Alix F.

If you have downloaded the en-US version of Firefox in the past few days, you might have noticed a change in the page displayed while you are downloading (i.e. after you click on the download button).

Why refresh it?

As part of the initiative to improve user activation, engagement, and retention, it is important that people who have made the decision to download and install Firefox do so successfully. Simple and visible instructions, and links to support, are key factors to achieve this goal.

What has changed?

• Added the Firefox logo to the page to increase brand recognition among new users, making it easier to remember what icon to click on to launch Firefox.

• More prominent link to start the download manually, appearing before the installation instructions.

• Three steps to install: the number of steps on the page has been reduced from 4 to 3. In addition, more details are provided, such as a warning that the download may take a few minutes. Finally, the 3 steps are aligned so it’s possible to quickly glance at them. It also makes it less likely that the download window will cover instructions or that someone would have to scroll to see the images or the text.

• Links to help and tips & tricks: the page now links to Mozilla support for users who require more help during the installation process. It also links to tips & tricks for more about how to use Firefox, something interesting to check out while waiting for the download to complete.

• Link to more about Mozilla: this gives new users one more opportunity to learn about how Firefox is made and why Mozilla is different.

• Updated flow to decide which version of the page to show. As you can see in the diagram, OS version and browser are now detected to show the steps that are closest to the user’s actual download and install experience. For example, Windows 2000 users will now see the default page and not the instructions for Windows Vista/IE users, because the installation prompts differ in each case.

• New warning page: if you’re downloading Firefox 3 from a platform that can be detected but is known to be unsupported, such as Windows 98, you will be warned of that fact and will have to manually start your download from that page.

What’s next?

• Monitor impact by looking at how visitors are interacting with the new page. If they click on support and view specific articles, it will help determine what issues users might be having and make the page even more relevant. In addition, Funnelcake tests show how many people click to download but never complete the installation. There should be an improvement in the next test’s results.

• Continue optimization: as we learn more from the success of the current pages and about the user experience during the installation process we will be constantly looking at refreshing the page to make it as effective as possible.

There is an opportunity to increase the number of active Firefox users by:

• making sure that people who choose to download Firefox successfully download and install it
• increasing engagement of new users so they continue choosing Firefox in subsequent browsing sessions

Check out past posts for a more complete overview of the opportunity: revisiting engagement and retention and from click to download to engagement.

After getting feedback from the community and analyzing the latest surveys and market research, we have put together a list of the next priorities for engagement and retention and proposed corresponding projects:

1. Remove any barriers to successful download and installation to make sure that users who intentionally click on the download button will get to the first run page. This will be done mainly by optimizing the download.html page and providing targeted support.
2. Make the availability of support top-of-mind so users who encounter an issue even months after installing Firefox will remember that help is available and contact Mozilla support instead of just switching browsers.
3. Help new users become power users of Firefox sooner so they can experience the benefits of Firefox right away and break old habits. Ex: more tips & tricks available via multiple channels (such as email or live bookmarks), more add-ons collections, more videos.
4. Engage the Mozilla community to help showcase the benefits of using Firefox, so we will be gathering more video and text testimonials. Because engagement starts at the acquisition stage (when someone decides to switch) we’ll also be asking for your help in getting the engagement message out: when you tell your friends about Firefox, tell them about add-ons and a few cool features too.
5. Facilitate the transition from other browsers to make switching simple. We’re going to provide a guide for switching to help potential new users become familiar with Firefox quickly and to answer their most frequent questions.
6. Address site incompatibility issues, whether real or perceived, to ensure users enjoy the best possible experience on Firefox.

I’ll be blogging about each point in the next few weeks, including some of the projects that have already started, but you can get a sneak preview on the Engagement wiki page, which has summary slides describing the opportunity and the priorities.

This is a follow-up to the post on revisiting engagement and retention which provides an overview of the following puzzle: what happens to people who click to download Firefox but never become active users?

To start the investigation, let’s look at the user experience from click to download to engagement.

Suppose you are currently using another browser, and have decided to switch to Firefox. You already know about Firefox and where to get it.

