Where Did Users Come From?
A good first step in reporting your website’s performance is explaining where your users are coming from. How did they find your site? People who searched for your products and clicked an ad will interact with your site differently than those who found you via a post shared on social media. Your marketing team will want to focus on acquiring users from channels that drive engagement, but to start, they need to know how traffic breaks down by channel in the first place.
In addition, you will also want to report if the channels driving users to the site have changed over time. Include historical data to show if specific channels have trended up or down over the past few months, the past year, etc.
Below we analyze a site which showed a decrease in traffic over the last few weeks. By tapping into Megalytic, we can create an effective report that will clearly explain what happened and why.
To start, we’ll use Megalytic’s Traffic by Channel widget to divide sessions using the “channel” categories from Google Analytics, which include high-level groupings of traffic sources such as Social, Organic Search and Paid Search. Note that in this example, we’ve customized Google Analytics to show generic and brand paid search terms separately. [You can set this up for yourself by following these instructions.] Once configured, we’re able to segment sessions that included searches where your company or product name were used, as these users have existing familiarity with your brand.
After creating the widget, we customize it by selecting a line graph for the chart type and choosing the channels we want to display from the Series dropdown.
As you can see above, we have chosen to compare Organic Search, Generic Paid Search, Brand Paid Search, Display, Referral and Social channels in our graph. Once we have selected these and closed the widget editor, we can view the final result.
This chart puts into perspective how each channel contributed to traffic as a whole, as well as the change in traffic over time. We can clearly see that Organic Search had the highest volume of traffic but also the largest drop over the past few weeks. We also see that Generic Paid Search, while dropping along with a budget reduction, shows a less significant drop than organic, while Brand Paid Search remains relatively steady. Based on the seasonal nature of this business, we find that general interest in services, and thus traffic, naturally decreases around winter, explaining the drop in volume.
To help the readers of this report understand the significance of this chart, we should summarize these insights in a Notes widget that we place directly under this chart.
How Well Are Users Converting?
Beyond looking at traffic, your report should include analysis of how well users are converting, whether you define a conversion as a content download, newsletter subscription, or ecommerce purchase. In addition, you should compare historical data to report how conversions are performing over the long term.
Continuing from our previous example, we should next report on how well the traffic coming to the site is continuing to convert during the expected seasonal downturn in traffic. To do that, we’ll add a Conversion by Channel widget to our report and customize it to focus on the top three channels.
Within this widget, we’ll look at conversion rate to see what percentage of sessions result in conversions over time. In the options, we select a line chart for chart type and Conversion Rate for metric. In addition, we choose to show Generic Paid Search, Brand Paid Search and Organic Search from the Series dropdown, as these are the three major channels driving conversions.
The paid search channels – both Brand and Generic – show conversion rates declining slightly, as expected as we head in to the off-season. The spike in Branded Paid Search conversions during late October should be explained in a Notes widget. Since the amount of Branded Paid Search is low – a small number of conversions can skew the conversion rate.
Most significantly, the conversion rate for Organic Search traffic has dropped from almost 3% to less than 0.5%. Your report should certainly call this out in the notes – as it might not jump out at the reader from this chart.
Including the trend lines for conversion rates makes for a compelling report, but we don’t stop there! Because we’re savvy marketers, we want to go beyond just looking at the past few months, comparing recent data with that of similar periods from last year. This is especially important for a business that sees seasonal changes in activity.
To ensure that we are looking at a data on a big picture level, as well as comparing change over time through the previous year, we can use Megalytic to add a table that shows performance by quarter, starting with the Metric Table by Period widget.
First, using the date range dropdown, we select the end date for the range we wish to analyze and select 3 Months for the period.
Next, we use the widget options to select Users, Sessions, Completions and Conversion Rate as metrics to show in the table.
In the final result, we see a table comparing these metrics by quarter over the past year, allowing us to quickly and easily compare recent data to historical data.
The table shows the third quarter (July – September) of this year witnessed an improvement in Users, Session and Completion compared with the same period in 2013. There was even a slight increate in the conversion rate. In our report, we should call this out, to emphasize that this year the website traffic is significantly outperforming last year.
In addition, our report should note that during the fourth quarter (October – December) of last year, there was a significant drop in traffic – similar to what is being experience this year. Point out in the report that this year’s drop is consistent with last year, and not a cause for alarm.
How Are Users Engaging with Your Site?
While conversion metrics that relate directly to ROI should constitute the core of a report on website traffic analysis, you should also look at how people are engaging with your site in general, as well as how this engagement changes over time. Generally, if people are spending time on your site and looking at multiple pages in a visit, they are more likely to be interested in your services than people who bounce off after a ten second session.
Pages/Session and Average Session Duration both relate to engagement, showing how many pages people look at and how long they spend on a site. By graphing these two metrics together, you can show trends for increased or decreased engagement levels. To do this, we’ll select the Double Axis widget.
From the Series dropdown in the widget options, we choose Pages/Session and Avg Session Duration and apply the selection to see the final results.
We expect these two engagement metrics to move in tandem – and that is what is happening, with the Avg Session Duration looking pretty much like an amplified version o the Pages / Session line. Both metrics indicate that engagement on the site peaked in mid-September, dipping down in November only to rise back up toward the end of the month. From this information, we can zero in on periods of time, based on marketing efforts or content pushes, when the traffic showed higher or lower engagement. It might be instructive to examine campaigns run during this period to see if any particular campaign drove traffic with an unusually high (or low) level of engagement.
Megalytic offers a number of reporting features to help you show your client or boss how a website is performing and point out any concerns they may need to address. If engagement is trending downward, a content team may want to revisit the blog post themes. If conversion rates are up, PPC and SEO teams should evaluate what sources led to the increase in conversion, as well as what tactics regarding ad copy, keywords, or linkbuilding should carry forward into future strategy.
Most importantly, when it comes to creating regular reports – make sure you lay out all the data that is needed, using charts and tables that are easy to understand. And don’t forget to include plenty of comment to explain the data in ways that make it accessible and call out the wins, as well as challenges.