Building useful reports

Paid Social

De-mystifying data visualisation

Ad people love to say the words “data visualisation.” It sounds very high tech, and very Big Data. But, let us blow your minds for a second… “Data visualisation” is just a picture that you instruct a robot (a piece of software) to draw. It’s usually pretty easy and the robot is usually pretty good at taking orders.

The robot we’ll be using is Google’s Data Studio. It’s free and easy to use. If you’re currently using a Google product like Analytics or Adwords (now called Google Ads), you can play along at home while we run through some helpful techniques. If you haven’t set up Data Studio yet, find out how to do that here.

Here are our most helpful tips we’ve found when visualising data for our clients.

Setting up a report

The first thing we do in any report is add a data source. You can do this in the right hand side of the report – just go with something simple for now, like Adwords. All kinds of data sources can be added, through direct integrations with Google, or third party apps. If you haven’t done so already, do some research to see how you can connect your CRM, EDM, or Facebook accounts to Data Studio.

It’s very useful to set up a date range selector at the beginning. This means you don’t have to adjust each graph’s date range individually and the report will be a lot easier to use.

 

Now, to the visual part of data visualisation. You can get Google Data Studio to draw all kinds of things for you: lines, bars, pies, donuts.

Your report is going to be most helpful when you are thoughtful about which visual format will tell you what you need to know to before you make your next business move. For example, for one of our clients, our goal is to constantly drive down cost per lead. A line graph makes it easy to understand change over time. Let’s look at how we’re doing:

 

We can see that this month has not been very consistent, and it’s hard to tell from this daily data if cost is going up or down. Adding a trend line will let us see if we’re heading in the right direction:

This is great news. The trend line is showing us that we’re achieving our goal of steadily driving down cost per lead.

Now, maybe you want to know how much money you spent on search compared to video. In this case, seeing your data over time might not be the most useful – it will just show you how much you spent each day, without giving you an indication of how much was spent on each channel, proportionally. For this measurement, a pie chart is most helpful:

It’s easy for anyone in the company to look at that and understand about a quarter of our budget goes to video campaigns.

These are just 2 examples to show you how which graph you choose can make your life easier or harder. If you want more information and examples, this is a great guide.

Using filters

Let’s keep going with the same example above. The natural next step from finding out that we’re spending most of our budget on search is to see if that extra spend is leading to extra conversions. This is where filters come in. Filters are super helpful when you’re running campaigns for different objectives. Setting up filters enables you to better understand your data and make more informed optimisation decisions.

The only thing we care about right now is “how many conversions”, so let’s set up a “scorecard” graph. This shows us one simple number:

This is the number of conversions across all campaigns. If you look in the right hand side menu, you’ll see the option to add a filter:

There are lots of options for creating filters and they’re mostly self-explanatory. For this example, I’m going to create a filter that includes any campaign, where the campaign type is video. Filters can be tricky when it comes to case-sensitivity, and they’ll be more difficult to use if you don’t have a consistent naming convention for your campaigns, so keep this in mind when you’re setting them up.

Now, once the filter is applied, we can see that the video campaigns led to exactly zero conversions.

This means the search campaigns have brought in 100% of conversions, so we’re definitely getting the most bang for our buck there.

Comparison

Usually, when you’re looking at data, it only means something when it’s compared to something else. You might want to compare this month against last month, or against this time last year, or NSW to Victoria, or Email to Search, etc.

Paying attention to small things like formatting and neatness will make it a lot easier for you to understand this information visually. Even something as simple as putting scorecards of different shades next to each other can help:

If there’s a metric you want to keep track of both month on month and year on year, a line graph with a comparison built in can help. In the below graph, you can see that CPCs have followed seasonal monthly fluctuations this year, but we’ve made significant improvements in cost this year compared to last.

Useful things that go together

You can also combine a number of metrics on the same chart. We find this helpful to gauge if things are going right at a glance. For example, you almost always want your leads to be going up while your cost per lead goes down. The below graph shows us quickly that something has gone wrong, and we need to address it.

Another example is how you almost always want your impressions and spend to move in tandem with one another. In the next graph you can see the lines are sticking fairly close together, which is a good thing. Note that we’ve set different axes for each measurement because we’re dealing with a totally different scale of numbers ($ and impressions). We’re reaching 20k people and spending only $30 on some days and if they were on the same axis, the red line would just be a flat line along the bottom. What we want to show is a relationship between any changes: are we getting more when we spend more? If not, why not?

Conclusion

We hope these tips have been helpful and shown you that data visualisation is just a way to make things easy to understand. It’s not hugely technical or difficult. The hard part comes when you have to synthesise all the pictures and numbers and make strategic decisions based on the patterns. Unfortunately, Google can’t do that for you yet (but they’re getting better every day)