The first thing you need to do is download Firefox. You visit mozilla.com click on the green download button, complete the download and the install, and launch Firefox. From then on, you look for the Firefox icon and click on it whenever you wish to browse the web. You even install cool add-ons to customize the look and functionality of Firefox. You are so thrilled you recommend Firefox to your friends, family, and colleagues!

Now, wasn’t that easy? It did sound quite simple, but only because each one of those steps went very smoothly. However, our data shows that this is not necessarily always the nice experience people go through.

Here are some hypotheses of where things might go wrong along the way, focusing on IE users on Windows XP. We’d love your feedback on whether or not you’ve seen this happen, whether it’s likely or not. You’ll see at each step if we are currently able to quantify the success rate.

1. You visit Mozilla.com and find the download button.
2. You click on the download button, you see the download box (no data on this), and you are redirected to the download.html page (98% success rate).
or:
• the download does not start and you don't see the text "Your download should automatically begin in a few seconds, but if not, click here." on the download page
3. The download completes successfully (no data).
or:
• the download is interrupted because of a problem with your internet connection
• you stop the download because it takes too long
4. You find and successfully run the installer (no data).
or:
• you can't find the installer
• you have an issue during the install process and aren't sure where to turn for help
• you don't run the installer right away, because the download took too much time and you started doing something else and forgot about it. You may or may not run it later.
5. You launch Firefox for the 1st time (57% success rate for download + install + first run )
or:
• you unchecked the "Launch Firefox now" checkbox in the Setup Wizard and can't find Firefox anymore
6. From now on, the next time you wish to browse the web, you remember you have Firefox, find it on your desktop and launch it (49% after 30 days).
or:
• you didn't set Firefox as your default browser and can't find it anymore
• you started using Firefox but stopped (site incompatibility? missing plugins? didn't import your bookmarks & settings during the install and aren't sure how to transfer them?)
• you forgot about Firefox after you installed it...

What do you think? Has any of this happened to you or someone you know? Are there other steps/scenarios you can think of?

More importantly, if half of the people who start a download don’t become active users, what do you think we can and should do about it?

We’re in the process of revisiting opportunities for increasing user engagement and therefore retention . You may remember the Funnelcake experiment , which showed that a significant number of would-be Firefox users may fall out at different stages between the time they start downloading Firefox, and when they become active daily users.

Here’s how we’re thinking about the steps a user goes through from click to download to active usage of Firefox:

• Acquisition: the user is aware of Firefox, visits mozilla.com (or another site offering Firefox downloads) and clicks on the download button
• Conversion: the user has clicked on the download button, views the download page and completes the download
• Activation: the user has downloaded Firefox, the install is successful, the user launches Firefox and successfully views the first run page
• Engagement: the user has installed Firefox, launched it once, and launches it again within 30 days
• Retention: the user is actively using Firefox after 30 days.

There is a clear distinction between the steps from click to download to first run page (acquisition->activation), and the subsequent use of Firefox (engagement and retention). Our hypothesis is that issues with the former are more technical (download speed/interrupted download, ease of install), whereas issues with the latter are linked to a change of habit (ex: not always clicking on the blue “e” to access the web).

In the first phase, the user clearly signifies an intent to download, install and run Firefox by clicking on the Download button. Therefore, this process is something we’re looking at very closely to understand why it may not be completed successfully. For example, we’re planning to update the download.html page to better reflect the steps users should expect to go through. More on that later…

We’re also starting to think about programs to specifically address the habit change. It would be great to hear what your experience has been in switching to Firefox or getting others to switch and continue using Firefox. What has made Firefox “stick”? What did you have to overcome to change your habit? Were add-ons key to opening Firefox instead of another browser? Or maybe it was getting to the point where you started to benefit from the cool features in Firefox like the awesome bar?

Finally, more funnelcake experiments are planned, which will help understand and monitor the download funnel, engagement, and retention over time. It would be useful to hear even anecdotal evidence of the conversion and retention rates for other downloadable software products. Having benchmarks will help us understand what is achievable and where we have room to improve.

In the next post, we’ll be looking at how we’re measuring each step, what is working, what’s not, and how we can get better over time.