Gmail at Nine: The Evolution of an Essential Web Service

In 2004, when Google introduced Gmail, it wasn’t instantly obvious to everyone why the search engine phenom would want to get into the e-mail business. Actually, given that the company made its announcement on April 1, some people wondered, momentarily, whether the news was a hoax. Even once it was clear that Gmail was the real deal, Google felt the need, in the Gmail FAQ, to answer the burning question “Why is Google offering email? I thought you were a search company.”

The era of people being confused by Gmail was so brief that it’s easy to forget such an era ever existed. Gmail quickly became not just one of Google’s core services but one of the web’s core services — the webmail client that came to define webmail clients even though it was far from the first one. Like most things Google does, Gmail hasn’t been without its controversies, but it’s still one of the company’s great success stories. I recently chatted with Alex Gawley, the service’s product manager, about Gmail’s first nine years — and a little bit about where it might be going.

First, to refresh your memory, here’s an infographic-based history of Gmail that Google distributed last month, complete with major milestones such as 2010’s Priority Inbox, probably Gmail’s biggest innovation since the ones that got it started:

One of the reasons Gmail took the web by surprise was that it wasn’t necessarily obvious that web-based mail was ripe for reinvention. Nine years ago, Hotmail and Yahoo Mail were good, popular and deeply entrenched; they seemed to have the market as it existed all sewn up. Gmail shattered that assumption because it was the first major web-based e-mail that aspired to be good enough to be someone’s primary inbox.

“From the very start, [Gmail] was about pushing the boundaries of what an e-mail product might be for people,” Gawley says. “Back in 2004, most webmail providers were giving people a few megabytes of storage. They were thought of as this interesting quirk. A gigabyte of storage was pretty much 200 times what everyone else had.” (Gawley is underestimating: Gmail’s original, groundbreaking 1GB — upgraded over the years to today’s 15GB, shared with other Google services — was 500 times Hotmail’s 2MB and 250 times Yahoo Mail’s 4MB.)

As you’d expect of an e-mail service from Google, Gmail also had far better search than any competitor of the time. Its anti-spam measures were unusually effective. And while it didn’t originate the idea of organizing messages in threaded conversations rather than as discrete messages, it helped popularize it.

Another Gmail innovation was the way Google made money from it: by scanning the text of messages for keywords and then displaying tiny ads related to those words, a twist on the ads which accompany its search-engine results. The approach was startling and controversial at first, but the kerfuffle was short-lived; only recently has it become much of a subject of conversation again, mostly because Microsoft keeps bringing it up and accusing Google of invading Gmail users’ privacy. (Outlook.com, the successor to Hotmail, doesn’t use keyword-based advertising.)

I asked Gawley about Microsoft‘s contention that consumers should be bothered by Gmail’s ad technology. “Ads across all Google services help keep them free,” he says. “Generally, users understand that we’ve been very clear in the way we’ve talked to users about what our systems do in order to show useful, relevant ads. We use the same sort of technology we use to keep spam out of your inbox and provide Priority Inbox.”

Which is not to say that Gmail users are understanding about everything. As a group, users of web services — or at least a noisy minority of users of web services — don’t deal well with change. And as one of the planet’s most widely used, essential web services, Gmail is probably second only to Facebook when it comes to examples of users going ballistic over interface adjustments. The most striking example: Google Buzz, Google’s short-lived attempt to turn Gmail into a Twitter-like social network, which prompted an outpouring of concern over privacy issues before withering away and eventually being superseded by Google+.

Me, I liked the lighter, airier version of Gmail which Google began rolling out in 2011…

…but some pretty smart people pined for the maximum information density of its more cramped predecessor, and continue to do so. And I continue to hear gripes about the new, smaller compose window Google first previewed last October and made permanent and universal in March.

So how does Google deal with the fact that it’s likely impossible to alter Gmail in meaningful ways without ticking off a measurable percentage of users?

“This is a tool people use a lot, they’re pretty used to it,” Gawley says. “Even with features that people become very excited about, [change] can still be a jarring experience.” And since Gmail is a productivity tool, folks can be sensitive about changes arriving when they’re trying to get stuff done.

