Digital Advertising Industry News


How To Use Variable Testing To Win Your Office’s Halloween Costume Contest

October 24, 2016 Kathy Bryan

It’s not easy to select a winning Halloween costume. A variety of factors, including the party setting, attendee demographics and cultural changes, can have a significant impact with regard to which costumes get the most votes. If you’re resolute about having the best costume at your upcoming festivities, we recommend the utilization of variable testing.

As marketers, we typically advocate A/B testing. However, since small, incremental changes are required to accurately evaluate results, it’s going to be too time-consuming of a process. With Halloween right around the corner, you’ll have to take some shortcuts. But if you set your mind to it, you can still get a lot of costume variations tested between now and the big event.
First, Understand Your Target Audience
The more you know about your target audience, the more likely you’ll be able to envision a costume your co-workers will appreciate. Consider your audience’s IAO variables, or interests, attitudes and opinions. If your costume connects with what is relevant to your audience, it is likely to be well received.

Then, Start Costume Variable Testing

The steps below are pulled from the DMS guide to A/B testing landing pages. Although we are recommending a streamlined testing process, if you follow these instructions, you should be able to narrow your costume choice to one that has a good chance of winning.

1. Define your variable testing objective:

Your testing goal must be measurable. While you can track and evaluate other key performance indicators, only the metric that relates to your testing objective should determine the results of your test.

2. Create a variable testing plan and schedule:

List out all of the costume variations you would like to test. Then, develop a hypothesis on the impact of changing those elements. Make sure to prioritize your testing schedule to have the biggest impact toward your objective as soon as possible. That way if you run out of time, you’ll still have tapered your choices significantly. Because each test may reveal new insights about how you should move forward with subsequent testing, expect to re-evaluate your testing plan and schedule after each winning variation is determined.

3. Create your variable testing variations:

Test just one thing at a time ― otherwise, you won’t know what delivered the results. Be patient during this process… but not too patient. Halloween is approaching fast!

4. Volume is key for variable testing:

Sitting alone in your office will not help your testing process. Make sure enough people see each variation to accurately pick your winning costume. Try walking through public spaces to quickly boost impressions.

5. Know when to stop variable testing:

To alleviate external factors that could impact the results, conduct the test for the shortest period possible.

Contact DMS

Important Disclaimer:

It is possible your co-workers will feel like you cheated the system and, therefore, not vote for your costume the day of the competition. If this happens, hold your head up high and know you did your best.

About the Author

Kathy Bryan

Kathy Bryan is the Chief Marketing Officer at Digital Media Solutions (DMS). In this role, Kathy is responsible for all aspects of marketing and communications for DMS, the leading global martech company leveraging innovative, performance-driven brand and marketplace solutions to connect consumers and advertisers.

Follow on Linkedin Visit Website More Content by Kathy Bryan
Previous Article
Snapchat vs. Instagram Stories: The Race Begins
Snapchat vs. Instagram Stories: The Race Begins

The world of social media is forever changing, and those who don’t evolve are lost in the dust. Likely with...

Next Article
There’s Still Time to Get on the Bus for the 2017 Back-to-School Season
There’s Still Time to Get on the Bus for the 2017 Back-to-School Season

The U.S. back-to-school (BTS) season is projected to rake in $857.2 billion in retail sales plus $74.3 bill...


I Want to Maximize My Campaigns!

Thank you!
Error - something went wrong!

Thank you!
Error - something went wrong!
× Streams