AdWords is now Google Ads. Our new name reflects the full range of advertising options we offer across Search, Display, YouTube, and more. Learn more

5K members online now
5K members online now
Improve your Google Ads performance and boost your ROI, CTR, and Quality Score
Guide Me

Using automated rules based on same day data

Follower ✭ ☆ ☆
# 1
Follower ✭ ☆ ☆



Does anyone have experience of using automated rules using same day data?


We have used it for some time now and find that the rule for changing the max CPC based on same day data (say cost per converted click) is not very effective. The actual CPC usually varies a lot if you check the report by "hour of the day" segment. When the rule kicks in (say to reduce max CPC if cost per conversion is high) and we track the actual CPC by the hour, it is almost impossible to see a correlation (actual CPC variation remains high and average does not seem to come down after the rule kicks in).


Are we missing something? Is it that same day data is not a reliable method for making changes in max CPC?





1 Expert replyverified_user

Re: Using automated rules based on same day data

Collaborator ✭ ✭ ✭
# 2
Collaborator ✭ ✭ ✭



The main problems with obtaining precision from your setup are :


a) The fact that the automate rules can be triggered at any minute during the hour which you specify . This is explained here under point 4 after expanding "Step by step guide ...... " :  



Did you know...

You can specify the hour of the day for your rule to run. A rule can start running at any time within the hour you select."


2) If you look at today's data to analyze if a rule did what you expected, there is a delay of around 3 hours in reporting in the interface. So to analyze better the effects of the rule you should look at data from at least 4 hours ago or later.


"In most cases, your account statistics (such as clicks, conversions and impressions) are delayed by less than three hours. For example, the total of all clicks received by 6 p.m. on Friday will be available by 9 p.m. Friday night, local time."


3) The segmentation you are analyzing,  report Time >>Hour of day,  is available only at the campaign or ad group level, so an actual CPC for the entire ad group is an average between many keywords. Maybe the automated rule just changed 5 values of MaxCPC from 20 keywords so the impact of running the automated rule is dilluted.




" Is it that same day data is not a reliable method for making changes in max CPC?"


I think it is reliable, but reporting is too slow to measure the effects.


Have you tried looking at the logs to see how many keywords have a lower MaxCPC when the rules are triggered ? Looking at the overal actual CPC many not be a good indication, as the actual CPC is almost always less than the MaxCPC. So you could have a lower MaxCPC but still a higher actual CPC depending on the competition at that hour.


An example : 


Let's say that your MaxCPC is 5 $ and your actual CPC is 1$ for hour 14 , When you look at the log the automated rule has decreased the MaxCPC to 3 $ but your actual CPC is 1.5 $ for the next hour 15:00.

Re: Using automated rules based on same day data

Top Contributor
# 3
Top Contributor

Hi am, further to Adrian's reply, I would suggest that normally reacting to same day data in AdWords is not a reliable or advisable optimisation method.  Unless your product/service converts at a very high rate - many conversions per hour, for each Keyword - it's usually better to look at longer term trends for CPC values and then, if necessary, alter these using Ad Scheduling, Location bid adjustments and/or mobile bid adjustments.


There are a few reasons why reacting in the short term is inadvisable.  Firstly, as Adrian mentioned, rarely is the short-term data reliable enough to make decisions.  Not only is there a variable reporting delay, but "Present day" has no consideration for which day is being referred to.  Most AdWords Accounts will see variations in performance day to day and these can often be very predictable so if Sunday is a poor performance day and Monday much better, by dropping the CPC on Sunday you're actually harming performance on the Monday.  Secondly, in most cases conversions don't all happen immediately after the ad click.  Depending upon the nature of the product/service, conversions could occur days or weeks after the initial click.  So, looking at one particular Keyword on one particular day might suggest it's not hit it's CPA target, but in 2 days time, when the visitor converts, that conversion (in AdWords) will be attributed to the first click.  The "Time Lag" report in Search Funnels will give you an idea of how long your typical conversion path is.


Unless you have sound reasons for trying to react in such a short period, I'd really recommend planning your rules to run over longer periods of "Last 7 days" or even "Last 30 days".



AdWords Top Contributor Google+ Profile | Partner Profile | AdWords Audits