The iPhone Announcement: Don't Redesign Yet, But Do Pay Attention

On January 9, at Macworld in San Francisco, Steve Jobs announced a phone. You have heard about it by now. What matters for anyone who runs a website is one claim from the keynote: the iPhone runs Safari, a real browser, on a three-and-a-half-inch touch screen.

Not WAP decks. Not the stripped-down ‘mobile internet’ that has let down everyone who tried it on a flip phone. The actual web, pinched and zoomed by fingertips. The best mobile strategy in 2007 is a fast site with honest HTML. That works on a Treo today and whatever ships next.

The same question arrived from every direction within the week: do we need a mobile site now? The honest answer is mostly no, not yet. The reasoning is worth walking through, because the same logic will apply to the next exciting device too.

What we know versus what we felt during the keynote

Keynotes are built to make announcements feel like reality. Let’s restore the distance. The device ships in June, so there are five months between now and anyone browsing anything. Pricing starts at 499 dollars with a two-year contract on a single carrier. That is early-adopter territory, not mass market.

Nobody knows what adoption looks like yet. Apple’s stated goal is ten million units by the end of 2008, which is about one percent of the phone market. Ambitious for Apple. A rounding error against your visitor logs.

Building for a single device has burned people before. Some of us remember the money spent on WAP portals around 2000, mobile experiences for a mobile web that never arrived as promised. The lesson was not that mobile does not matter. It was that betting a budget on one vendor’s vision, before users vote, is a gamble dressed up as a strategy.

What the announcement really changes

Trajectory. Until January 9, browsing on a phone was a compromise everyone tolerated and nobody chose. The iPhone’s premise is that people will do real browsing on a phone if the browser is real. That resets what the industry thinks is possible, and competitors were taking notes before the keynote ended. Whether this device wins hardly matters. The bar for mobile browsers just moved for good.

Two details from the demo deserve attention. First: no Flash. The browser is Safari, the plugin is absent, and Apple has given no sign that will change. If your site’s navigation or media runs on Flash, a growing class of premium devices will render it as empty space. We have argued against Flash foundations before. The iPhone adds a blunt commercial reason.

Second: touch. Fingers are not cursors. They are less precise and they cannot hover. Interfaces built around tiny click targets and hover menus, which describes most navigation built this decade, will meet this input model badly.

The preparation that costs nothing

Here is the position we are taking with clients, and it has the nice property of being free. The right iPhone prep is the same prep we would prescribe with no iPhone in sight.

• Standards-based markup. Safari on iPhone uses WebKit, the same engine as desktop Safari, and it handles clean, semantic sites well. Standards-built sites are already iPhone-compatible. Sites built on hacks and Flash are already not. June is five months of runway to close that gap for reasons that were valid anyway.

• Light pages. The device runs on AT&T’s EDGE network, which is not broadband. Heavy pages that merely annoy desktop users will truly punish phone users. Page weight discipline is about to get a fashionable new reason.

• Layouts that don’t assume 1024 pixels. Not mobile-specific design. Just the absence of rigid width assumptions. Flexible structures cost nothing extra during a normal build and age better against every screen size.

When the math changes

There will be a moment when device-specific work makes sense for some clients. It arrives when your own analytics show real traffic from capable mobile browsers, and when your audience’s tasks fit the pocket: directions, schedules, quick lookups.

That moment is measurable, so it does not have to be guessed. We are adding mobile-browser segments to the analytics reviews we run. For a university whose applicants skew young, the line may move fast. For a manufacturer whose buyers browse from office desks, it may crawl for years. Both answers are fine. Both are decisions with data under them.

What we would avoid in the meantime: the vendor pitches already going around for ‘iPhone-optimized’ microsites, five months before an iPhone exists in anyone’s pocket. Enthusiasm is not a requirements document.

Three questions for your analytics between now and June

What share of visits already comes from anything identifying as a phone? Establish the baseline before the excitement. Which of your tasks would someone do from a pocket? That is your future mobile priority list, assembled free. Does anything on those task paths depend on Flash or hover? That is your exposure list.

Twenty minutes of digging now buys you an evidence-based June instead of an anecdote-based one.

Our bet, in writing

We think this announcement is the important kind. Not because ten million units would reorder anyone’s traffic. They would not. But the premise will prove out: people want the real web everywhere, and January 9 was the day the industry stopped pretending otherwise.

We expect several credible pocket browsers within a couple of years, from several vendors, at several prices. The sites that do well across all of them will be the ones that never bet on any single one.

Build for the web, not the gadget. It is the cheapest advice we give, and June will start telling us whether it was also right. We have already decided the agency is buying one. Strictly for research, mostly.