This makes sense, especially given the context of Material Design. As each result has gotten larger (read: more information included, like ratings), they've also started to blend together. These "cards" can be used to help the user separate out each result. I also wouldn't be surprised to see them test an infinite scroll.
So many conventions were established by search early on and it's served Google very well to stick to their simplistic roots. But I can tell that they're starting to see opportunities to enhance the experience and break a few of those conventions. Great to see them doing this through tests, rather than just rolling it out. User-Centered Design has become a big focus for them.
That being said, adding a border doubles spacing. Adding a gap, triples spacing.
Tripling spacing means less results on a screen.
So count me out.
But why do we need compressed search results? Honest question, really. It's not like anyone's going over the first page.
More spacious results are worth the cost if they enable cards with richer information. Historically, relevance has been confined to the first ~6 results, so I'd rather see more about them than seeing more results.
This exactly. Google have become maniacally focused on relevance and quality over noise and quantity. This is especially evident in the UX and quality focused changes that have come with the Panda, Rank Brain, and Mobile-Friendly algorithm updates.
The overwhelming majority of clicks occur within the first 3 results, and as relevance increases, so do those clickthrough numbers. Google doesn't necessarily want to serve you more results. They want to serve you the right result and the best experience.
Guys, no offence, but I'm not going to explain why I don't want less search results on a single screen instead of "a card-like look", not to a product designer and a senior ux designer anyway.
You should be able to figure that request out on your own.
No offence taken. I don't have access to any of Google's data or research (and neither do you), so this is all speculation, which is essentially meaningless. The one thing that I can say is that as a designer, I've found that my personal preferences for a design can sometimes be very different from my audience's preferences. There is a chance that either you or I could be running into that right now.
Hence the fact that Google is running this as a test, rather than just rolling it out.
That said, I'm actually not particularly attached to either design. Rather, I was just recognizing a few patterns:
- I can appreciate the user-centered and experimental approach that Google is taking
- I can see how this could be a natural part of Material Design's evolution within the company
- I can understand why they may be hypothesizing that "more results" may not always equal "better"
But as I said earlier, I don't have any of the context or data. This is why I can both understand why you may want the design to stay the same, but also not be shocked at the tests they're running and shout "But there should always be more results! It's a UX rule!" Over time, you learn that once you go beyond the basic principles of Cognitive Psychology, there are no rules in design. It's all contextual. And your personal preferences or conventions that you've learned in the past won't always hold up.
You shouldn't have been surprised that your design convention would be questioned by a Product Designer and a UX Designer. That is quite literally what we do.
Fits with their material design guidelines, but I'm in agreement with previous comments about it feeling more difficult to quickly scan results.
Having used it, I don't like it. I could feel myself not being able to scan the results as quickly as I would normally. Google results might not look the prettiest, but they work really well.
A more minimalistic design available at google.ro: http://imgur.com/v50vaR2
Posted this three weeks ago... https://www.designernews.co/stories/69332-google-seem-to-be-ab-testing-a-new-search-results-card-layout but yeah definitely seems to be the thing to do with Material design. It's also easier to differentiate between results so an improvement in my opinion