CASE STUDY: How To Improve Your Facebook Ads Using Google Analytics Data

Anyone who has run an online store knows that sending traffic to your site is one thing, but getting people to buy is another thing entirely! So much time and effort goes into sending the right people who are ready to buy so that our time and money isn’t wasted. In today’s post, I’m going to run through a new way to look at your existing data to give you some new insights so when you serve your ads, it’s more likely to be the right people.

Now, if you’d explored Facebook ads before, you might know that Facebook can actually automate finding the right audiences for you a little bit. When you select the conversions ad objective: Facebook learns from the people who buy from you, and, using it’s plethora of data and AI tech, serves your ads to more people like that.

However, in order to use the Conversions objective in Facebook ads to help target those who convert, there is a minimum of 50 conversions in 7 days in order to best use this — which means it won’t work for every business. So if you sell anything under 50 products in a week on your online store, Facebook will not be able to optimize your ads based on people who buy from you.

For stores with less than 50 sales per week, we need to do it the old-school way, deep diving into data. For this, we can use Google Analytics.

What can Google Analytics tell me?

So, if we’ve got a smaller online store, we can’t tell Facebook to automatically target people who are likely to buy from us, but we can learn a surprising amount about the people who are buying from us, and the contexts in which people buy from us, using Google Analytics.

So, how do we identify the people who have bought from us? We look at the traffic on our “conversion completion” page: for many of us, that will be a ‘thank you for purchasing’ page. It’s the page that people can only get to on your site if they have successfully purchased one of your products, or signed up to your service.

Now, the caveat I want to start off with is that to draw conclusions from our data, it helps to have a lot of data. If you’re dealing with single digits, there’s nothing to say that individual person pressed ‘buy’ by accident when their dog demanded their attention or, on the flipside, their kids distracted them mid-purchase making it look like your sales page isn’t working. Having a larger dataset means you’ve got clearer patterns to work with.

I’d suggest waiting til you’ve got at least 100 sales, and even then, draw your conclusions with a grain of salt, because you might be completely different patterns when you have 1000 sales. But, we all need to start somewhere, so use what you have and use that information to begin testing and learning from your data.

CASE STUDY: My Online Store

Did you know I’ve opened up an online store? I sell gorgeous social media templates and tools for businesses, freelancers and teams. It’s been a little fun little side-project, and, like all good side projects, I’ve been starting small, testing, and expanding as I go. So, I’m going to use my online store as my case study for how you can use Google Analytics to learn more about your audience, and better target your Facebook ads.

So: if I’m going to run Facebook ads for my store, I don’t just want to know who clicks on my ads, but I want to know who buys. And to do this, I need to really learn about what’s going on behind the scenes. And since I have less than 50 sales per week, I need to climb under the engine and look at it through Google Analytics.

Finding our audience in Google Analytics

For this example, I’m looking at just the people who have 100% completed a purchase. To find the people who have completed a purchase, I’m looking at my Purchase Confirmation page traffic. So, let’s navigate to this page.

Head to your Google Analytics, and, from the Behaviour menu, head to the Site Content toggle, then All Pages.

I like to toggle to the Page Title view, so I can just type in my Page title, but you can search by Page URL, too.

I search for my ‘Purchase Confirmation’ page….

And, voila, my data just for my Purchase Confirmation Page appears excluding all other page data:

Now, in the top right-hand corner, we need to choose a date range to study. It might be one week, one month, or all time. Ensure it’s a time frame where you can get the largest amount of data you can, especially if you’re working with only a few sales a week.

The next step, as I’ll outline in my case study below, is to add a Secondary Dimension to this data to find out more about who is purchasing.

Filter By Country

There are a couple of ways you can filter by Country. You can use the Secondary Dimension filter to break your data into the countries your traffic comes from, or you can segment your data to only look at your data through the lens of a specific country.

Let’s start off with Secondary Dimensions.

Once you’ve navigated to the page you’d like to deep-dive into, use the Secondary Dimension feature to break down your data.

Simply click on the Secondary Dimension filter, and type in ‘Country’, and apply the filter.

From there, you can see which countries your sales are from, broken down.

Some of you might be wondering ‘Is this all real traffic’? Not always. You might find you get some ‘spam’ traffic from countries which are unexpected. For example, if you are a local business in Australia and you sell handmade crafts locally, but you might wonder why you get some traffic from another part of the world. It’s possible this is ‘spam’ traffic that has just landed on your Purchase Confirmation page, rather than actual purchasers. One tip to discern if it’s real traffic is to look at the average time on page — if it’s especially low (like one second, rather than one minute), you can assume it’s not a real human.

The other approach to sorting your country data will also eliminate this traffic if you only focus on a specific country. You can add a Segment to filter your data, to only show traffic from a certain country, and exclude potential spam audiences. If you only sell to a specific country, this might be a useful approach.

To do this:

Scroll to the top of your page, and add a Segment.

It’s this one right here (that currently says ‘All Users’):

Apply a Country segment, and all your data will appear just from traffic from that country once you hit ‘apply’.

Ta daa! Now, you’ve just looking at traffic from that country only.

Now, remember that this Segment will stay active as you travel around Google Analytics, so switch it off once you’re done with it!

How is this useful for Facebook Advertising?

Instead of targeting my ads globally, I can target my ads just to the countries interested in buying from me.

Or, I might have assumed all my sales were from only one country or region. However, with the beauty of the internet, I also have an international audience. From this, I can get an idea of what markets I might have overlooked, or could consider expanding into (like Finland. Hei to my Finnish customers!).

Desktop vs mobile?

