An A/B Testing Mad Lib
One of the great pleasures of working at Optimizely is getting to see so many people having so much fun in their work (not just on our team, but with our customers too). Work and fun are not often caught hanging out together, so to speak. But the world of A/B testing on Optimizely lends
One of the great pleasures of working at Optimizely is getting to see so many people having so much fun in their work (not just on our team, but with our customers too).
Work and fun are not often caught hanging out together, so to speak. But the world of A/B testing on Optimizely lends itself well to moments of great joy and celebration in the workplace.
Each day, I am delighted to hear from colleagues, customers, testers, and the online marketing world at large how excited they are to discover a better way to connect with their site visitors through testing (or re-appreciate an old one that has yet to be bested).
By far though, one of the most awesome results I see day in and day out is the knowledge sharing and learning that comes with testing. There are many ways to share the insights you glean from any particular test with the people around you – whether testers are celebrating their learnings with a strategically placed foghorn and siren in the office, a wall of fame for anyone who proposes a winning test idea, an internal social network, or any other variety of ways our customers share with their teams – the power of sharing the stories of meaningful experiments should never be underestimated.
After all, the more fun your team has making your workplace succeed, the better it is all the way around.
There are plentiful ways to make your knowledge sharing happen.
In the spirit of sharing with you, I have fused a great activity from my past – the Mad Lib – with the notion of sharing your testing learnings, and crafted below a sample Mad Lib email as one way to have some fun while you share some learning.
Dear [Endearing name for your co-workers],
I wanted to send along a quick note to share a success story of [tall mountain of geographic relevance to you] proportions with all of you.
Recently, I overheard a conversation between [name of opinionated colleague] and [name of another opinionated colleague] discussing ways that we could increase our [business critical online conversion metric].
My ears perked up like [celebrity dog name] and I immediately realized this was an opportunity to resolve this debate with data.
Armed with a couple of ideas and my handy login to [A/B testing platform ending in ‘ly’], I set out to find a definitive resolution to this quandary.
I quickly crafted an A/B test as follows:
On the [web page(s) being tested], I set up two variations, described below:
Variation 1, which I dubbed [opinionated colleague] variation, had [brief description of the variation].
Variation 2, which I dubbed [opinionated colleague 2] variation, had [brief description of the variation].
I defined [business critical online conversion metric] as the goal to measure for this experiment and was ready for launch.
Within [short amount of time it took to get your A/B test up and running], I had set the wheels in motion to discover which idea would stand tall as the winner.
Admittedly, I felt a bit like [famous explorer], charting my own course towards Crushing-It-Ville.
[Amount of time the test ran until a winner was determined] later, I received an email notification to check in on my test results – I really appreciate these heads up notes that [A/B testing platform ending ‘ly’] sends me.
Channeling my inner [academy award caliber actor], I took a dramatic pause before clicking the link to take me to the results page.
After [total number of visitors in the A/B test] of our [endearing term used to describe your site visitors] came to the site to help us really get to bottom of this, I am proud to officially share with you the following:
[Opinionated colleague 1] variation: [Results]
[Opinionated colleague 2] variation: [Results]
What’s exceedingly cool about this, other than the fact that we have settled yet another debate with data, is that we can now deploy [winning variation] and reliably know that our [business critical online conversion metric] will get a [percentage increase from winning variation] lift.
I have given a bit of thought to why [winning variation] performed better. I’d love to hear yours – anyone who wants to contribute to the conversation is welcome, just click here [link to convenient place where you and your colleagues can share ideas and discuss tests with each other].
If you want to settle any scores, be sure to let me know. Testing is awesome. Even more awesome than [something that is really awesome].
And if you want to see if you can best the brilliant idea of [colleague who proposed winning variation], lets do that, too. After all there’s more than one way to [activity that there is far more than one way to do].
How do you share your testing wins at the office? Let us know, in the comments below.