There are times in life where being lazy is a virtue. In some cases, laziness can be a desirable trait in a co-worker. Why? Because you can be sure that lazy person will find ways of accomplishing something that’s either hard or time-consuming by developing methods or tools that will effectively maximize productivity while providing the least effort possible. Sometimes, laziness translates into genius or pure art. After all (and bear with me), mankind has evolved and survived up until now mostly because we developed tools to make things easier: from the silex knife to the wheel to the lightbulb all the way to the device you are using to read this article right now. All these tools and inventions are focused on simplifying, enhancing or automating tasks we would accomplish manually. Because our bodies and minds have limits. Because the outside world is dangerous. Because… meh. Don’t feel like it. I’d’ rather be doing something exciting. I’m pretty sure that if laziness had to be translated to a math equation, it would involve parameters such as motivation, effort and return on investment.
So many hats
It comes as no surprise that this is true also when using a web/digital analytics platform. Most of the time, and unless you’re a full-time digital analytics professional like me, looking at analytics reports is neither your top priority nor an activity to which you can devote considerable amounts of time. In the case of small/medium businesses, you are very likely wearing many hats: CEO, salesman, stock management, shipping, marketer, customer support – and the list goes on. Long story short: you’ll be lucky if you can set aside an hour per week to look at your performance reports and hopefully conjure meaningful insights that will move your company along towards a brighter tomorrow (and more online revenue). So you don’t have time to use analytics? It’s OK, we’ll just tell analytics to send you information about your site’s performance when you need it. We will not quite reach the point where we can click and forget but we’re getting close.
Google Analytics custom alerts to the rescue
For the purposes of this article, we’ll be focusing on Google Analytics since that’s the analytics solution most people use on a daily basis / have the most exposure to. Of course, the methods I describe in this post can apply to other solutions such as Adobe, Webtrends, AT Internet and sundry. Let’s not start a vendor war quite yet. Let’s get right to it, shall we? Open Google Analytics and head to the left hand side menu and click Intelligence Events and then Overview. The interface will show you a list of the alerts that are detected by Google Analytics automatically, heuristically, based on past data. For instance, if your usual weekly traffic is at 1000 visits, a quick jump to 2000 will tell analytics that your site’s performance goes outside of the usual value brackets and therefore will register as an automatic alert. These alerts are broken down by a variety of factors such as acquisition data or geolocation. Clicking on details to the right will bring up a popup contextual view of the alert: Then you can click Go to report to see a more detailed view. Unfortunately that kind of data-driven alert is neatly hidden within the GA interface and does not warn you when any such alerts being triggered. You have to manually and deliberately log onto the reporting interface and go to Intelligence Alerts. So very low on the lazy scale. Don’t get me wrong, they can be very useful; they’re just not as interactive as the other type of alerts, which we’re going to introduce next. Let’s go back to the Intelligence Alerts > Overview screen and now let’s click the Custom Alerts tab that is sitting above the list of alerts. You can now see (surprise!) Custom Alerts which you have defined for your current view and that were triggered over the reporting period. But let’s create a custom alert in order for you to get a better idea.
