How to Use Google Analytics Annotations for Real-World Correlations

“Wow, look at that traffic spike on January 15! Is that the day we announced the new product?”

“Conversions were really down the weekend of Feb. 22-23. Maybe that was when we were doing our website maintenance.”

“We received lots of submissions about properties in Oakville three weeks ago. That’s when we launched our print campaign, right? Or was that after the town ratified the tax abatement?”

The Danger of Now

When something is occurring, the chronology is clear: it’s happening now. “Now” steadily fades with the passing days, weeks, and months, quite harmlessly for most people.

For Web analysts, however, this kind of oblivion can be fatal. How can we begin to measure the impact of any circumstance on website (or mobile app) behavior if we can’t remember when it happened?

This fundamental principle of Web analytics is often overlooked. We can’t blame Google Analytics. As illustrated below, Google Analytics makes it very easy to annotate reports by date.

01-annotations

Occurrences to Record

You should annotate any occurrence that may positively or negatively influence activity on your website, such as:

  • marketing campaigns
  • website design and content changes
  • website outages
  • industry developments
  • competitor activity
  • general news
  • weather
  • any other time-specific factor that could affect website behavior

After you create an annotation, a small caption icon will appear below the corresponding date. You can click that icon (or the small arrow tab again) to display that annotation, as well as any others for the selected time period.

02-annotations

List All Annotations in Your Google Analytics View

To list all annotations, as shown below, you can click Annotations in the admin panel for your view.

03-annotations

Additional Points about Annotations

A few additional points about annotations to consider:

  • To create annotations in a Google Analytics view, you need only basic Read & Analyze rights. So, anyone who can access a view can annotate it.
  • The default visibility setting for annotations is Shared (as shown in the first screenshot above). If you don’t want anyone else to see an annotation you’re creating, select Private.
  • Annotations are replicated among reports within the same view. For example, if you create an annotation in the Landing Pages report, the caption icon will appear in the All Referrals report.
  • Annotations are not replicated among views. If you and your team work with multiple views for the same Google Analytics website or app property (which in many cases is a good practice), make sure you’re clear about which view will house all shared annotations.

Maintain an External Timeline

In addition to (or even instead of) creating your timeline directly within Google Analytics, you might want to record events in a separate spreadsheet so you can categorize, color code, or add one or more columns for status and follow-up notes. For example, you might want to note that two months after a modification to your website’s top navigation, you will check specifically for changes in pages/visit, goal conversion rates, and other relevant metrics.

The advantage of keeping the annotations within Google Analytics is that they’re in context. With the caption icons appearing directly in the reports, it may be easier to connect your data with the occurrences you have recorded.

Monitor Metrics Proactively

Annotations can help you retroactively make sense of your data. If, today, you notice there was a sharp increase in e-commerce transactions two months ago, you might be able to attribute that change to an annotation you made at the time.

However, you can take a more proactive approach to the annotation process by purposefully monitoring specific metrics that the annotated occurrences may affect. The table below lists several examples.

google-analytics-table

*This actually happened to my website this week. I’m monitoring the immediate impact; and a few months from now, I’m going to rely on the annotation to explain the decrease in visits.

Use Intelligence Alerts to Complement Annotations

As a complement to your annotations, you can configure Google Analytics to send custom Intelligence Alerts by email (or by text to United States phone numbers) when a metric threshold is reached for a given time period.

For example, after you feature a product on your home page (using one of the items in the table above), the following Intelligence Alert will generate an email if traffic to your product page increases by more than 10% compared to the previous week.

04-annotations

Keep in mind that the custom Intelligence Alert will be in effect until you delete it, so any factor can trigger it, not just the one you annotated. In fact, quite independently of annotations, Intelligence Alerts provide a useful mechanism for actively monitoring key metrics on your website.

Correlate Your Data with the Real World using a Chronology

Whether you maintain a chronology inside or outside of Google Analytics, and whether you create corresponding Intelligence Alerts, a chronology is not a recommendation, but rather a must. Without a chronology, there will be no correlation between your data and the real world.

You can’t be effective in Web analytics or marketing optimization if you don’t maintain a coherent, real-world timeline that helps to provide the narrative behind your data.

As with most things in life, good analytics begins with the basics.

About the Author: Eric Fettman is Analytics Trainer & Coach at E-Nor. He is also the developer of googleanalyticstest.com, a free resource for Google Analytics training and GA Individual Qualification preparation. He recently authored the Google Analytics Universal Guide: Best Practices for Implementation and Reporting and the 36-Point Google Analytics Implementation Checklist). He regularly provides GA pointers at gatipoftheday.com.

  1. Great Article. All Tips is Really Helpful for every Blogger. Thank you very much fro Sharing With us. Keep up it.

  2. Thank you for the insights here…

    As a practice, sharing knowledge is one of the most beneficial traits of successful organizations. Hopefully internet marketers from every segment will put these insights to use.

  3. Very nice article. Love it and Keep it up….

  4. Thanks for this Google Analytics Insight….. I was actually not aware of this…. Will use them for sure…

  5. You mention “marketing chronology” in an offhand way. Would love to hear more in-depth about this topic. That’s something I’m doing all the time for clients and it ends up in Excel–very frustrating. There’s got to be a better way….

    • Gretchen, I’m sorry not to be in a position to advice on dedicated systems for this. A spreadsheet with some drop-downs and color coding can be very effective if you want to maintain a marketing chronology externally. To measure the effectiveness of specific marketing campaigns within GA, you can start by making sure that you have used campaign parameters on all applicable inbound links (and that you have Autotagging enabled for AdWords). Clean attribution, a comprehensive chronology, and the metrics that GA provides should cover most ground in measuring the effectiveness of marketing campaigns.

      Not sure if this answers your question directly, but hope it helps.

  6. I think this is a great article. Thanks for sharing..!! Great insight. Cheers..!!

  7. Nice post Eric. Another tip – you can add annotations in advance by creating new annotations in the admin area, under the current view > personal tools and assets > annotations and use the big red button to create a new annotation. This allows you to choose dates in advance. When the date arrives, the annotation is already in the timeline.

    This technique is very handy for pre-populating all the activities planned in a campaign schedule. Yes they may change as the campaign occurs, but they can be edited. Using this method is easier than waiting for each date and activity to add them in the main reporting area.

    cheers
    Jon

  8. Great post. I really like the idea of implementing “real world correlations” in GA. It reminds me of Google Trends feature where you can see the impact of external factors on keyword trends. Here I wrote a short post about this: http://goo.gl/fG3mUa

    If with Google Trends we can connect external factors with the “macro world of web searches”, Google Analytcs annotations could do a similar job allowing us connecting external facts with our “micro world website” and perform interesting analysis. I think annotations is largely underutilized feature in GA. Let’s go experimenting!

13 comments

Please use your real name and a corresponding social media profile when commenting. Otherwise, your comment may be deleted.

← Previous ArticleNext Article →