search
 


Blog

Entries in multivariate test (36)

Wednesday
Mar302011

Why can't I use my homepage as a landing page?

A lot of marketers send ad campaign traffic to the home page, or to deep links on a company website, but that’s not really the best strategy for maximizing conversions. At ion, we advise these marketers to use campaign specific landing pages to achieve better results.

Here’s Shelby Nousain, Account Development Manager, to talk about why this is a much better strategy for maximizing conversions.

Have a question about post-click marketing? Send us an email at ask@ioninteractive.com and we might answer it in our weekly Ask ion video! 

Two weeks ago we did a video on when to run an A/B versus a multivariate test, and last week we did a video on how long you should run an online test for.

Thursday
Mar242011

How long does it take to get online test results?

The time period it takes to get statistically significant results from your A/B and multivariate tests will vary depending on how much traffic you’re sending to the pages, and how many variations you’re testing (in the case of MVT).

Here’s ion account manager, Audrey Ross, with all the details:

Have a question about post-click marketing? Send us an email at ask@ioninteractive.com and we might answer it in our weekly Ask ion video! 

Last week’s video was on when to run an A/B versus a multivariate test.

Thursday
Mar102011

When should I do A/B testing versus multivariate testing?

We often hear the question, “when should I run an A/B test versus a multivariate test (MVT)?”

A/B and MVT are both part of a strategically sound testing program and when applied in the the right way can lead to big improvement in your campaign results.

Here’s ion account manager, Jessica Eck, with all the details:

If you have a question, feel free to send us an email us at ask@ioninteractive.com or leave a comment below!

 

Wednesday
Jan122011

How many landing pages do you need? 

how many landing pages do you need?How many landing pages do you need? To conclude the Creating Effective Landing Pages blog series, we’ll wrap up with a commonly asked question, “how many landing pages should I have?”

That’s near impossible to answer because it’s an ever-evolving number. To take a stab at it: if you’re a global brand running lots of ads, I’m guessing it should be up there in the hundreds or even thousands. Seriously.

You should have at least one landing page per ad campaign you’re running. On top of that you’ll need “challenger” pages that you test against your “champion” pages.

A champion page is your current landing page, and the challenger is a second (third or fourth) page that you test against the champion. The winner is the higher converting page. In many cases your challenger will beat the current champion and become the new champion. When this happens, testing isn’t over; you simply start again by either testing elements within the new champion (multivariate testing) or by creating a new challenger to test against it (A/B testing).

Because testing is a never-ending process and because your ad campaigns are ever-changing, the number of landing pages you need will vary all the time. This is why landing page software is so important – landing pages are disposal — your champion today may be outdone by a new challenger tomorrow. Marketers must have a tool that helps create, launch and test landing pages on the fly without a ton of investment in time or money. With landing page software you can create as many landing pages as you need without spending a boatload of time on the creation process. You just don’t want to get caught up in the design or coding of a landing page, when the important stuff doesn’t happen until after the page is launched. At the same time, you don’t want to launch a half-baked (read: ugly) page that will likely be low converting. With landing page software you’ll work off of a branded and templated system that gets you going in minutes or hours, not days or weeks.  The focus won’t be on getting the page launched, but rather catching conversions and analyzing why the non-converters walked away.

And that concludes the Creating Effective Landing Pages blog series. Looking forward to your feedback in the comments and on Twitter @ioninteractive.

Sunday
Jan092011

How to get your landing pages ready for testing 

Are your landing pages ready for testing?Are your landing pages ready for testing? In the first part of the Creating Effecting Landing Pages series we talked about the difference between tracking click-thru rates and conversion rates, and why tracking conversion rates is more important. In part 2 we’re going to talk about how to get your landing pages ready for testing and optimization.

While “best practices” are often shunned, and rightly so, there are some “best practices” that should be adhered to, particularly when it comes to the user experience (UX) of your landing pages.

I’m not suggesting you usability test your landing pages because seeing what a test group does won’t tell you how your landing pages are going to perform with a variety of different visitors once the page is live. However, I am saying there are fundamental UX principles that should be considered before you even begin testing.

Here are four things to consider before you begin testing your landing pages:

  • Load time of your page
  • Messy, multi-layered navigation
  • Browser optimized
  • User-forgiveness

Once you have these basic elements nailed down, now it’s time to start testing.

Here are 5 quick ways to start testing your landing pages:

  1. Behavioral targeting – keep this simple: give second time visitors different information than you gave them the first time (obviously they’re interested if the returned, now give them something new to convert them).
  2. Copy – change your messaging, headline and body copy, and experiment with bullets and paragraph lengths.
  3. Images – experiment with people (male vs. female, group vs. individual, looking left vs. looking right), product shots, animals, charts, etc.
  4. Geolocation – test providing local content such as mentioning specific city names and showing images that reflect the location the visitor is in.
  5. Content – experiment with different content elements (tabs vs. no tabs, still image vs. video, bullets or no bullets, etc.