May 16, 2017

6 ways to dramatically improve your Facebook targeting using Google Analytics

By Supermetrics

FacebookGoogle AnalyticsPerformance Marketing Analytics

[ Updated May 16, 2017 ]

10-MINUTE READ · By Jon Quinton on May 16 2017.

Facebook offers an intimidating wealth of targeting options, putting new advertisers on a steep learning curve before they figure out what works for them. What if you could start that learning process with a head start, leading you to a profitable campaign more quickly?

If you know where to look Google Analytics offers new advertisers much needed context, and direct pointers for improvements to targeting on Facebook. This post is geared at showing you how to get that information.

For the purposes of this post I’m using Google’s demo account which you can find here. For those wanting to dig around inside a well populated GA account it’s a great resource, so I’d highly recommend getting access.

That being said, if you’re completely new to advertising on Facebook I’d highly recommend checking out people like Jon Loomer, listening to some of the Facebook related Social Media Marketing Podcast episodes, or reading through Shopify’s excellent guide for a basic introduction.

So here we go – some commonly asked questions, and how you can answer them using Google Analytics!

Which demographics should I hone in on?

Setting your paid campaigns to target all people of all ages can lead to unnecessary waste in spend, which with a bit of pre-existing knowledge could easily be avoided from the get go.

I’ve seen dramatic increases in performance by narrowing demographic targeting before, and thankfully Google Analytics makes it super easy to get a head start.

To get these insights you’ll need to make sure you have demographic reporting enabled within GA. That’s a simple matter of hitting ‘enable’, but make sure you have your privacy policy updated to reflect your use of this data.

Once you’ve collected some data, navigate to ‘Audience’ and then ‘Demographics’. Below you can see a report with the ‘Converted Users’ segment applied:

Demographics on Facebook Ads

Let’s say for example you’re running a campaign in Facebook that’s optimized for conversions, understanding which demographics are likely to convert is very powerful knowledge. Using the data above to fine tune your demographic targeting will mean a more relevant audience, and more chance of picking up people most likely to convert.

Below you can see how the data from GA’s demographic report can be easily reflected in your Facebook targeting:

Google Analytics demograpgics

What locations & languages should I include?

Far too often people only focus on the markets and languages that seem most obvious, or they take a scatter gun approach when the time comes to expand campaigns into new markets.

Why not find suitable candidates for international expansion before you start? Within Google Analytics this is really easy to do. Click ‘Audience’, followed by ‘Geo’, followed by ‘Location’ to get the following view:

Google Analytics locations and languages

Here we can see that in terms of traffic, the top three countries are as follows:

  1. USA
  2. India
  3. United Kingdom

However, sorting by transaction revenue paints a very different picture:

Google Analytics locations and languages sorted by revenue

When sorted by transaction revenue the top three countries are:

  1. United States
  2. Canada
  3. Mexico

The difference could of course be down to issues with international shipping or onsite translations – all factors that should be considered prior to running paid campaigns. For a bit more on this side of things, take a look through Dan Barker’s post on Smart Insights.

Checking the same report for ‘Languages’ rather than ‘Locations, you can see a dramatic fall off in languages beyond English. It goes without saying, but before you start targeting new locations and languages, make sure the site caters for them and that you stand a chance of winning.

Should I optimize for conversions on mobile?

Of course we should target mobile users! Well, have you checked? It sounds a bit mad in 2017, but quite a few sites still struggle to convert on mobile even if mobile is great for traffic.

Before throwing all your advertising dollars at mobile users, it’s worth double checking that the site converts well for that segment. Again, in Google Analytics this is really easy to see.

Click ‘Audience’, then ‘Mobile’, and then ‘Overview’ to get a sense of how mobile and desktop stack up against each other:

Google Analytics device category

As you can see above, mobile performs drastically worse than desktop when it comes to revenue.

With 90% of Facebook’s daily use being on mobile you definitely don’t want to ignore it, but in this case you may want to consider optimizing for traffic on mobile, and conversions on desktop (Facebook will allow you to remarket to a cross-device audience).

In the bigger picture you’ll probably want to look at ways of optimizing your website for higher conversions on mobile though.

How long should I wait to see conversions?

If you’ve ever set up a Facebook conversions campaign you’ll have noticed the option to optimize for conversions within the following windows:

  • 1 day click
  • 7 day click
  • 1 day click or view
  • 7 day clicks or 1 day view

The conversion window that you select above will determine the data Facebook uses to optimize your campaigns towards achieving conversions, so you’ll want to make sure you select the most appropriate option for your business.

To get a sense of how quickly your customers convert, navigate in Google Analytics to ‘Conversions’, then ‘Multi-Channel Funnels’, and then ‘Time Lag’:

Google Analytics conversion time lag

As you can see, 60% of revenue for this website is occurring on the same day as the first interaction. Therefore, optimising your conversion campaign on Facebook for a 1-day click is perfectly OK.

Which blog posts should I promote?

One of the bigger traps people fall into is either paying to promote every post they produce, or holding back until a new batch of content is created when a historic set of content already exists.

Paid campaigns on Facebook don’t work like magic, and if a blog post has tanked organically it’s highly unlikely to work with paid spend behind it. Given you’re considering precious budget to promote your content, it always pays to be selective.

My first step is to look at what content has driven the highest quality traffic via Facebook, either to get a sense of what new content should be produced, or to find some quick wins for campaigns whilst new content is being created.

Go to ‘Behavior’, then ‘Site Content’, then ‘Landing Pages’, and then apply ‘Source’ as a second dimension. The report you get should look something like the following:

Google Analytics landing pages

Note that the traffic from Facebook in the demo account was terrible, so in this example I’m using the report from my own site

Given that in the majority of cases when promoting blog posts, I’m looking for engagement and traffic rather than conversions, what I typically look for are articles that have resulted in a higher time on page.

What’s a sensible CPA to aim for?

This is one of the harder questions to get a definitive answer on via Google Analytics, as your Facebook campaigns will almost certainly behave differently to other previously explored paid channels.

As a result, this is really about getting a benchmark to compare performance against, and one that would ideally be reviewed regularly until you build a picture of what ‘good’ should look like for your business.

To get an estimate, I would suggest looking at non-brand paid search and display, calculating the CPA for each channel, and setting your initial (to be reviewed!) target somewhere in the middle of those two.

If you’re onboarding a new client then ask for details of spend on each channel to help with some context, but to get a view on paid search and assuming you have Adwords and GA linked, click ‘Acquisition’, then ‘Adwords’, then ‘Accounts’:

Cost per acquisition Google Analytics

If you’re working with a business that has some historic data on Facebook performance then that’s clearly the ideal place to start, but for those that don’t then reviewing other channels and general conversion data will at the very least give you a sensible starting point.

The list of insights available from Google Analytics goes on and on, but hopefully this is a useful start for new advertisers.

Good luck, and happy campaigning!

About Jon Quinton

Jon Quinton runs Overdrive Digital, a consulting business specializing in paid social, paid search and SEO. Working with clients in both B2C and B2B, Jon enjoys the challenge of optimizing campaigns for greater conversions, and always focusing on the fastest way to make an impact on the business. If you need a hand with your digital marketing, get in touch for a chat!

Turn your marketing data into opportunity

We streamline your marketing data so you can focus on the insights.

Book Demo