Closed Bug 683196 Opened 15 years ago Closed 14 years ago

[FX Download Testing] Create 2 new home page graphics to test without monsters

Categories

(Marketing :: Design, task)

Product:
Marketing
Component:
Design
Platform:
x86
macOS
Type:
task
Priority:
Not set
Severity:
normal

Tracking

(Not tracked)

Status:
RESOLVED FIXED

People

(Reporter: christine.brodigan, Assigned: smartell)

References

(
URL
)

Details

Hey Tara, So, we'd love to test two different types of images on the /new experience without the monsters to see if they impact conversion and affect brand perception. 1.) a combination of desktop & mobile devices, see https://skitch.com/chrissiebrodigan/fiypn/screen-shot-2011-08-30-at-10.09.35-am for an example (ideally, we'd be able to improve the user's ambient understanding that Firefox is mobile too without words) 2.) a photo of the real people who make Firefox, ideally the iconic photo from the Summit It would be great if we didn't need to add much copy to explain the images, so that if they work well we could also implement across the European locales easily - but, if needed, no worries :) Thanks so much for helping. If possible, this is a P1 for Q3 and we'd love to get the running before the all-hands, since that will be a rigorous week.
We should really brainstorm the right set of options here, but definite +1 for doing this.
(In reply to John Slater from comment #1) > We should really brainstorm the right set of options here, but definite +1 > for doing this. Yep, yep, please all add your thoughts! I've put my two down in the description. Blake has referred to wanting to test idea 1. a combination of equipment and the web team (during our onsite) advocated for testing idea 2. a photo of Mozillians v. the creatures. Laura, John, Tara, etc. can you add your thoughts - ultimately, John's decision on what gets created here. Big thanks! ~cb
I vote for showing our product offerings. something like this: http://www.flickr.com/photos/29142435@N08/6026337253/in/set-72157627271263053 Except that I'd rather not just force in an existing graphic, but make sure it works well within the homepage. The device graphics are compelling, but usually very heavy and also drown out the standard screenshot. So we should try and show the screenshot for the desktop within the context of a laptop or something. I also like dynamic content. You can show picture ofmozillians inside the device, and also rotae to show other sites.
(In reply to Tara from comment #3) > I vote for showing our product offerings. something like this: > http://www.flickr.com/photos/29142435@N08/6026337253/in/set-72157627271263053 > > Except that I'd rather not just force in an existing graphic, but make sure > it works well within the homepage. The device graphics are compelling, but > usually very heavy and also drown out the standard screenshot. So we should > try and show the screenshot for the desktop within the context of a laptop > or something. > > I also like dynamic content. You can show picture ofmozillians inside the > device, and also rotae to show other sites. +1 Tara, I really liked that on the sync page. We'll still want to test a second image that doesn't involve any devices. Ideas on that?
What if we waited to show the tablet / phone product shot as an introduction to the new mobile UI when it hits? http://ianbarlow.wordpress.com/2011/08/30/firefox-for-tablets/ I think it'd be awesome to introduce that in the product shot promo style. ----- re: photo - I'd like to have a go at adapting the basic template to allow for hard edge photographs. It currently works well with illustrations that don't have hard edge boundaries, but I'd like to see what we can do with photographs. Also: A photograph will increase the Kb size of the homepage by quite a bit. What should our limit be for image size / performance hit?
(In reply to Sean Martell from comment #5) > What if we waited to show the tablet / phone product shot as an introduction > to the new mobile UI when it hits? > > http://ianbarlow.wordpress.com/2011/08/30/firefox-for-tablets/ > > I think it'd be awesome to introduce that in the product shot promo style. > > ----- > > re: photo - I'd like to have a go at adapting the basic template to allow > for hard edge photographs. It currently works well with illustrations that > don't have hard edge boundaries, but I'd like to see what we can do with > photographs. > > Also: A photograph will increase the Kb size of the homepage by quite a bit. > What should our limit be for image size / performance hit? RE: new mobile UI - Sean, do you know when it will launch? I would hate to hold back on a test, waiting for a product launch. Would love to update the graphic with new stuff from the launch post- launch though. Great question, the new graphics should be the same size as the current monster graphic.
