Yahoo showed us 30 days of logos. Here’s the one consumers liked best | Hacker News Reader

Yahoo showed us 30 days of logos. Here’s the one consumers liked best

survata.com

193 pointsawenger98 comments

I think the weirdest thing about Yahoo's logo is the color. It wasn't until I started working for Yahoo that I first heard that the corporate colors were purple and yellow[1], later changed to purple and white.

The logo on Yahoo.com was -- from birth until around 2008 or so -- red. That was the color everybody outside of Yahoo knew. But inside the company everything that could be a color was purple, people were emotionally invested in it, and debate raged on getting Jerry and Filo (who were apparently the holdouts on red) to change the home page logo to purple.

It seems to me very emblematic of Yahoo that they chose their corporate color first entirely at random and then hung on to it forever out of sentimentality. That's the kind of place Yahoo is: fuzzy and loving but not terribly sensible, business-wise.

Changing the logo is a cliché for a struggling company that doesn't have a lot of new ideas. I actually don't think that's true of Marissa Mayer; a whole bunch of her ideas are along the lines of "do whatever Google did", but then Google is very successful, and at least part of what Yahoo needs to do is get better at the things that Google is already good at. But the acquisition-frenzy in mobile seems smarter, and the acquisition of Tumblr smarter still. Time will tell whether it's enough to turn the ship around.

But I still think it's weird that the corporate color is purple.

[1] Chosen essentially at random because they were the cheapest paints available when they painted their first offices. Purple and yellow do not go together at ALL, incidentally.

Purple and yellow do so go together!

http://www.bhg.com/gardening/design/color/yellow-purple-flow...

this and LA Lakers

Pretty sure Datadog out-purples y'all http://datadog.com

As accent colors, sure. But in an office setting as the primary colors you'll be looking at all day? Not so pretty.

Also see the Minnesota Vikings and the LSU Tigers colors.

As someone living in minnesota, I don't see how the vikings are a vote for purple and yellow. :D

Pretty fancy:

http://www.imdb.com/media/rm3892224512/tt0067992

They are almost opposites or complementary (the complement of yellow is actually violet, not purple, but purple is close enough), so just like blue and orange.

Complementary colors used in combination give the highest contrast.

Having trouble finding the link, but I remember an article on why the founder of DailyKos settled on burnt orange. The rationale was there were already lots of shades of blue and other pleasant colors on the Internet. So while burnt orange was pretty ugly, it was at least unique and memorable. And for a blog trying to get repeat visitors, a memorable look was important.

Granted, uniqueness or memorability is no longer necessary for a company as well known as Yahoo, but there's something to be said for ugly.

My point was not so much that the color is bad, so much that the process by which the color was selected -- ignoring customer familiarity in favor of something that made employees happy -- was very characteristic of Yahoo.

I wonder if the next thing Yahoo! does is build a huge office complex. That's the thing big silicon valley companies do before they die, although Yahoo! should have done it in 2005-2007 on this timeline.

It was being discussed in 2009; http://www.businessinsider.com/yahoos-humongous-new-hq-see-i... Not sure if there's any recent news on it though. It's pretty close to the new "SF" 49'ers stadium so it was probably a good real estate move on that front.

Isn't having a strong corporate attachment to any color the weird part about all of this?

It's the exclamation mark that's always bothered me for some reason. Would make contracts and proposals look silly. "We will develop X for Yahoo!"

Dropping the exclamation mark would only improve their logo.

Purple and yellow actually go together quite well.

Purple and Yellow are complementary colors.

"When placed next to each other, they create the strongest contrast and reinforce each other. They are widely used in art and design."[1]

http://en.wikipedia.org/wiki/Complementary_colors

I agree. Yahoo! logo is red. That's it. This change is equal to Facebook changing the corporate color to green.

I think Paul Rand would turn in his grave at the thought of "Thinking of a new logo? Don’t guess; test." There is a lot more to logo design than having a first impression test well. I really do hope they hired someone good to do this.

I guess we'll see the "real" version tomorrow.

Edit: Let me say it another way. If the real logo is just #31 in a series, then they will have screwed up big time. The buzz about the redesign will just be "I liked #N better". I'm hoping there is more to the redesign than just an exploration of the typeface choice. If that's it, they shouldn't have let us see 30 variations. Paul Rand just used to present his one choice.

The winner seems very similar to the old logo, sans serifs.

I wonder if the proper conclusion is that this is the best new logo, or that this is simply the most familiar.

I was thinking the same thing. They should have included the old logo in the test also.

Perhaps the best is the most familiar because when people see Yahoo they expect and want to see Yahoo.

