TLDR: Identify trends in user behaviour by analysing the performance data on a time of month, week and day level. Implement time bid adjustments and automated rules to increase the efficiency of your media spend or increase engagement with dynamic time-sensitive ads.
Analysing the change of user behaviour over different periods of time has a strong potential to help:
- Save budget by bidding down in quiet periods
- Maximise last-click results by increasing prominence in peak times
- Make the most of your traffic by introducing time-sensitive offers
- Increase relevance and CTR by tweaking ad copy at certain times
So, in order to reap the benefits of time-based optimisation, you will need to dive into the data.
TIP: Make sure you only include the conversion action that you would like to optimise towards e.g. sales or leads. If you are looking at multiple conversion actions, analyse them separately in order to distinguish between trends relating to different stages of the funnel.
TIP: Use a long period of time (at least 1 year) for statistical significance and to avoid seasonality differences skewing your data.
Time of Month
- Use your historic data to optimise monthly budgets
Export your Google Ads data over the period you’ve chosen, number the days in a month and sum the results for each day from all the months. This way you will look at the “traditional” month of performance for your account.
TIP: Use conditional formatting to turn your table into a heat map and easily spot hot/cold periods.
TIP: Split the month into periods based on the trends you notice in impression volume, CTR, conversion volume, conversion rate and revenue if applicable.
Based on your findings, you can implement automated rules which bid up on a certain date each month and bid back down on another.
In our example, there are three periods of the month, the beginning of the month being a research stage (high on-site engagement), and the last 10 days of the month being the high-value sales period (highest revenue per sale), potentially due to users being paid around that time.
TIP: If you are bidding up by a percentage and would like to bid back down to return to your starting point. Use the formula 1 / (1+x) where x is the decimal form of the initial percentage bid change.
Time of Day/Week
- Control presence through ad schedules
Use the “day” and “hour of day” segments on Google Ads to look for trends in performance.
Implement a suitable ad schedule which bids up/down at certain periods.
TIP: If you are looking to preserve budget, use the ad schedule to bid down at certain times and let your keyword bids determine your presence at all other times. Bidding up and down at the same time for different periods makes it harder to review and evaluate the results of your strategy especially if you have other bid modifiers like devices, demographic, audiences etc.
Recently, Google made the ad scheduling more precise by allowing advertisers to bid in 15 mins increments as shown below:
TIP: This can be useful if you are aligning bidding precisely with specific events, for example, increasing bids at the exact time your TV ad is showing which could increase your brand searched.
- Engage users in the right moment with relevant dynamic ads
Use the findings from your time of day/week analysis to tweak your ad messaging accordingly thus increasing relevance and CTR.
This can be done using dynamic ad customisers that switch certain ad elements (e.g. headlines, descriptions) based on a set rule. Making your ads change in different time periods can make you stand out among the advertisers in the search results and also add a bit of character to your ads.
For example, you can have ads highlighting specific time-sensitive offers or countdowns.
The following ad showing on the days of the week when orders tend to be lower:
Or the following countdown ad on a Monday evening to boost sales:
Affinity Audiences are audiences available to target on Google Ads that consist of a group of people with similar lifestyles, passions and hobbies based on a series of online signals. Advertisers can utilise these [...]