Popping your alert cherry
Start by clicking the Manage Custom Alerts button. Yes, you are taken to the Admin section momentarily but fear not, we’ll be out of there soon. In this example, I only have 2 custom alerts defined for the current view – but you should see the alerts I have defined for some customer accounts 😉 But let’s go ahead and create your first Google Analytics custom alert. As I click the red “New Alert” button, I can now define how my custom alert gets triggered. In my case, I’m going to focus on my contact form conversion rate and I want to be alerted whenever the conversion rate for Goal #2 drops by more than 10% So at this point I just need to give the alert a name such as “Sharp drop in contact form conversion rate” then define on which view the alert applies. If you need the alert to apply to more than one account or property, you’ll have to define the alert again. You can also select other views for which the alert applies. Next, select the alert interval – in my case, weekly. If you need the alert to happen daily that’s fine too. Note that if you select a daily or monthly interval, you can trigger the alert either based on the data from the previous day or month or based on the same month the year before. Check the box that sends you an email alert otherwise there is not much point in calling it an alert, especially since the mobile phone option next to the checkbox only works for United States cell phone numbers. Don’t forget you can add alert recipients at this stage. It can be a good idea to add a distribution list alias for instance, such as [email protected] Great, now that we’ve defined the alert’s format, we can focus on the alert traffic type criteria. As seen below, you can select a traffic category from the green box: All Traffic, Country = UK, Browser = Firefox, etc. In this example below I’m only considering my alert in the context of “visitor country set to either UK or France” (using a regular expression). The best part about this green box is that you can select an advanced segment, either out of the box “system” segments or custom advanced segments. Finally, I need to select which metric is going to trigger the alert. Clicking in the blue box, I can select just about any metric available in Google Analytics, with the exception of Universal Analytics Custom Metrics and Calculated Metrics – but let’s wager this is going to be fixed soon 😉 In my example, I’m expecting an alert message whenever my conversion rate for my contact form drops by more than 10%. There you have it! Click Save alert and when the alert triggers, you will receive e-mail saying that… your alert triggered 😉 Go to your trusty e-mail solution and open the Google Analytics Intelligence Alert e-mail and voilà: I can click on the alert title in the e-mail to be taken to my Google Analytics to examine the alert and – hopefully – take action. The whole process of setting up the alert should literally take you under one minute and save you countless hours which you can spend being productive. Make sure the criteria you select matches your goals and you’ll be golden.
Best practices are other people’s practices
Now that I’ve shown you the basics of creating a custom alert, you can go crazy and create your own alerts with criteria that matter to you. Here are some of the super useful alerts that I use on a regular basis:
- Alerts for newsletter or campaign traffic > 0 sessions Ideal to debug a campaign launch and be alerted without staying stuck on the real-time report 😀
- Alert when hostname soandso.com starts registering sessions Also great for site launches and to indicate the switch between testing and production/live environments
- Alerts for conversion rate drops. Self-explanatory.
- Alerts for when e-commerce revenue drops
- Alerts for sharp increase in bounce rate / decrease in pages/session or time spent on site Denotes drop in website engagement.
And I could add a lot more! Again, these work for me and quite a few of my analytics clients but don’t take it from me: try your own alerts and see which help you the most. Of course your mileage will vary and with experience you will adjust the criticality cursor and your metrics increase/drop percentages will reflect your comfort zone.
TL;DR and conclusions
Great, you’ve made it this far. Setting up alerts sounded pretty easy, right? As mentioned before and because your job does not necessarily allow you to spend too much time in front of the Google Analytics console, set up alerts to save time and be alerted as soon as something goes well/wrong. At least, get notifications when your alert metric reaches acceptable levels of negative performance for you to take corrective action. Just remember that there is no such thing as real-time corrections. You will always lose/waste some time before you realize something is wrong and the moment when you can fix things. Other avenues for alerting include using the Google Analytics Real Time reports or its API to get to-the-second metrics. Be warned that those reports will keep you mesmerized (“Ooooh shiny…”) for the best part of an hour if you’re not careful – without yielding any valuable insights as to how to fix your website. Don’t get me wrong, I work in analytics full time and I too need to be pinged every now and then when things are going well or not so well. Too often the value of the insights is a direct factor of how much time you spend analyzing data – but it doesn’t have to be. If you don’t have that much time available, at least equip yourself with useful tools such as Google Analytics custom alerts that will save you a lot of time and effort while nurturing your lazy bone. And yes, it’s that much extra time you can spend doing stuff that matters more 🙂 In my next piece, I’ll talk more about other productivity features in Google Analytics, but first let me hear about you. How do you use your alerts? What had you been using until now? Do you have questions or feedback? Let me know in the comments!