Not only that, but in the breakdowns by attribute, users actually preferred serif logos in general over non-serifs, which suggests that they would actually have preferred the current logo over any of the new ones.

Perhaps familiarity is in fact part of the definition of the "best "logo. After all, a logo serves no other purpose than recognition. Companies routinely spend a great deal of money ensuring that their logo, colors, interior decor, etc etc, all make people go "Oh yeah! That's [X]!"

Precisely. What if they had shown the logos in different colors, with a different word, with different styles?

In which case the question would be the wrong one to ask. It doesn't really matter which logo people "prefer" - their personal aesthetic taste has nothing to do with an effective logo (well, assuming your logo isn't goatse).

It would be more effective to ask people what descriptors they'd use to describe the company behind the logo, using a fictional company name.

The rest is millions of dollars of marketing.

True, but there is also potential for some re-branding here. Maybe one of them is more "hip" and "fun" but people don't associate it with the Yahoo! brand.

Ideally you would have different people use different logos and then you'd survey them after a year or 6 months; once they're more used to "their" new logo.

> After all, a logo serves no other purpose than recognition.

If recognition was the sole purpose, Yahoo should have kept the logo exactly the same. I think logos (or styles in general) are also an expression of the attributes that a company wants consumers to associate with the brand.

The uneven baseline in Yahoo's logo express playfulness, as do the colors in Google's logo, for example. That's how Google and Yahoo contrast themselves with boring old business-oriented companies like Microsoft, IBM and SAP.

Like potatolicious pointed out, it is dubious how worthwhile it is to ask your existing customers for feedback if you want to change your image. After all, these are people that are already familiar with you, while the purpose of an updated image is, presumably, to attract a different/wider userbase.

Logos in fact have two purposes: Recognition, and so C-level empty suits can feel like they're accomplishing something by changing them.

Yeah see the "failed" Pepsi logo redesign[1]

[1] http://blog.logodesignguru.com/wp-content/uploads/2011/12/Pe...

A great victory has been won this day in the name of sans serif fonts.

You read my mind.

It's the most similar to the old logo, IMO.

It would have been interesting if they would have included the original logo in the test. :)

It's because the old logo was properly created by some ad agency to pass the idea of the "Yahooooo" sound.

Hence the best logo chosen by the people (which i think is not the official word by yahoo inc) also has that "feeling".

All the logos with a fixed baseline, or jagged baseline that goes against how western read the "yahoooo" sound, just feel like you changed the font in MS Word and typed Yahoo!. Which is all most of those are...

I think the overall idea of 30 logos, one per day, is terrible. It cheapens the brand, and the seriousness of the process of creating the focal point of a multi-billion dollar company's branding for the next decade and beyond. I understand they want to underline that they're making changes. But people want to be told a compelling story, not asked which of 30 stories they prefer.

This feels like a huge A/B test, and -- even if not true -- seems like an experiment in releasing the tiller to see where the wind takes the boat. They should make a strategic design decision and execute it consistently -- have a strong opinion. Given the paradox of choice, the fact that people ultimately prefer a logo similar to the current one isn't surprising.

They should make a strategic design decision and execute it consistently -- have a strong opinion.

Amen! Generally this disturbs me about a lot of web stuff (probably because it is the one area in which I'm exposed to it) - caring more about what people expect or want, rather what you wish they would want, because you like it yourself. The winning logo is so ugly, I'd use it as an glaring example why you can't poll the masses on style or taste, not as a guideline. Not that I think it matters that much; the world is full of ugly logos, but as long as they stand for something want/need, we tend to accept or even like them. But still. Yech.

The board of logos looks like something you'd expect if you went to 99designs and posted: "Billion dollar company needs new logo; winner gets $125! Must use one color, be tiny as hell, wording only, no art. Winner will be selected after 30 days of voting by end users. Please, no imagination; we do not want to scare Grandma away from her 'google'."

Its incredible how we think popularity means correct. The "best" logo via this methodology is very similar to the current one. Yahoo has this "old" and "cookiecutter" image that's bad enough. This just makes it worse. I can't imagine real designers thinking these designs and this popularity contest mentality is remotely good. Seems like something a ma and pa company would do because they didn't know better.

Also, try a color other than purple. No one likes purple.

I like purple.

Objectively I think rainbows are worse than purple, but it works for Google.

I think the purple really sets them aside. HN is orange, Reddit is light blue, Facebook is darker blue, Yahoo is purple. shrug

Most of their submissions looked like they were done in the same afternoon using a shitty torrent of 'hot fonts' from the pirate bay.

