Data is the heart of strategic decision making and can provide insights for your business. Below is a checklist for making data-backed decisions.
-
Ensure your sample size is large enough.
This prevents your data from being skewed.
Secures a normally distributed population.
-
Make sure your testing is long enough.
1-2 weeks is considered the minimum amount of time for testing, preferably a month.
The longer the testing, the more insights for your business.
-
Use appropriate attribution modeling.
First Interaction: credit goes to first channel
Last Interaction: credit goes to last channel
Linear: equal credit
Time Decay: more credit to channels closer to conversion
Position Based: credit divided between first and last
-
Make sure conclusions are unbiased.
Do not let opinions cloud your judgment.
Always trust the facts from testing and research.
Hi Beth,
I really enjoyed the title and excerpt of your blog which made me want to read your blog! I like how your blog is very simple but very straight to the point. I like how for every heading you had a short but sweet explanation! I agree with everything you have said and I believe it is super important not to make conclusion biased. We have two based our conclusions on our previous research and data. Good job!
Beth,
I have to admit, the excerpt encouraged me to give this post a read, and I’m glad it did. I loved the checklist, the information was relevant and concise, and I was entertained by your use of GIFs (especially the one from The Emperor’s New Groove haha).
Hi Beth,
Thanks for the awesome post on digital analytics. I think you did a good job of summarizing what we learned in class. Your point about making sure the sample size is large enough reminded me to do this whenever I am creating a survey to ask people. I think you did a good job of keeping it short and informative.
LOVE the layout on this and your use of GIFs!!! So fun and interesting to look at!