So, let’s find out which devices people are actually buying from the site on — is it deskop computers or smartphones?

Head to your good old Secondary Dimension and search “Device Category”.

From here, I can see that in my case, 90% of sales are on Desktop:

What does that mean for my Facebook ads?

We can target our Facebook ads based on Device. While most users are on Facebook on mobile, but a lot of sales actually happen on Desktop Computers (although, this is rapidly changing). You might be directing cheaper traffic to your site by using mobile ads, but you could be wasting your money if people purchase only on desktop.

Instead of spending money on ads on mobile devices because they are cheaper, I could be spending my budget on desktop placements only because they result in more sales.

Obviously, it’s not to say mobile ads aren’t useful for awareness, but it’s about knowing what tool is right for each job, and this job is sales.

Apple or Android?

If you do have a lot of mobile users purchasing, you can also see whether Apple users or Android users purchase more. There is an older study from 2014 that claims that Apple users are likely to purchase more (which could be due websites being presented better on iPhones, or those who can afford iPhones may have a higher disposable income) — so it’s worth checking out your data to see if this is the case.

Again, this is a category we can target in Facebook Ad, so it’s useful to see if you can find out if this applies to your audiences. If you type in “Device Branding” into your Secondary Dimension, you can see which device they were on.

And while 90% of my sales were from Desktop, the sales that were from mobile devices, were all, indeed, on Apple devices:

What does this mean for my Facebook ads?

If your data shows a high number of purchasers on mobile and on a specific device, focus your budget toward them.

Obviously, in my case this data is single digits, so I’m not going to pivot my business based on this.

Time Of Day

For Lifetime ads on Facebook, you can set your ads to a schedule. Different groups are likely to buy different products at different times of day. If you are selling B2B, you might find your sales are during business hours. If it’s consumer-facing, it might be during primetime in the evening. You can conserve your budget by just serving ads during the times you know your customers are buying.

Let’s look at time of day in Google Analytics by searching “Hour” in our Secondary Dimension, and see if there are any patterns. Using classic 24-hour time, Google Analytics looks at this based on the timezone of your Google Analytics account.

In this case, you might want to Segment by Country so you’re looking at only one timezone at a time and have a little play with a timezone calculator if that country is a different timezone to yours.

Just looking at my Australian stats, while there isn’t a lot of a data there, I can see that the majority of purchases are between 11am — 1pm, so just before, or during, lunch.

What does this mean for my Facebook ads?

Instead of serving my ads 24 hours a day, why not just serve them during the hours I know people are more likely to buy? Again, with limited data, this is a useful indication of something I cantest in my business.

Day of the week

Is your product something people buy during the week, or on the weekend? Are there patterns to the day of the week people are likely to buy?

In Secondary Dimension, if you search “Day Of The Week Name” you can see the day of the week your conversions have happened.

When I look at my data, for Australian customers, I can see that the vast majority of sales happen on a Tuesday and a Wednesday. I can see that no one wants to think about work on the weekend, or on a Monday!

What does this mean for my Facebook ads?

Again, the Lifetime Ads allow you to serve your ads on a specific day of the week. If I know my sales are mostly on weekdays, limit my ads to those days.

City or State

If you focus your products in a specific country, you might find that looking at a City or State level will give you new insights into which audiences to focus on. By searching the Secondary Dimension filter for “City” or “Region” you can isolate these attributes.

I can see that I’ve got some latte-loving fans from Melbourne as my primary purchasers for my Australian audience. Shout out to Melbs!

What does this mean for my ads?

Instead of targeting the whole country, I might just want to target my ads to the cities where I know I am getting the highest conversions.

Age And Gender

Did you know Google can also show your data based on age and gender? If you’re logged into a Google Service like Gmail or YouTube, Google knows your demographic data. Google will do it’s best to share demographic data, like age and gender.

Tip: I’ve looked at these stats a few times of the past few months and this is the first time it’s had enough data to give me any insights, so don’t be discouraged if you don’t yet have data on demographics — it might just be a numbers game!

Back to our BFF, the Secondary Dimension filter, type in “Age” or “Gender” and see what audiences you’ve got.

How does this help me with my Facebook ads?

Demographic targeting is one of the staples of Facebook ads! If you know your primary purchasers’ age and gender, you don’t need to waste your budget on audiences not interested in your product! I can see I’ve got a key audience in females ages 25–44.

This will also help with your ad creative: if you know your key audience is a specific type of person, perhaps use images of them in your ad, or create copy that speaks just to them.

Conclusion

By looking at my Google Analytics data, I’ve learnt a lot which I can use for targeting my Facebook Ads.

I have new insights into:

• Targeting my audience: I know what countries and cities my purchasers are more likely to be from, so I can target my ads to them.
• Creative and content: If I also get age and gender data, I can rework my ad copy to appeal to that audience.
• Placements and devices: if I know people buying from me are more likely to be on desktop or mobile, that helps me define not only the device, but the placements, I can use.
• Spend Type: if you want to focus on a specific day, or time of day, you may want to use Lifetime Spend in order to capitalize on the peak times your audiences are buying.

Now, I’ve only got a small sample size because I’m just starting out. But there’s no reason I can’t do this exercise again next month, or in six months, as I gain more sales and more data so I’ve got a better picture of what’s happening.

Getting across Google Analytics is a handy little trick to help level up your Facebook ads, so well worth exploring this tool to help give you some insights into who your audiences are.

Awesomeness In Your Inbox

Sign up to Rachel’s newsletter, The Beaney Beat to receive first-in deals on the latest courses, access to exclusive content and the run-down on the latest that’s happening in the world of digital and social media.