When we talk about updating a brand, one of the most recent examples of solid execution was mail chimp.

http://jessicahische.is/monkeyingaround

I was thinking that Yahoo should hire Jessica Hische for the redesign. I think they'll go with a typographic logo, and who could be better? She could do a great script Yahoo.

Honestly, it would be a great choice - - I mean, if they didn't hire me for it :)

The whole mood and perception of Yahoo needs a shift. In a sense they should prepare a few key properties and release a new brand, design and suite all at once instead incrementally like they have been.

So sad. So instead of hiring anyone, the CEO did it herself (with some internal designers). The $0 logo.

Not too surprising it won't do much in terms of rebranding.

I cannot un-see "Mail Chinys" anytime I look at that logo.

It's important to note that this article and survey were written and executed by Survata, under no guidance from Yahoo, and that the results are not the "Yahoo sanctioned" winner of all the logos. I think at initial glance this isn't obvious.

OP here. That's correct. This is work of Survata not Yahoo. I see how the title might suggest a connection, but that was not our intent.

Right. And they only tested 5 logos picked at random from the first 28 that Yahoo presented. They may not even have picked the "best" 5 out of that lot. So really... this whole blog post was click bait. I'd never heard of Survata before... but now I have... for better or worse.

The way I understand it, each user surveyed got five random logos, but not the same ones. Over the entire population, all the logos were surveyed and the "winningest" ones are presented as the highest rank (see the last chart with the full matrix of which logos won against which others).

Ah. They could have worded that a little different then. It did sound as though they picked 5 logos and then surveyed the people. I can see it the other way now as well. I've never known a survey to be run like that. Seems odd to me.

This still generates an issue with sample size - since today was the 29th day, that logo had from days start through time of writing/submitting the article, while the logo from day 1 had a much larger sample size. There is some evidence in the results. Secondly, was it fully randomized, or were things like multi-armed bandit or other a/b testing methods used to provide enough variation. The chart is biased towards earlier numbers, with the outliers of the really bad logos being dispersed towards the end.

The data also comes from those highly annoying survey walls, so take with a grain of salt. As in, "before we let you view the content you want, complete this survey."

Personally, I find these rather blogspammy.

As a designer, this whole process is degrading, frustrating and reminiscent of the worst clients I've had throughout my career. I'm not sure if they are trying to make it look like they want to appease the people, but if I were working in-house over there, watching all of these awful designs roll out would just crush me and make me question what they think my worth is. I'm hoping the whole thing turns out to be a joke.

There is possibly more than just design at stake here. It can be a really difficult technical and cultural problem to get such a large website to the point where they are capable of changing their site every day - and ostensibly compare the impact of each change on their metrics.

It can also be difficult for a company so firmly rooted in one brand identity to make a 'big bold change'. Forcing this type of churn onto the company could make the organization more comfortable with change, and could lead to an actual 'big bold change'.

I don't know if any of this will happen, but large organization often have a lot more to balance than just having one designer come up with what they think is right.

I was at Yahoo! for the last logo change (from web safe red to purple), and that I was seeing the correct logo every day across the network is pretty impressive.

It's not really about the design though. They (Yahoo) are trying to connect with people and get them engaged.

This is similar to the "design a bracket for a jet engine in 3D" campaign that GE ran - that resulted in absolutely horrible contest entries from people who do not understand materials/stress/strain/vibration... but looking visually appealing to a non-techie. I am sure GE had no intention of even considering any of those "designs" for anything other than pure entertainment.

Right, but I think that most people can see right through the "look guys, we're cool! we care!" crap when it's this tacky and confusing. Quite frankly I thought they would be way more outlandish and unusable in the context of being a logo simply because of the name and the whimsicality of the current iteration. Downloading a bunch of type from Dafont does not a good logo or even fake contest make.

And here's the final they posted - http://yahoo.tumblr.com/post/60332693287/introducing-our-new... - along with a now laughable quote about it from the post 30 days ago:

> The new logo will be a modern redesign that’s more reflective of our reimagined design and new experiences.

Isn't the "winner" the current crumby logo? I would suspect that beyond any kind of validity questions from the surveyed cohort that there is a significant bias built in. Humans are kind of silly that they will always choose that which is familiar over that which is not, no matter how stupid or silly the familiar is. People have to heavily be told what they want, because people also don't like thinking. It is the very core of our society in particular, but also generally so.

