If I asked if you wanted an orange to eat, would you say yes? What if I asked you if you wanted a fresh navel orange, hand-picked from the orange fields of California? You’d be more likely to accept the offer, wouldn’t you? Information is everything, and having all the information to make an educated decision is one thing that lies at the heart of any successful endeavor.
The same is very much true when you are collecting data about your website using Google Analytics. Every day as I try to dig into new areas of Google Analytics, or try to mix and match Google Analytics data in creative new ways, I always discover something new and amazing. This actually happened this month as I was taking a closer look at other dimensions that are available inside of Analytics. Two of the most interesting — and in my view the most valuable — dimensions that are available are “Hour” and “Day of the Week Name”.
In this article, I’m going to explain what those two dimensions tell you, and how you can combine them with other data to reveal a lot of amazing things about your website traffic, reader behavior on your site, and even how to better optimize the website when you publish certain types of articles or other content on your site. Just like on Twitter or Facebook, timing is often everything. The same is true about when you publish new content to your own site, and these Google Analytics dimensions can help you optimize that.
Understanding Hour and Day Dimensions
As people visit your website in greater numbers, there are certain patterns that will emerge if you know where to look. I covered a whole lot of those in our MUO Google Analytics Guide, which I highly recommend. While you can certainly view things like pageviews or bounce rates by the hour or week in the charts, that only tells you overall traffic trends by the hour or by the week.
What the Hour and Day dimensions give you access to is longer term patterns. Those dimensions will take traffic, social data, and any other metrics available on Google Analytics, and then calculate overall patterns over a longer range of time, and in my opinion, the longer the range the better. The more data you use with these dimensions, the more likely you’re going to identify extremely useful patterns that can improve your odds of success.
But first things first. Since you’ll be looking at hourly data, time zone matters a lot. So first check the time zone setting in Google Analytics by going to the Admin area, and clicking “View Settings” under Profile.
Under “Time zone country or territory”, you’ll see your current time zone setting. This is really only important if you’re looking at a website with a publishing schedule set up in a different time zone. You’ll need to correlate the time zone of the Analytics data with the specific time zone you’re thinking about to publish certain types of content.
Working With the Hour Dimension
The Hour dimension can be accessed by building a custom report. You access those by clicking “customization” in the top menu on Google Analytics, and then clicking the “New Custom Report” button. When editing the new report, you’ll find “Hour” available under “Other” in the dropdown for the Dimension field.
Even just a general Pageviews report based on the Hour dimension can tell you a great deal about your visitor patterns. Again, examining this data over a longer time period is best for finding patterns, so I’ve made my own analysis span over six months. What the data reveals below is that my highest traffic time periods are the afternoon, from 13:00 to 17:00. And then, interestingly, there’s another surge very late at night, around 22:00 (probably after the kids have all gone to bed and parents are getting online to do some research).
This is really only the tip of the iceberg through. Moving beyond pageviews, just think about what other metrics tell you when you look at it in terms of hourly patterns. For example, at what time of day do you seem to have the highest exit rates from your site? Mine appear to be worst just before and after midnight.
What time of day do you get the most new visitors? Using the % New Visits metric by hour, you can see when you have the highest rate of new visitors. In my case I get maximum new traffic before and after midnight, and well into the early morning. It’s very common for a high rate of new visits from places like Google Search to correlate with higher exit rates, because a portion of those new visitors are “speed-clickers”, just clicking through to see if your article answers the question or interest they had when they conducted the search.
These are just a few examples of hourly trends you can use to learn more about visitor behavior and traffic patterns, but if you’re creative with the metrics or combination of metrics you look at, the possibilities are really unlimited.
Working With Day Dimension
If you want to look at larger patterns, an awesome one to explore is the Day of the Week dimension. You can find this one under “Other” in the Dimension drill down box when you’re creating your custom report. The name of the dimension is “Day of the week name”.
Like the hourly report, the day of the week report by pageview is a quick way to see what day of the week you get the most traffic. Surprisingly, my own blog follows a pretty clear pattern from the start of the week having the most traffic, to the end of the week having the least. However, this definitely isn’t always the case. I’ve seen other sites where the middle of the week is clearly the busiest. It all depends on the niche or subject matter you cover, and when the people interested in your content are most likely to be spending time online.
Again, like with the hourly report, the weekday report sorted by exit rate can tell you what day of the week people tend to leave your site upon initially visiting it. You might think this should directly correspond with traffic levels, but that’s not the case. As you can see here, the entire weekend represents a time period with high exit rates on my blog, while more people tend to hang around at the start and at the end of the 5-day work week.
However, there’s a lot more that you can do with this beyond just standard metrics like pageviews and exit rates. What about learning what days more people share your articles, or what days your site seems to struggle more with page load speed?
Other Patterns You Can See
There are so many things you can analyze about your site using the day or hour metric that it’s a bit mind-blowing to consider it. There are a few more examples I wanted to share just to whet your appetite. When you have your report listed by day of the week, you can drop down the Secondary dimension box and select “Visitor Type” to see the specific days you have more returning or new visitors.
Doing this reveals that I get the highest number of new visitors at the very start of the week, from Sunday through Tuesday. Those also happen to be the days I get the highest return visitors, showing that both numbers correlate to overall traffic patterns. Nothing surprising there.
However, if you do something like sorting the report by other metrics, you can extract some pretty interesting data. For example, if I sort by average time on page, I learn that returning (loyal) visitors spend the maximum time on my pages on Thursdays, Sundays and Tuesdays, while new visitors tend to stay on the site the longest over the weekend, from Friday through Sunday.
This information can be really helpful for targeting those types of readers during the correct days of the week. Offer your loyal readers content that you know they’ll enjoy most on Thursday, while offering new reader-targeted content that might encourage them to become loyal readers, over the weekend.
Another example might be selecting the “Actions per Social Visit” metric, which could reveal the days of the week you’re getting the most social activity on your articles — i.e. the days when people tend to share them the most.
Running through this custom report, I’ve discovered that I get the best share rate on Thursday. This provides me with an opportunity to not only target readers that are more active on social networks during that particular day (posting social network promotions), but it also shows you what other days you may need to work on for increasing social activity on the site.
One last example might be checking how your web server is performing during the week. You can see this by choosing the Avg. Page Load Time metric on the custom report editor.
Sorting by this metric in the “Day of the Week” report reveals that my server really strains on Fridays and Sundays. When this doesn’t correspond directly with traffic, you may want to explore other causes for it, as there could be solutions that could improve page load on those days. Page load has SEO implications for your site, so it’s an important metric to follow (and improve).
You might even break down the page load metric by the hour, to see what time of day you seem to be having the most issues. Again, this should correspond with traffic.
When it doesn’t, it’s a clue that you could have other issues going on with the server during those times — maybe you’re on a shared web server and other sites are bogging it down during those times.
As you can see, hourly and daily dimensions can help you break down your website metrics into much more meaningful data. By mixing and matching that data, you can better understand a whole lot about how and when traffic is arriving on your site, and how readers are behaving at different times. You could explore things like the particular days you get specific referral traffic, the hours of the day certain sections of your site tend to perform the best, and so much more.
Do you use hour or day metrics when you explore Google Analytics? If so, what sorts of things have you analyzed and what did it tell you about your site? Share your own experiences and ideas in the comments section below!