The truth is that even the best-planned, most expensive search marketing campaign can produce unexpected or unintended results. You have to build agility and flexibility into your plans that are closely connected with the tail end of the SMARTER goal-setting process—evaluate and reevaluate. Those two steps are in there specifically to manage risk. We will talk more in depth about risk management and key troubleshooting steps for any search campaign in new series of articles.
Join me on the journey where we’ll reveal the solutions to common data ailments such as data disappearances and anomalies, low traffic volume, low-quality traffic, high-quality traffic, and low conversions.
Let’s kick off with a problem that can happen any time with search accounts:
Your conversion data has a sharp decline from expected performance.
Conversion data is some of the most important information in a search marketing campaign. A big drop in data isn’t necessarily a bad thing because it can show marketers how to adjust and iterate on a current account. But you’ll want to figure out what caused the drop off as soon as you can. By taking a methodical, well thought out approach you can usually find the cause before the problem results in lost revenue. Below are six steps to effectively diagnosis the issue.
1. Revisit the marketing goal of the campaign. This includes making sure you know what actions you want to track as conversions. Are you properly capturing these actions within your analytics tool or has something changed in your implementation? Do you have a clear understanding of what other events might be happening in your company that affect your paid search accounts? I like to keep a promotional calendar handy just for instances like these. It is easy to assume a dip in conversions automatically means that something is wrong, when in reality it could be something as simple as a promotion has ended. Making sure you’re clear on these points can help you identify what’s really “off” and what might be a natural rhythm in the data as you start diving into the reports.
2. Determine when exactly when the drop-off started. You can do this either by looking at daily conversion reports, or if you are an Adobe Analytics customer, check out the new anomaly detection feature. When you are ready to pull your conversion report, be sure to pull a large enough slice of data at a daily granularity to determine exactly when the decline began. See if you can narrow down the root cause of the issue. Is there a specific account, campaign, or ad group that is the troublemaker, or is the decline something that spans the entire portfolio of accounts?
3. Examine your campaign-level settings. Is the budget depleted? Are the location, device, and network targeting settings where you want them, or did it somehow get changed? Next, drill down to the settings and data at the keyword and ad level, and look at items such as keyword and ad status. Was there an increase or decrease in CPCs or position, and quality score?
4. At the ad group level, see if you can determine which ad group is showing the decrease in traffic and/or conversions. If you know this, you can then dig deeper into keyword-level performance. Are the keywords driving traffic to the right page, and is that page still active? Look at your high-converting keywords. It’s possible one or more of these aren’t converting anymore. We will discuss what do with these finicky keywords a few weeks. Here are some other things to look for: Have any ads become disapproved or new keywords introduced? Maybe some negative keywords are now blocking traffic or keyword additions to other ad groups are receiving the traffic instead of the original ad group/keyword combination.
5. If all the items we have just discussed are as expected, the next thing you’ll want to do is run a change history report for the relevant timeframe. Alterations to the website and pricing or product changes can all cause conversion data to skew. So can changes to the range of keywords you’re using. If more than one person is using the account, the change report can show you who made a particular change.
6. When I use a tool that relies on tracking, such as Adobe Media Optimizer (AMO), I also look at ratio of search engine clicks to media instances. This metric refers to the number of times the AMO code is loading on the website, and it should always match search engine clicks within 10 percent. The 10 percent accounts for issues such as someone not loading the page properly or backing out before the page loads. At that point, you will need to dive into your analytics implementation to determine if anything is amiss there.
Additionally, your data drop-off could also be caused by landing page URL errors such as typos, 404 errors, or redirects. Keeping your technical SEO in top order can prevent this kind of problem from cropping up.
As a final note, it can be helpful to have some strategies for how to interpret line data in context as you look at all these reports. I recommend this article on Search Engine land.
So there you have it: an overview of my process for troubleshooting a decline in conversion data. Stay tuned for next week’s article on what to do if your data completely and unexpectedly disappears.