>Humans are kind of silly that they will always choose that which is familiar over that which is not, no matter how stupid or silly the familiar is.

through the human (and well before the human) evolution the ones who prefer unfamiliar (to try to eat, to touch, to waddle in, to make friends with, ...) have statistically been leaving lesser number of descendants than ones who prefer familiar, and thus we're here - the result of that statistical process

> because people also don't like thinking.

instinctive or very well learned/automated behaviour (i.e. familiar situation/stimuli, trained response) is much cheaper for the brain energy-wise, thus it will be preferred by selection until the situation changes to where thinking presents better price/performance ratio. And then new cycle of bottom-line optimization of pushing newly thought out behaviour in that new situation down to the level of well trained/instinctive response and back to square 1.

it's a lot like the current logo but without serifs.

Is it just me, or were none of these logos inspiring in any way? The winner logo even seems to have a glitch (too much space between "Y" and "A").

They only tested variations on type face and layout. They didn't offer any completely different logo options that split from the familiar.

Is it just me, or is this winner logo the one most similar to the original logo?

It's not just you. That's actually the whole point of the article. Did you even RTFA?:

"Our findings from Yahoo’s “30 days of change” campaign suggest it should stick to something familiar. So, for Yahoo’s sake, we hope the new logo announced tomorrow will remind users of its graphical heritage."

In which case, they really should just stick with the original. If they really want to change their imaging a bit, they need to go hire Hardgay.

Hardgay goes to Yahoo!: https://www.youtube.com/watch?v=KSD8edviajE

Seriously, I might use Yahoo! for something or another for the first time in fifteen years, if they lead their imaging campaign with him.

Here is the complete list:http://www.yahoo.com/dailylogo

While I think they were all slightly boring, they were quite varied in design. I like Day 2.

Also, how is that spacing a glitch? Maybe it is by design? Since the whole point of this is to play with typeface, letter spacing and other creative ideas.

I also like the clarity, readability, and simplicity of 18 and 28. Most of the other options are either hard to read, or just seem like a jumble of shapes at first glance. But yes, I like the mostly boring options.

The only one I remotely liked was Day 11. I hate the old one and think they need a completely fresh start that signifies a new era.

I think a huge problem in today's big-data mentality is that few have the guts to make a bold decision based simply on good taste. Everything gets over-analyzed to ad nauseum, much like this ridiculous choice for a new logo. This process that Yahoo went through for the logo is death by committee x 1000. And the irony is that people preferred the original logo the best, or didn't care between the two logos that looked most similar to the original logo.

This reminds me of Komar and Melamid's Most Wanted Painting series, where they surveyed[1] residents of various countries to find aesthetic preferences, and then made the paintings[2] that should appeal and repulse those residents the most.

[1] http://awp.diaart.org/km/surveyresults.html [2] http://awp.diaart.org/km/painting.html

"...for Yahoo’s sake, we hope the new logo announced tomorrow will remind users of its graphical heritage."

That the logo should "remind" people of the old one seems correct, but too close to the old one would be a mistake.

These days, it seems like Yahoo management is caring more and more about its image to non-users. For that purpose, they might be better off keeping some of the testably preferred characteristics, but taking a sharp right on others to rewire the brand.

Same great Yahoo, new chocolate filling.

Isn't this just subtle confirmation bias? The favorite is basically the old logo with some things pushed around. Show someone 30 logos , they will subconsciously like the one that triggers a familiar pattern.

I think Peter Griffin said it best. Oh my god, who the hell cares?

Everyone, without even knowing it.

That's a really good take away for all the people who look down on stuff like this. EVERYBODY cares; they (you) can't help it. Repetitive manipulative messaging (think sexy slim people doing sports in beer ads) works even if you're aware that it's a complete ruse. It persuades your lizard brain and there's nothing you can do about it. This sort of brainwashing is one of the most disheartening and insidious things about human nature and, by extension, the marketing industry.

Don't get me wrong, there's good marketing. There are companies doing amazing stuff and they can tell you about it honestly with a take-it-or-leave-it mentality. It's a noisy world and when enough money gets involved (enough is usually a surprisingly low bar) being honest quickly becomes not good enough.

After spending a couple years in marketing and seeing this kind of research over and over I had to get out. (I left an online marketing agency to focus on design and tech.) I think on some level, it's taking advantage of human weakness. It's in the same ballpark as the studies that show that the placebo effect still works a little bit even if you TELL THE PARTICIPANTS they're in the placebo group. WTF, brain? It's so frustrating.

Anyway, You can bet your life that as much aggregate exposure as the Yahoo! logo gets, that selecting a proven one will impact their bottom line.