Here's my take on how we should break it down: 1. Product shot - I like the direction Tara mentioned in comment #3, and also Sean's idea about adding hard edges to our imagery. When we do this, though, let's remember that people are coming to this page to download the desktop version so that should be primary. I worry a bit that showing multiple products in an array would confuse people...we don't want to do anything that would give even a moment's hesitation that we're offering a desktop software product so that's very key. The main challenge here is to find a way to make a screenshot interesting. Not easy, but I know our team can do it! 2. People shot - I'm thinking this should be shots of people to serve more as stand-ins for the user rather than a group shot from the Summit. I think people are more interested in seeing how the product affects them rather than seeing the people who made it. To pay this off, we'd definitely need to tweak the headline as well. To make it a little more visually interesting, we could do a stack of pictures on top of each other, so you'd mainly see the primary image on top but others would be sticking out to add color. Just a thought though. The main goal here should be to give the browser a human feel and make the viewer identify with the pictures so they could imagine themselves using Firefox. I don't want to use stock imagery here, but maybe we have some community pics we could use.
Considering the current monster shot is mostly line art, we may have problems getting a photograph down to the same size. Just a heads up there.
(In reply to Sean Martell from comment #8) > Considering the current monster shot is mostly line art, we may have > problems getting a photograph down to the same size. > > Just a heads up there. Yeah, will be tough! Got magic?
(In reply to Sean Martell from comment #10) > If this works, yes: http://www.websiteoptimization.com/speed/tweak/jpegmini/ This is going into a test, so no matter what we'll learn from the results. I'm more concerned about page load post-merge, because we've added 2 seconds to page load with the new redirect. We'll do what we can, Martell magic-enabled is a big plus in this case, and I'm excited to see what happens. Blake cc'd in case he's got magic up his sleeve too. Thanks for everything Sean, in spite of the hard edges, you're still a big softie.
Blocks: 683377
Assignee: tshahian → smartell
Tara/Sean, Do you have enough from the team to take a first pass at this? Ideally, we'd love to start testing before the all-hands. Thank you! LMK, CB
Summary: [A/B Testing Downloads] Create 2 new home page graphics to test without monsters → [FX Download Testing] Create 2 new home page graphics to test without monsters
Um, testing before the all-hands is a little ambitious. I will let Sean respond.
(In reply to Tara from comment #13) > Um, testing before the all-hands is a little ambitious. I will let Sean > respond. If the graphics are complete, the testing is a breeze to set up, turn on and turn off. Are you estimating that these two graphics will take more than two weeks to fit into the workflow? Should we look to a contractor? I have a number of wonderful designers who could definitely field these two graphics quickly and brilliantly. Also, these are just to test and they won't go to a full 100% audience until after a refactoring and polishing session. LMK, CB
I say it's ambitious because I"m getting tons of requests within the same timeframe and our designers are pretty strapped working on various high-ish priority items, in addition to prepping for the all-hands. So, not sure what you meant by ready to test before the all-hands... but given that we have about 1 week to do this, and it's a design direction we're still exploring and would need time to iterate on...it seemed tight. Generally speaking, I prefer to keep these types of projects in-house vs. be quick to outsource as it's not always easy or efficient to do that. Transferring files and knowledge takes time and I prefer to keep pages like our main download page reserved for our in-house team to pour their creative talents into. That said. I'll let Sean chime in on how he feels about timing and what works best for him. I'd like him to work on this if he's still interested to take it on.
(In reply to Tara from comment #15) > I say it's ambitious because I"m getting tons of requests within the same > timeframe and our designers are pretty strapped working on various high-ish > priority items, in addition to prepping for the all-hands. So, not sure what > you meant by ready to test before the all-hands... but given that we have > about 1 week to do this, and it's a design direction we're still exploring > and would need time to iterate on...it seemed tight. > > Generally speaking, I prefer to keep these types of projects in-house vs. be > quick to outsource as it's not always easy or efficient to do that. > Transferring files and knowledge takes time and I prefer to keep pages like > our main download page reserved for our in-house team to pour their creative > talents into. > > That said. I'll let Sean chime in on how he feels about timing and what > works best for him. I'd like him to work on this if he's still interested > to take it on. Hey Tara (Sean), I'm 100% comfortable using a designer I've worked with to generate these two