands<\/a> or spreadsheet hacks<\/a> or a Craigslist vertical<\/a>. Now it\u2019s even simpler<\/a>: choose a massively successful product with a large user base, hone in on a specific segment of their customers, and simply ship brilliant product design work aimed specifically at them. Free of the burden of having to be relevant to billions of very different customers, these \u201cbest of breed\u201d products are now at the frontier of product design innovation.<\/p>\n You\u2019re probably familiar with disruption theory<\/a>: essentially the strategy of smaller companies building a simpler, cheaper alternative to an existing product.<\/p>\n Best of breed products are similar, but come from the opposite direction: rather than cleaning up with the low end of tiny customers who want something simpler, they can cherry-pick the most sophisticated (and often most lucrative) customers from higher up the market. Think of it as disruption from above.<\/p>\n “Such is the enormity of the market for business software today that even a small proportion can sustain a significant business”<\/p><\/blockquote>\n It\u2019s now clear that a decent chunk of users who spend all day using (and feeling the daily frustrations of) those older tools will happily switch for even a marginal improvement, often layering several best-of-breed solutions on top of an existing mediocre one.<\/p>\n And such is the enormity of the market for business software today that even a small proportion can sustain a significant business. If Superhuman can win over even a tiny percent of Gmail users with a more focused product, they will be very happy indeed.<\/p>\n Whatever the tipping point, it\u2019s clear that buyers are willing to pony up for a better solution. How else can you explain video conferencing service Zoom\u2019s recent eye-watering valuation<\/a>, given that a great many of the customers undoubtedly already have access to Google Hangouts or similar for free? It\u2019s because the design of Zoom is so much better<\/a> than Hangouts, that it\u2019s worth it. Good design is good business<\/a>.<\/p>\n But still, \u201cgood design\u201d is pretty vague. How should designers make decisions when designing best of breed products specifically?<\/p>\n Any product designer worth their salt will start by understanding two things: their audience, and their competitors.<\/p>\n Let\u2019s assume that you have identified an audience of demanding, discerning users who value product quality and their own time. Let\u2019s also assume that your main competitor is some insanely huge, successful, resource-rich company like Google who are already deeply entrenched with a broad offering like GSuite.<\/p>\n At first it seems like an impossible David vs. Goliath proposition, but then you quickly realise some huge advantages you hold:<\/p>\n This is liberating. You can do things they can\u2019t. You can build product that\u2019s fantastic just for your niche. You can cover much less surface area, but go way deeper. The ceiling of complexity that might work for your customers is probably higher, so you can build more powerful products and expect your users to take the time to learn it, because your product will give them new superpowers.<\/p>\n Let\u2019s look at an example. Slack, Superhuman, Stripe, and Linear all use a similar pattern to allow for keyboard input. From anywhere in these apps, you can hit Cmd-K to invoke a modal input, into which you can type an array of commands. Here\u2019s what it looks like in Superhuman, the Gmail competitor:<\/p>\n From here you can type to act on the current mail (reply, archive, remind, move), navigate the app (compose, switch inbox, search), or take actions (copy recipients, open calendar, switch theme).<\/p>\n I love this UI pattern: it\u2019s fast, learnable, and makes you feel like a badass hacker from the future (an oft-overlooked feature). It\u2019s clearly a power user feature \u2013 you could never imagine mainstream Gmail users navigating like this \u2013 but once you get used to it, you can plough through email in seconds. It\u2019s random access navigation<\/a> applied to inbox triage. And it\u2019s good design.<\/p>\n Why? Because Superhuman have focused all of their efforts on people who want to be able to process large amounts of email quickly. People for whom a 20% increase in efficiency would be life-changing. For whom that gain represents $30 per month worth of value over the free thing they already use.<\/p>\n And so they go hard on it. They hammer home the speed point on their marketing site. The near-total reliance on keyboard navigation means that the screen UI can cut away a ton of cruft, making it faster again. Any time you attempt to use the mouse, the app almost admonishes you for doing so, teaching you the faster way.