It's not a hair worth splitting for your start up with 30 customers, but once you start dealing with billions of slices of attention, shit gets weird and unintuitive. See Google's infamous testing of 41 shades of blue [http://www.nytimes.com/2009/03/01/business/01marissa.html]

This is simply the wrong way to do branding. Even it's just for testing logo design. Come on. I know branding from strategic angle could be very simple to explain. It's all about how human beings perceive a "thing" and generate associations. However, executing one part of the branding job like this makes the test less meaningful. A great brand or whatever thing it is fulfills people's needs by leading them to a unique journey, not by asking specific place they want to go. Feedback is very important, but there is other ways to achieve it better. Moreover, testing is better to be done from a branding's perspective. Trees and forest are both crucial.

People's stated preferences may be totally uncorrelated with what Yahoo really cares about which is how their logo affects people's use of Yahoo properties vs their competitors.

So while I agree with the statement "test don't guess" wholeheartedly, it isn't clear that what was tested matters or that the results are actionable. People often respond to things they would subjectively describe as ugly in very positive ways wrt key conversions such as signups, sharing, long term engagement.

In general test behavior, not opinions if you can. People lie to themselves and pollsters constantly. This is an awesome article about how the Obama campaign used testing to design their email campaign:

http://www.businessweek.com/articles/2012-11-29/the-science-...

Big takeaway: you have no idea what is going to work, measure behavior not opinions.

By a large margin, users prefer a slanted exclamation mark, serif font, uneven baseline, and uppercase letters.

Now, the selected logo was almost identical to the current logo, except with a sans serif font. But wait... don't users prefer serif fonts?

So wouldn't a truly representative test also measure the logos against the current logo? Did Yahoo perform this kind of sanity check when trying to choose a new logo, I wonder?

And as a side note, I just want to say... some of those Yahoo logos are truly terrible.

They spent 30 days on logos meanwhile butchered their sports site. So many errors and so many inconsistencies, you can't even look at player game logs anymore. One page is all in black, while others have the old look(better) in white.

And the kicker is that they role it out a 2 weeks before fantasy football starts. You cannot even move to another site in time.

All this time spent on self promotion, and so little spent on execution.

I am a frustrated consumer, and all I get are logos for 30 days....who gives a blank.

Why employ communications experts and designers if the final decision is left in the hand of the consumer?

Thankfully this has simply been an exercise in PR. The proof is in that logo. It's goddamn awful. I imagine it appeals to a nostalgic memory. It's appeal, much like a bad movie sequel, is base.

Remember a logo is not a brand. So this decision will not have any meaningful repercussions beyond hype.

What I take from these findings is that people prefer the current style of Yahoo! logo, so for their sake, the new logo will want to be very similar to their current one. Obviously Yahoo! were blanket testing what worked best and presumably have even more detailed statistics about why people liked which logo.

This seems like it could have been a terrific genetic algorithm experiment in picking the new logo - create populations of the logo with variation in shape, font, spacing, and whatever else they could think of. Use ratings as the fitness function, and iterate.

That would create the blandest option, the epitome of "meh".

For a similarly inspired experiment in art, see http://denisdutton.com/most_wanted.htm

I suspect that winning online surveys is not a good fitness function for a successful logo.

I'm reminded of the spat over the London Olympics logo. Say what you will about the chosen design, many of the "better" alternatives offered to much acclaim were completely unsuitable. In particular, I remember one was a takeoff of the London Underground logo, which, while doubtlessly sure to be popular in online polls, rather missed the point of creating something distinctive.

Interesting that the most preferred logo was not one of the choices. The logo in the choices that is closest to the preferred one is different in some key ways, and a better logo IMO. I wonder which of the two logos users were actually asked to judge.

Personally I don't think it will look good until they drop the exclamation mark from their name.

This is a great case study for why design by committee fails and why design should not be decided by the masses.

From The Wire -

Stringer Bell changes the name of the product - http://www.youtube.com/watch?v=KbbZc2pab9k

Wait, that logo thing already finished? a) I've been too busy. b) Yahoo need a better marketing department, I didn't hear about it after day 1.

I now know what it is like to be a complete simpleton in the face of a courageous nerd rage and be completely baffled why the dude cares.

Interesting gestalt shift.

I personally think that they should have published logos in different colors as well, some of the sans serif fonts looked better in darker colors

This is what happens when you unleash an intern on http://www.dafont.com/

It's weird to me to assume that the "best" new logo is the one that people tell you they like the most.

Yahoo is overthinking this.

People liked that logo because you can read it easily, and it looks like a logo not just a word.

This article wasn't written by Yahoo!, and to my knowledge the company who did this study was not commissioned to do so by Yahoo!. So it's not Yahoo! overthinking anything (yet), it's a company showing off what service(s) they offer.

YAHOO IS RED!!!! doesn't anyone remember! <old crank sees himself out>

so now you can really tell that Mayer's running the show

This reminds of the famous tea kettle parable.

Reply on news.ycombinator.com