Optimizing Your Paid Search Bids Around The Holiday Rush
Retailers often underbid PPC campaigns pre holiday and overbid when the rush ends. Columnist David Rekuc explains how to maximize holiday bidding.
The yearly holiday rush is often the happiest season of all for us marketers. Not only are customers more likely to buy, but demand and customer acquisition both typically reach their annual peak.
However, this surge in sales mean retailers are often pulled in several different directions, strapped for time, and more likely to fall victim to oversights that wouldn’t be an issue at other times of the year. One commonly mishandled phenomenon? A rise in conversion rates, specifically those attributed to paid search ads.
Without careful forethought, many retailers will underbid during the holidays and, later, overbid once the shopping rush concludes. This is a result of the way most marketing professionals calculate a “good” bid. Whether you use a bid rule, cost per acquisition (CPA) algorithm, or manual adjustment, bidding is almost always based on past conversion rates.
Here’s what I mean: on Cyber Monday, when we see the holiday season really kick off for e-commerce, most retailers will be basing their bids on conversion rates from the last 30 days. To make matters worse, on December 26th (or when you’re no longer able to process orders in time for Christmas), many retailers will base their bids on some of the busiest shopping days of the year, during one of the slowest months for sales.
Segmenting The Season
To combat this, I suggest most online marketers break the annual holiday period into three distinct phases. Bigger retailers working with larger sets of data can get more nuanced with these phases, but for most retailers, these stages will cover their seasonal bases:
Phase 1: Pre-Holiday Rush
This is the period in November leading up to your first day of significantly increased conversion rates. For some online retailers, especially those with a brick-and-mortar presence, this may be Black Friday. Others may not see the increase until Cyber Monday. In either case, you can use data from the previous year to find out when this windfall occurs for you.
Phase 2: Holiday Shopping
During this period, sales are up and so are conversion rates. Higher bids are still able to produce profitable sales. Pinpointing both the start and the end date for this period is critical, as it’s going to fundamentally change the way you treat your bids.
Also, make sure you’re comparing apples to apples; use a fiscal calendar so the days of the week line up and keep in mind that dates for Thanksgiving, Black Friday, and Cyber Monday will change from year to year.
Phase 3: Seasonal Cessation
The final period will be slightly different for every retailer. Though many will have strong sales days in January as they clear out inventory, almost every single one of us will see an immediate drop off in conversion right after the holiday rush comes to an end. Usually, the date your conversion rate plummets is when you have to put up a message that boils down to, “We can no longer guarantee delivery by December 25th.”
Conversion Rate Change
To start analyzing when these conversion rate changes occur for your brand, create a simple line graph of your non-branded search conversion from mid-November until January.
Once you have your graph, you should see where the traffic spikes up and where it falls off, and from there you should be able to segment your traffic into one of the three phases of the season we discussed previously. Now you’re ready to get to work.
Step One: Compare Phase 1 To Phase 2
What’s the increase in conversion? Is it statistically significant? Use a simple spreadsheet like this to ensure you’re dealing with statistically significant results (XLS download).
Once you understand what the increase in conversion is for you during the holiday time period, you can calculate your bids as you normally would. The trick is to then apply a percentage increase to your bids to account for the increase in conversion rates.
So, if you have a 30% increase in conversion rates from phase 1 to phase 2, you would calculate your normal bid and then increase it by 30%.
Step Two: Compare Phase 2 To Phase 3
When calculating what your bid would ordinarily be for this portion of the season, make sure you’re only using dates that occur in phase two during the holiday rush. If you use dates outside of the holiday rush to calculate your bids, you’re going to incorrectly calculate the difference in conversion rates between the phases.
Let’s say you determine that conversion rates fall by 50% when you put forth your end of season messaging. Calculate your bid based on the previous phase, and then reduce it by 50% to find your newly adjusted bid.
Average Order Value
A change in conversion rate occurs for almost all online retailers from November to January, but a change in average order value (AOV) is not always a given. However, it is important to remember that an increase or decrease in AOV can impact your profitability just as much as an increased conversion rate.
With that in mind, you can take a look at your AOV with the same bucketing approach described for conversion rates above. If there is a statistically significant difference, make sure you account for it in your bid adjustments as well.
The Countdown Is On
For most of us, the increase in demand will naturally increase sales during the holiday rush. But why stop there?
If you’re conscientious enough to adjust your bids upward during the right time of the year, you can see an even higher rise in sales volume than you would if you stuck with the status quo. If you remember to lower bids when the season slows, you can save resources for other important times of the year.
Finally, while you’re at it, don’t forget to take some time to think about the content of your ads (any killer sales to tout?) as well as holiday-specific keywords (think “gift ideas” and “product name sale”). Following these guidelines will ensure a very merry marketing season online.
Opinions expressed in this article are those of the guest author and not necessarily Marketing Land. Staff authors are listed here.