graphics. These graphics are only going to be seen by a few thousand users in a test cycle. Totally appreciate that there's a ton of work for the in-house team. This seems like a great opportunity for us to onboard a trusted contractor who can be helpful towards our developing culture of lots of A/B testing to really learn what we can and deliver awesome to our users. This is the kind of thing we'll be doing more of, so having someone in our pocket who can generate awesome test-friendly graphics would be ideal. It would be a shame to hold back on testing and optimizing, and it would work against our ability to complete our Q3 goal. Happy to share both emails and portfolios over email - just let me know what you need, so we can move forward on this and launch before all-hands. Best, CB
Thanks. I'd like to wait and see if it's something Sean can tackle first. Then discuss alternate options and next steps from there.
(In reply to Tara from comment #17) > Thanks. I'd like to wait and see if it's something Sean can tackle first. > Then discuss alternate options and next steps from there. I think that's the right approach. Let's see what Sean says and then go from there (Sean, please weigh in). From my perspective, I'm definitely very excited about this test as I think it's a big part of the future direction of the site. But, in terms of immediate priorities (i.e., in the next week) vs slightly longer term (i.e., in the next 2-3 weeks) I think it's ok to downshift on this if we absolutely have to. Would be great if we didn't have to, of course, but sometimes you have to make tradeoffs.
I'd prefer to do this in-house. I'll tackle this. I'm clearing out a backlog of work now, but please put this on my radar and assign to me. Thanks!
Thanks Sean. Does that mean next week is doable then?
Sure. LET'S DO IT! \o/ *cues Rocky music, drinks more coffee*
Coffee fixes everything. Thanks Sean!
Awesome, thanks Sean! What's the next step here?
(In reply to John Slater from comment #24) > Awesome, thanks Sean! > > What's the next step here? Feedback: V1: *Are we staying with the soft feathering v. hard edge, so we could more easily swap images out? V2: *Really different, which is great. The red doesn't feel friendly and can have negative connotations for users (here is an example of an A/B test that tried this unsuccessfully: http://mat.usc.edu/MAT/index8.php http://mat.usc.edu/MAT/index7.php (still winning and going on 2 years) Client still wants to get more red into the winning design: http://mat.usc.edu/MAT/index1.php http://mat.usc.edu/MAT/index2.php http://mat.usc.edu/MAT/index3.php ***** given that the next page is the /fx page it could be very jarring even for this small test. I think anything is worth testing, but not everything is worth releasing even if it performs well. *If possible it would be great to see this without the red as well.
I think we should go ahead and test both of these...I'm sure either will need additional tweaks if it performs, but this will be helpful in determining future direction of the site.
Yes, the blood orange color is based on the brand video, so it'd be good to see if it is a dominant colour we should pursue. Might as well test it!
(In reply to Sean Martell from comment #27) > Yes, the blood orange color is based on the brand video, so it'd be good to > see if it is a dominant colour we should pursue. > > Might as well test it! I think anything is worth testing. Backing up a bit, we had a similar experiment with the Join Mozilla landing page and I've mentioned my experiences with red and it's negative impact on conversion. It's really good experience that we have on the table. Since mozilla.org won't be a landing page in the sense that mozilla.org/firefox will always be, so sticking in the theme for the /firefox design is sensible and could really help with Q4 download strategy. With scant resources, I'd really love to ensure we have a blue theme to test. We've been pushing this since early Q2, so it would be great to put together a great package - it might be the only test we run. Next steps seem to do a round of design revision and then move over to a testing plan. Sound OK?
Blue version: http://cl.ly/192b3t3K0j3o1B3r3s2y PSD: mozilla.seanmartell.com/engagement/mocodotcom/testing/home_nonfxusers_revised.psd I made the bottom white again to keep it lighter. Thoughts?
Blocks: 689746
Looks good to me. Can we commence the A-B-C test?
Can we close this out?
(In reply to Tara from comment #31) > Can we close this out? Not yet.
Ok thanks. Can you please provide a status update then so I know where things stand on this bug? Thanks
(In reply to Tara from comment #33) > Ok thanks. Can you please provide a status update then so I know where > things stand on this bug? Thanks Waiting on making sure we have all the assets for bug 689746 . I'm sure we'll close out next week. Thanks for hanging tight.
Hi there. Looks like we're good to go here, right? I'd like to close this bug :)
Status: NEW → RESOLVED
Closed: 14 years ago
Resolution: --- → FIXED
You need to log in before you can comment on or make changes to this bug.