“Sometimes, ‘right now is not a good time,'” he says. That’s one reason why significant changes to Gmail’s look and feel usually arrive first as optional previews, allowing users to put off the transition. The previews also allow Google to gauge feedback on changes, and it sometimes tweaks further based on what it learns.

Gmail controversies tend to involve the desktop-browser version of the service — its original version, the one with the most features and the one which gets updated at the fastest clip. But today, that’s only one variant of Gmail. The service was mobile from early on: Google launched a version for phone browsers in 2005 and a Java app, compatible with a wide variety of the era’s phones, in 2006. Today, in the age of iPhones, iPads and Android gizmos, mobile Gmail is feeling less like a fundamentally secondary way to use the service and more like the one that could edge out the classic desktop browser one over time.

Actually, though, Google doesn’t think of Gmail for phones and tablets as one thing. “There’s mobile devices and there’s tablets, and we tend to think about them together as an industry because of the commonality of operating systems,” Gawley says. But phone Gmail needs to be designed for on-the-go use on a tiny screen; tablet Gmail has more screen real estate and less of a need for ultra-mobility.

I mentioned to Gawley that in all its phone and tablet forms, Gmail still isn’t anywhere near as feature-rich as it is in its traditional desktop incarnation. I wondered: Is that a statement by Google that these newfangled versions of Gmail need to err on the side of simplicity?

“We’re still very much in active development on those platforms,” he says. “The pace of change on Android Gmail and iOS Gmail will accelerate as we build more and more of the features people need going forward. But at the same time, we want to be really thoughtful.” (If I were in charge, I’d decree that desktop Gmail’s utterly indispensable Send and Archive button — which spend almost four years as a Gmail Labs experiment before becoming an official desktop Gmail feature — would be added to every version of Gmail immediately.)

Even on the desktop, making Gmail both powerful and simple involves an endless parade of judgement calls, in part because there’s no such thing as a typical user. “There is a very broad spectrum of people using Gmail,” says Gawley. And “even with individual users, their use is also a spread across that spectrum.” In many cases, Gmail aims to strike a balance by offering advanced features but keeping them out of the way unless you need them.

For instance, “even the most powerful users send short messages with no formatting,” he explains, which is why the new compose window hides most formatting options when you open it. But Google’s research showed that 50% of messages with one bit of formatting included multiple pieces of formatting; that’s why the formatting toolbar stays open once you’ve used it.

(Occasionally, I’ve been mystified by Google’s decisions about which features mattered: for a while, Gmail suggested people you might want to add as recipients. For me, at least, it was useless clutter — I knew who I wanted to send messages to, and the suggestions were never among them — but there was no way to shut the feature off. One reason I’m a fan of the new simplified compose window is because it eliminated this “feature.”)

As I wound up my conversation with Gawley, I asked him about the future of e-mail — or, more specifically, the possibility that e-mail’s future might be less eventful than its past. For years now, surveys have shown that young people are less dependent on e-mail than their parents, older siblings and others who lived through the medium’s glory days. Gmail, like all e-mail, faces competition from everything from text messaging to video calls to Snapchat.

Gawley rejects the theory that e-mail is on the wane: “People are using e-mail more and more and more. I don’t subscribe to the theory that it’s becoming less important.” At the same time, he says, “We have to think about e-mail differently than we did nine years ago — it’s part of an overall electronic communication mix.” He points to Gmail’s integration of Google+’s Hangouts feature as an example: it lets you launch a ten-person video chat from your inbox.

Thinking about the future of Gmail, I told Gawley that Gmail Labs, the service’s collection of experimental, optional features seemed pretty quiet lately; no longer did Google seem to implement every feature that popped into anybody’s head, such as Mail Googles, which let you choose to have Gmail force you to do math problems to prevent you from sending late-night drunken e-mail.

He acknowledged that Gmail Labs is, indeed, less of an exercise in stream-of-consciousness engineering than it once was. “Across Google, we recognize that we should be investing in experiments that will have the biggest possible impact. At the same time, we want to make sure that we don’t lose the change that comes out of something that seems small when you start, but grows to be something world-changing.”

I’m not sure what Gmail could do in 2013 or beyond that could be as world-changing as the service was in 2004 — but it would be fun if Google came up with something, and it’s good to know it’s trying.