The Yule Blog: Silver Bell Curves and other Overlooked Holiday Metrics
Published: November 18, 2013
Author: Jay Stampfl
Holidays are upon us! And if you are running campaigns for ecommerce clients, I am sure this has not gone unnoticed. Everyone is aware that holidays are a period of big numbers. High impressions, high CPCs, high CVRs.
But I want to move a little deeper into the analytics. I have seen pretty consistent patterns that exisit outside of these surface-level observations. While it is intuitive that conversion rate increases, it is perhaps less so that average order value (AOV) increases along with this metric. This is important if you are working off a CPA model and not taking into account revenue. I recommend modifying your CPA goal based on the percent that AOV improves.
Our last column shows the percentage of multiple conversions per click over single click. The higher that number, the more likely someone who has purchased will come back to buy again. Note that this number peaks in December. There are two main takeaways:
1. If some has purchased recently, they are more likely to buy again than at any other time. You should incorporate this knowledge into your remarketing and RLSA campaigns.
2. Instead of buying all at once, people are leaving the funnel and then reentering—what is the deal with that?
The answer stems from the difficulty of shopping for others and the need to hunt for sales, specials, etc.
This is the “What should buy?” and “How can I can find a deal?” user. This is the person who will research for the best deal and then make big purchases. This is the person who buys a gift, shops around on other sites, and then swings back to make purchase number two.
With all this hopping around, we would expect that there would be more ad clicks per conversion than usual. Well let’s pull those numbers…
And that is exactly what we see. The percentage of conversions that occur with only one click is lower during the holidays. This isn’t the only interesting behavior that we see emerge from the sale pressure. We also see that people make their decision to purchase faster than they would normally in order to catch the sale or have the gift ship in time for Christmas.
In this chart, you’ll see that the number of users who purchase after 12 days (bottom figure on the chart) drops from 24% to 14% during the holidays. So while people are clicking more, researching more, and bouncing around the internet, they also make decisions in a shorter amount of time. It is also important to note that while the number of conversion after day 12 drops, the revenue associated with those conversions is almost equal to the pre-holiday segment. This means that people take more time to make higher-value purchases.
Hopefully I have pointed out some interesting metrics that you can use in your holidays efforts. In these numbers are strong implications for how you bid, what creative you use, and when you push in remarketing/RLSA. All these numbers are taken from real ecommerce accounts, but either way it is important that you are tracking these metrics in your account and are aware of how they change during different seasons.