Along with being actionable, your Google Ads performance reports should be grounded in the overall context of your account. Consider the world that your account is operating within.
Detect anomalies in your Google Ads account by being aware of baseline performance
Anomalies in performance, either good or bad, help you decide what actions you need to take. In order to detect outliers, you’ll need awareness of what standard performance looks like. It’s also important to understand the significance of those anomalies - are they one-day blips, or indications of a larger trend?
The more history you have, the more familiar you should be with typical performance. There are natural ebbs and flows to performance due to things as simple as day of the week.
A particularly useful visualization to use when identifying baseline performance is time comparison on a line chart. Use it to focus on your most important metrics and see how performance is trending.
Once you identify anomalies, see what secondary metrics might be changing along with conversions and further understand why performance is trending differently from the baseline you’ve established.
Tip
Remember to factor in time lag when you’re establishing baseline performance for conversion metrics. If it typically takes seven days for users to convert, wait at least seven days to review performance.
Familiarize yourself with your performance, especially in the context of what you’ve done in the past and whatever external climate you may be in. The earlier you spot anomalies in performance, the quicker you can take advantage of new opportunities or fix poor performers.
Tip
The top movers report will detect large changes in your account, some of which you may expect based on seasonality. Keep an eye on it to maintain awareness alongside your own viewing habits.
Choose the right time periods when analyzing Google Ads performance
Google Ads accounts are almost like living, breathing entities. They interact with the world around them and are affected by things like seasonality and external market pressures. Don’t overreact to something that could actually be a result of a temporary, external event.
Although it’s impossible to control for all changes at any given time, there are things that pop up time and again:
| External forces | Intra-company considerations | Google Ads account changes |
|
|
|
Awareness of the different external or internal pressures that could be affecting your numbers will allow you to choose the appropriate date ranges for running your analyses. As an example, if you’re on a monthly budget, week-over-week comparisons won’t be especially useful when you’re ramping spend up or down the first and last weeks of each month.
A particularly valuable analysis to help understand external pressures is Auction insights. When you segment Auction insights by month you can get a view into what other advertisers are doing and how it overlaps with your own internal account changes. Use this analysis to stay on top of your competitive landscape. You can keep an eye on their aggression via bid changes, and you can also make changes based on who’s entering or exiting the auctions that you care about.
Ensure that your ratios have the right amount of data to be legitimate
An especially important piece of establishing context for your analyses is being confident that the insights you’re generating are legitimate and reproducible. Are your numbers stable, or are they inflated/deflated because you haven’t collected enough data points? This is going to come down to two key components:
- Do you have long enough time periods to have substantial data?
- Have you incorporated supporting data into your analyses? (Things like time lag for conversions)
It’s important to use analyses to drive action, and it’s equally important that your analyses drive the right actions.
Early returns can be misleading. By waiting for enough data, you can be confident that the changes you make are corroborated by the facts. A new keyword may convert on the very first click that it drives (giving it a 100% conversion rate). That doesn’t mean that it’s now one of your top performing keywords. Sustained performance is what will determine that keyword’s long-term value.
For example, imagine that you’re testing out bidding a handful of your keywords to the top of the page for the first time. You’ll want to wait on a couple of things: until you have enough clicks to make an informed decision, and also until your standard conversion time lag has passed before reviewing results.
Tip
Keep in mind that conversions are back-dated to the day that the click happened. For example, if someone converts on May 21st, that conversion would be recorded when the ad was initially clicked on May 6th.
Wait until you have the right information on hand before determining the final outcome.
