Why Google Killed Your Favorite Feature

Jon Wiley, head of Google Search, explains why Google sometimes kills your favorite features.

We’ve all had a favorite Google product or feature that just disappeared one day. (For me, Google Reader comes to mind immediately.) And when it happens, it can seem senseless and strange. Why would a company as large, profitable, and known for supporting some strange projects ever throw out a tool that people liked?

In a Reddit AMA (Ask Me Anything) yesterday, Jon Wiley, Lead Designer on Google Search, explained why too many small features can not just bloat the UI of a product, but actually compound into major problems for a company.

One thing that’s almost always guaranteed with product design: when you add a feature, no one complains about it outright; if they don’t love it they mostly just ignore it. Whereas if you take something away, you’ll hear about it if people relied upon it . . . loudly and often. With something like Google Search, even if just a small fraction of people miss a feature and an even smaller fraction says so, that can still be tens of thousands of people. It can seem like a tidal wave of opposition to the removal: "look at all these people who want it back!"

So it would be much easier to leave in everything that’s ever launched. But then you end up with bloatware: an unwieldy array of ill-fitting modules that don’t work well with newer technologies (e.g., the shift to smartphones, or upgraded security, or touch screens, etc.) and don’t really serve most of your users well either. And nothing comes for free—every feature must be maintained, supported in multiple languages, on multiple devices, and the additional complexity must be accounted for in testing so that the entire service remains reliable. And that cost gets balanced against the impact: is this feature solving an important problem for lots of people?

There are many, many such features that you always have to make tough choices about. We’ve actually cut features that I love. This is one of the toughest but most important parts of designing products—deciding what to trim as you move forward. Sometimes you over-trim—we work to measure the impact and aim to strike the right balance. Sometimes we get it wrong, so it is important that people speak up. We really do listen, and we prioritize according to what seems to satisfy the widest needs given our capabilities.

The rest of Wiley’s AMA is worth reading, too. Our favorite piece of gossip that he revealed? Google is working on a voice search feature for bilingual people that can automatically detect which language you're searching in without having to change settings beforehand, giving everything a more natural flow.

Read more here.

[Image: Google search via George Dolgikh / Shutterstock]

Add New Comment

9 Comments

  • I wonder if there another way to deal with some of these features and products that can demote or divide them off. I understand the needs for a clean, UI, focus on the highest priorities, etc. However If thousands to millions of people are using them and they've been created what other options are there? It's fun to think about, because clearly there is some value there and some unmet needs on both sides. Open sourcing is one option, archiving but not maintaining is another but what if something could meet the needs of both users and Google in a new way? What if that feature could be demoted and handed off to a GreaseMonkey script or some sort of other catchall Plug-In library? What if the product was spun off to an employee that was leaving anyway with a contract option to reacquire?

  • I wonder if there another way to deal with some of these features and products that can demote or divide them off. I understand the needs for a clean, UI, focus on the highest priorities, etc. However If thousands to millions of people are using them and they've been created what other options are there? It's fun to think about, because clearly there is some value there and some unmet needs on both sides. Open sourcing is one option, archiving but not maintaining is another but what if something could meet the needs of both users and Google in a new way? What if that feature could be demoted and handed off to a GreaseMonkey script or some sort of other catchall Plug-In library? What if the product was spun off to an employee that was leaving anyway with a contract option to reacquire?

  • adamjameslaw

    They killed thumbnails in Google Patents but there were many many complaints. They brought it back.

  • The problem then is, when a new feature is added, how on Earth can you decide which feature to become reliant on when there's no way to know which features will be popular enough to maintain? You're at the seemingly-capricious whim of a large company that tells us, flat out, that the opinions of tens of thousands of people don't ultimately count in the grand scheme of things.

    So you'd just better pray that those features you use every day are popular, or you're in for trouble.

  • Todd Cliff

    Wait, did they take away the search feature? Isn't that what Google does best?

  • denfam4

    The opportunity then is for Google to communicate WHY a product and/or feature was cancelled. Talk to us Google...after all in the end it's about the overall experience.