I attended the Elisa-DBI Web Analytics summit late 2012 and was struck by a presentation by Pere Rovira on visualisation. While being aware that I need
to make my information pretty for the purposes of storytelling to stakeholders and management, I have rarely strayed beyond linecharts, barcharts, piecharts and the like. I love infographics and videographics, and think it’s a great way to present information, but I’ve always done the analysis using spreadsheets
(or similar), then presented the analysis as prettily or powerfully as possible using visuals. This is changing. With Google Analytics you can now create powerful visualisations to aid your analysis, not just to get your HiPPO on board! This post outlines a simple example to get you started on the road to visualisation glory using Google Analytics motionChart.
How are my campaigns going?
This use case lets you see at a glance how Campaigns perform over time in terms of volume (visits) and a success metric. I’m using goal conversion rate because of the nature of the website I’m using as an example. You could easily use
eCommerce conversion rate, or if your website isn’t commercial then page views or time on site; whatever goals you have set.
In Google Analytics, go to Traffic Sources>Sources>Campaigns and select the previous two weeks and a WOW view. If you have some particular campaigns you wish to look at, filter for them at this stage, then click on the icon containing 3 circles highlighted below.
By default your visits will be on the x axis along the bottom, so you just need to configure your y axis to your metric of choice, in my case goal conversion rate.
Without using any of the other features in motionChart, you already have a world of insights in front of you that it might take you a few hours to decipher from rows of data in spreadsheets, that’s if you spot them at all!
From the chart above I can see that I have a clear route that I need to move my little dots. First I need to optimise under-performing campaigns (drive them up vertically) then I need to push volume for the highest performing campaigns. I have a clear outlier here above in red. Highest volume of the bunch, but with the lowest goal conversion, ie: I’m wasting money. That needs immediate attention. Instant actionable insight!
What else does this visualisation tool do?
A lot! Here’s a list of some more things you should definitely try now if you’ve followed the steps above:
1: Press the play button to see the WOW changes for each of the campaigns in your report, animated.
2: Click on a bubble to attach the campaign name as a label, to highlight areas of specific interest. This is really useful if you see something of interest and want to look at it closely, or compare a couple of campaigns.
3: Assign a metric which changes the size of the bubbles. For me, this can get a bit ‘tmi’, but interesting to weasel out correlations.
4: Assign a metric which changes the colour of the bubbles. Again, for me this is getting a little too complicated to be a useful view.
5: Try segmenting by other values, like keyword medium and source, exactly the same principles apply.
I’ve caught the visualisation bug now and have been trying out Gephi and some other tools as well as motionChart. They’re cool!… but most of them are still more like university projects than business tools, built by devs for devs. It would be great to hear about any of your experiences of visualising data for analysis, particularly where you’ve had great success!