<\/p>\n This reasonable little message appears on screen whenever you click the \u201cMark as Done\u201d button:<\/p>\n But here\u2019s what you get when you simply select some text with your mouse:<\/p>\n And here\u2019s how Superhuman feels about you hovering a link like some kind of goddamn caveman:<\/p>\n It\u2019s even a little bit weird, but\u2026 aren\u2019t you glad that a product today is doing something weird at all? It\u2019s not for everyone, but it doesn\u2019t have to be. If, however, you\u2019re the type of person who loves this type of thing, well, you\u2019ll love it. It\u2019s picking a stance that\u2019s great for some, and it\u2019s sticking with it. That\u2019s what a best of breed product is.<\/p>\n Of course, Gmail has some keyboard shortcuts. But Gmail can\u2019t berate you for using your mouse. Best of breed apps are designed to help you get the most out of them. They don\u2019t have to dumb things down. Design decisions are tightly connected to the product strategy and business goals (e.g. our main selling point is speed, so we\u2019ll force our users to be as fast as possible, so that we actually deliver on our selling point).<\/p>\n You may be looking at this Cmd-K pattern and thinking it looks familiar. That\u2019s because, as with all design, there\u2019s plenty of prior art here. You might be reminded of Spotlight search (or Quicksilver, or Alfred). You might think that Superhuman vs. Gmail is essentially command-line interface<\/a> vs. graphical user interface<\/a>.<\/p>\n And you\u2019d be right. This is where being a student of the history of software can pay off. You don\u2019t even need to invent brand new interaction patterns to design best of breed products. Innovation doesn\u2019t necessarily imply invention. The application of an existing solution in a new context can be highly innovative.<\/p>\n “The Canon Cat was an early example of what we might now call a best of breed product”<\/p><\/blockquote>\n As I used Superhuman I was stuck by how much it reminded me of the work of legendary designer Jef Raskin<\/a>, in particular the Canon Cat computer<\/a>.<\/p>\n Although best known as the originator of the Macintosh project<\/a> at Apple, the clearest realization of Raskin\u2019s vision for an \u201cinformation appliance\u201d arrived in the form of 1987\u2019s Canon Cat. (Raskin later captured his underlying ideas in his classic book, The Humane Interface<\/a><\/em>.)<\/p>\n The Canon Cat was an early example of what we might now call a best of breed product. Perhaps the first in tech. It was a response to an existing successful mass-market incumbent that was enjoying broad appeal.<\/p>\n The Macintosh had succeeded by billing itself as \u201cthe computer for the rest of us\u201d and introduced the masses to simple, accessible concepts like a GUI and a mouse. While wonderful for most novice users, Raskin was frustrated by what he saw as the trading off of power in service of broad accessibility.<\/p>\n Raskin saw the opportunity to make a computer for a smaller niche audience, \u201cparticularly office people,\u201d who wanted to \u201cmake mountains of work disappear faster and easier than ever before.\u201d Sound familiar? By identifying a specific audience and selling point, he was able to be radically opinionated in creating a product that was far, far better for that core set of users.<\/p>\n As we\u2019ve seen, best of breed products can be better by doing less. The Canon Cat did away with the mouse, an aggressive move in favour of the faster input afforded by keyboard input (just like Superhuman). It did away with the desktop metaphor and filesystem, booting straight into a single multi-purpose document that could perform word processing and spreadsheet-like calculations inline (just like Coda).<\/p>\n It even did away with cursor keys. But as we\u2019ve already seen, best of breed products also add things to make the experience better. In particular, it introduced two new keys:<\/p>\n Photograph: Marcin Wichary<\/a><\/em><\/span><\/p>\n These \u2018Leap\u2019 keys don\u2019t just perform basic cursor movement. Holding one of these modifier keys while typing will immediately shift input to the nearest instance of the typed letters \u2013 much faster than using cursor keys of a mouse. Again: kind of weird. But good weird! If you\u2019re curious, the Internet Archive have an in-browser emulator of the Canon Cat<\/a> you can play with.<\/p>\n![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-Superhuman1-1.png\")
\nSuperhuman, a Gmail competitor <\/em><\/span><\/p>\nBest of breed products<\/h2>\n
Designing best of breed products<\/h2>\n
\n
Cmd-K<\/h2>\n
![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-Superhuman-2.gif\")
<\/p>\n![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-Superhuman4.png\")
<\/p>\n![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-Superhuman2.png\")
<\/p>\n![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-Superhuman3.png\")
<\/p>\nEverything old is new again<\/h2>\n
![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-Mac.png\")
<\/p>\n![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-CanonCat.png\")
<\/p>\n![](\"https:\/\/www.intercom.com\/blog\/wp-content\/uploads\/2019\/07\/CmdK-CanonCatKeyboard.png\")
<\/p>\n