Three tips for a/b testing your nonprofit online marketing efforts inspired by the Obama campaign

As the nation's capital cleans up from the weekend's Inaugural festivities, it's worth asking how we got here:

How was it that President Obama came to be standing on the steps of the Capitol being sworn for a second term rather than his rival, Governor Mitt Romney, being sworn in for his first?

Presidential Inaugural Parade [Image 11 of 11]

One undeniable factor was that Obama's Digital Team ran one of the most successful online campaigns ever.

Curious about how? I certainly am, which is why I was so excited to see the insightful blog post by Kyle Rush, Deputy Director of Frontend Web Development at Obama for America, called Optimization at the Obama campaign: a/b testing. It's really worth a full read, so I suggest grabbing a coffee or tea and digging in.

Laboratory GlasswareThe optimization details Rush shares are fascinating. The specific results aren't necessarily going to be the same as yours but its the process that I find most interesting. After working a full 14 months to optimize their donation form, Obama's team could have just said "Hey, we're done optimizing, it's as good as it's going to get".

But they didn't.

Instead, they busted out their lab goggles and got to work experimenting like mad scientists. They took risks. They tried things that were "out of the box" and might not work. They took a really scientific approach to their efforts and tested, innovated, iterated and tested again to squeeze out every increase in performance they could with some truly remarkable results.

Want your nonprofit to jump into the lab and do a/b testing yourself to concoct some online marketing success of your own? Here are three things you'll need:

  1. An actionable and measurable goal No need to go overboard at first, just pick one clear ask/request with an end-result that can be completed online (usually something like clicking a button or completing a form).
  2. A variable to test and a hypothesis Pick one thing to test at a time, maybe it's different copy, a different layout or a different photo. Don't try to change too much at once.
  3. A testing tool Does your organization use Google Analytics? Then you have an a/b testing tool in your arsenal already. Check out this post on how to run Content Experiments within Google Analytics. Yes, it's going to take some figuring out, but it's not cost prohibitive to get started.

If your organization is already using a/b testing or other optimization strategies, we'd love to hear about your successes and challenges! Let us know by commenting below or drop us a line.

Photo credit: DVIDSHUB on Flickr and LoveInTheWinter on Flickr