Recently I’ve been discussing barriers that prevent digital testing program success. Latest conversations have focused on Resource Barriers and Data Barriers. Both barriers are extremely important and can be attributed to many fruitless testing and optimization efforts. Today, I will be exploring the third and possibly most important barriers: the barriers of expectations.
After you have established what success looks like, quantify success based on any trending data you have available. Set realistic quantitative goals and deadlines for these expectations. This will make you, your team members and your agencies more accountable for results. For example:
- Increase online sales for our Initech Personal Software by 5% in 6 months
- Decrease churn rates Initech Monthly Software Product Subscriptions by 15% in 3 months
- Increase average order value of Initech by 7% by September 2012
This is just the very beginning. Once you understand the expectations for the final result, you can now step back and formulate a plan to get there. You now will need to take calculated steps that produce incremental wins, meaning that you need to establish a long term testing program that focuses on the micro-metrics for conversion points throughout your conversion funnel. Improving these points one by one will help to get you there. Give yourself a realistic timeline based on current conversion counts and traffic. Setting these realistic quantifiable expectations is a necessity for any successful testing program that is built to win.
You can’t realistically expect to make improvements with a “one and done” approach.
The final end all be all killer of digital testing programs is the infamous “one & done” test approach, meaningwe test once, we get a quick win, and we use that winner for now and forever – testing completed! One test is not the solution and does not provide a sustained lift. Testing is an ongoing, disciplined, iterative process. It is a discipline that becomes a part of a culture. Testing allows us to learn more about our customers and increases performance of digital marketing efforts.
Many one-and-done testing efforts will see initial spikes, only to trail off after 30 days (sometimes less). This can be due to a variety of reasons – fatigue, change in product trends, personalization, competitive innovation, seasonality, etc.
My point is that you must always be learning about what is working now. Always be curious and always be focused on continued improvement. Do something simple if you can’t create a robust test plan. For example: 1-list the items that you want to test; 2-schedule those tests; 3-create a new list of testable ideas; 4-when you are half-way done with your original list, repeat. Your customers and company will thank you for it.
I hope that you have found these three barriers insightful, if not at least therapeutic knowing that you share some of the same pains as others when taking on digital testing programs. Please let me know if you have experienced any challenges I may have missed. In the meantime, good luck overcoming these barriers and happy testing.