So you’ve launched a few campaigns on Google Adwords in order to increase sales on your website. But you notice that your campaigns are limited by budget and your cost per acquisition (CPA) has increased, while the number of conversions has remained the same. What is the best strategy to optimize your campaigns’ performance while having a target CPA?
In this article, we’d like to share two of our strategies that have proven successful with our clients’ campaigns:
- Using ad scheduling based upon on the number of conversions and the time of the day.
- Setting bid adjustments based on the time of day.
Ad Scheduling by Conversions and Time of Day
Website sales can be subject to strong seasonality. Therefore, clicks, conversions, and cost per conversions (CPC) can also vary significantly depending on the season, day, and even the time of the day. Here are steps we’ve taken in the past to better optimize our campaigns.
For a client we needed to create an effective strategy that would fix the problem of having the CPC higher than the target CPA with a negative ROI. After exporting the available data from Google Adwords we created a pivot table and chart to display clicks and conversions based on the time of day.
Figure 1: Google Adwords Export
From our data set we were able to see that the number of conversions was at 0 from midnight to 5 am, which is when the CPA is higher. This struck our curiosity, making us want to go deeper in our analysis.
We used data analysis software (SPSS) to test if there was a statistically significant difference between two periods: midnight to 5 a.m. (Period 1) and the rest of the day (Period 2); we performed an analysis of variance (ANOVA). The results showed that there was a statistical difference in both the average conversions and the cost per conversion, with less than a 0.1% margin of error.
By examining the conversion rate by time of day we were able to isolate a key indicator of online user behaviour. In this case, our clients were more inclined to convert during the day (5 am to midnight) than early in the morning.
As the graph above indicates, we had been spending the budget during Period 1 while the ROI was nil; these costs increased the CPA. As a consequence, we decided to schedule our ads for the first campaign between Period 2 (5 a.m. and midnight). The objective of this strategy was to decrease the CPA and focus our budget on the hours when the ROI was positive.
Bid Adjustments Based on Time of Day
In another campaign we had a more constant conversion rate throughout the day. However, there were significant variations in the CPC. We had determined a target CPA of $45 with our client, which had been higher ($55) during the times between midnight and 7 am.
Using SPSS, we confirmed that there was a statistical relationship between CPC and time of day. Therefore, we decided to set a bid adjustment based on that indicator. We were able to decrease the bids by 22% between midnight and 7 am.
After a one week follow-up, we observed a significant amelioration in our campaigns’ performance. More precisely, there was an 18% increase in conversions, 18% decrease in cost per conversion, and a 15% increase in conversion rate! This is solid proof that ad scheduling along with bid adjustments by the time of day are both beneficial strategies for optimizing PPC campaigns.
This post was originally written by Mehdi Fehri.
