We are moving our dashboards up a notch! In the process of creating a customized dashboard for reporting mobile traffic, we had some interesting discussions about which metrics to include and how this dashboard should be different from the existing site-level dashboard? Through this post I am presenting my personal set of metrics that
was proposed to the team:
1) Mobile Sessions: This is the most rudimentary of all KPIs that no dashboard can do without. This metric is great to understand some key attributes of your visitors by using different dimensions. Which devices are your visitors using? What percentage of these devices are tablets? Which brand of devices do they prefer? Which service providers do they use? My personal favorite though is to apply the ‘Mobile and Tablet Traffic’ segment to the location reports (especially cities) to understand how mobile traffic from different cities is similar or different. I also take a quick look to see the sessions report broken down by the mobile operating systems and browser (set as the secondary dimension).
2) Mobile Session as a Share of Total Sessions: One of the metrics that I always try to keep track of is the relative share of mobile and tablet traffic when compared to the total traffic that arrives at the site. Of course, we are expecting this share to creep-up gradually, however trending this data point over the past year tells me if there have been any unexpected spikes or falls. I typically do this by charting the weekly share of mobile sessions till date, starting with the comparable week of the previous year.
3) Bounce Rate: The next metric on my list has to be bounce rate. This is to do a quick health check about how effective (or ineffective) the site is in retaining customers. Obviously, we will go beyond the topline number in this case. We will look at how the bounce rate varies for different traffic segments, like we did in the case of Mobile Sessions. We will also look at how the bounce rates vary across different landing pages.
4) Share of New Sessions: I picked this metrics not just because it is important for us to know what Share of Sessions is from those who have never visited the site before. True, the accuracy of this number is debatable in scenarios where we are unable to track a visitor across multiple devices. However, my intention of using this metric is to observe the trends and understand if the ratio of sessions from new users to repeat users is changing when compared to earlier time periods.
5) Campaign CPC: While I am interested to review all metrics that measure campaign performance, the CPC metric is of particular interest to understand the cost incurred for the first step of acquisition. What I am interested to know is how the CPC for mobile traffic varies when compared to desktop traffic at the channel, engine, campaign, ad group, and keyword levels. This metric is important to arrive at the specific optimization tactics that are to be applied to the mobile campaigns.
6) Share of Campaign Traffic: I have included this metric as I feel it is important to understand what percentage of mobile traffic arriving at the site is due to marketing programs. This is important due to the differences between how users use search, and can be targeted based on location, when browsing from a mobile device.
7) Pages per Session: A metric to understand how deep visitors go during a session. This metric will also be used to understand how usage varies for different segments for e.g., campaign traffic vs. non-campaign traffic, by device type, by screen resolution, etc.
8) Average Session Duration: To be used together with Pages per Session to understand engagement of mobile traffic as a whole and also the various sub-segments within mobile traffic described earlier.
9) Conversions: We cannot leave out the conversion metrics, can we? But what if the objective of acquiring mobile traffic is to move the visitors through the conversion funnel and not necessarily to complete a transaction? Even in the second case we have to clearly define what outcomes we are expecting to generate from the mobile traffic. In this case we can define engagement goals for the mobile traffic. Irrespective of whether the definition of a conversion is the same as the desktop traffic or different, the absolute number of conversions is important to understand how a site is performing vis-à-vis mobile traffic.
10) Conversion Rate: To extend the logic for tracking conversions further, we should also track the conversion rate for the mobile traffic. This will help us compare and contrast the various mobile sub-segments. We can also study how the conversion rate varies for comparable segments (same search engine, paid keyword, etc.) between the desktop traffic and mobile traffic.
So, that is my list of top 10 metrics that I proposed for adding to the new mobile dashboard. What do you think? Are these metrics sufficient, or did I miss anything important? What will you change in the list above and why? I’d love to hear and learn from your wisdom and experience, so why not leave a comment below and let me know your thoughts. Thanks!