Multivariate Testing and Optimization of Flash

5 Comments

by MrOptimization in Basics

How do you do Multivariate Testing and optimization of Flash?

YOU DON'T!

Flash is a dead technology. Some people just can't bring themselves to accept that yet.

But if you help nudge along Flash's inevitable death inside your own organization, you'll look like a genius in the future.

Flash and Optimization don't mix

Non-standard, proprietary objects that look and work differently based on the whims of each designer are called art.

If your digital channels are all about creating art, then you can probably skip this and keep using Flash.

But if you're interested in using your digital channels to make money and improve customer satisfaction, read on because Flash is doomed and you need to start purging it from your digital environment now.

What killed Flash?

Standards, plus a heaping helping of its own sucky-ness, killed Flash.

Flash DID help move the web along way back when, and we thank it for that. For example, if you wanted to create a banner ad that turned into a madcap penguin sashaying across the screen, Flash was by far the best choice for a long time.

Very soon though, HTML 5 browsers will include almost everything Flash used to be needed for.

So you might be asking "if the same functionality is free, built into the browser, fast and based on worldwide standards, why would I use slow, proprietary, crash-prone junk like Flash?"

Bingo! And the really smart people, like Mozilla VP of engineering Mike Shaver, agree:

“The future of the web is the web, and betting against the web is a bad idea.”

Nowadays, if you're betting on Flash you're betting against the future of the web—and possibly the future of your career.

Big guns now aimed at Flash

Mike Shaver at Mozilla is in good company. The parade of smart and influential people predicting Flash's demise is deafening.

One of the leaders of the Flash doom parade is none other than cultural and technical icon and all-around object of worldwide worship Steve Jobs himself.

Last month, Steve even climbed down from his mountain of money to explain why Flash needs to die.

Here's the shortcut version of why Apple doesn't allow Flash on its devices and, by extension, why Flash is doomed in the larger market:

• "Adobe’s Flash products are 100% proprietary. They are only available from Adobe, and Adobe has sole authority as to their future enhancement, pricing, etc."
• Apple's hardware and OS are proprietary, but for content and distributed digital experiences, "Rather than use Flash, Apple has adopted HTML5, CSS and JavaScript – all open standards."
• "Apple’s mobile devices all ship with high performance, low power implementations of these open standards."
• Apple open-sourced its WebKit HTML5 rendering engine and now "Almost every smartphone web browser other than Microsoft’s uses WebKit. By making its WebKit technology open, Apple has set the standard for mobile web browsers."
• "HTML5, the new web standard that has been adopted by Apple, Google and many others, lets web developers create advanced graphics, typography, animations and transitions without relying on third party browser plug-ins (like Flash)."
• Video encoded in H.264, a format supported directly in mobile hardware, Blu-ray DVD players, etc. can be played faster and with less power drain without Flash
• "Flash was designed for PCs using mice, not for touch screens using fingers"
• Flash is a crappy way to develop and deliver applications

For those who purr over his every word like a newly fed kitten, here are Steve's complete Thoughts on Flash.

Multivariate Testing of Flash

Used wisely, MVT can help you make the case to flush Flash from your world faster.

Start by asking "do we really need Flash for this?"

In most cases the answer was always no, but Flash looked snazzy at the time so you went ahead and used it.

Now Flash may have infested many corners of your digital properties where it doesn't belong, such as:

• Splash screens. Flash or otherwise, these shouldn't exist at all.
• User interface components such as navigation bars, form input fields, etc.
• Search results display
• Shopping carts
• Content presentation elements
• Functional widgets that can be implemented in other technologies (calculators, sorters, etc.)

All Multivariate Testing technologies require redevelopment work if you want to optimize within Flash components.

Why not use that time more wisely? Instead of throwing more good time and money into the doomed pit o'Flash, develop standards-based replacements for the Flash components to run through multivariate tests.

For example, even when they look and behave almost exactly the same, it's been proven that AJAX components tend to outperform their Flash-based predecessors in the same locations.

Where Flash is still a necessity, quarantine it

"But we have a mission-critical Flash-based [insert thing that HTML 5 can't do yet] !"

Ok. If it absolutely, positively needs Flash today—such as certain types of video, animation, banners and games—you should be doing it as shown below.

Surround the Flash disease with healthy, future-proof standards-based tissue, like so.

Quarantine and Kill Flash

And what if your whole site or application is built entirely in Flash? Don't panic! There's still time to get another job before the angry mob fires up the torches and comes looking for you.

No new Flash!

You should stop developing new Flash components as soon as is practical. At the same time, use Multivariate Testing to prove why you need to progressively rid your digital environment of its existing Flash infestations.

So long as a standards-based HTML 5 equivalent is good enough, use that instead.

Your digital audiences—and your career—will thank you.