You want to be on Facebook – right?
Maybe you’re afraid your competitors are one step ahead of you? Let’s be honest – that’s a really annoying feeling – you want to be in front!
Most people get confused by the huge reach you can get on Facebook and how easily you can reach millions of people.
Is this post a sermon about how to avoid Facebook advertising?
Definitely not – on the contrary! This post is meant to help you better understand the factors at play when you start the Facebook engine. Also, it applies to all other social media platforms.
Let’s get to the point!
Facebook conversions
You can create a specific objective on Facebook called “Website conversions.”
During the setup, you decide when this goal should be “triggered” – you can do the same in Analytics if you run a webshop by setting a goal for people to visit your “thank you for your order” page.
The problem is that Facebook claims credit for all your sales, whether people click or not – and even for a “viewed” conversion.
If you run two campaigns – one with the goal “website conversions” and one with “clicks to website” – your click campaign will show fewer sales in Analytics than Facebook’s website conversion campaign. This is because Facebook uses a “first interaction” model – more on attribution models later.
Advanced setup
You can now specify your Facebook tracking code much more precisely than before.
This means you can set multiple rules for what counts as a conversion for you. Generally, my recommendation is to always set up your own tracking (via UTM parameters) so you know exactly what you spend on marketing and how much revenue you generate – in other words, how much you earn.
UTM tracking & Google Analytics
Goal setting
Just like you can see your Adwords campaigns in Google Analytics via google / cpc, you can also see your Facebook campaigns through simple tracking.
Example when data comes in
This is what it looks like when data comes in for Facebook / CPC
All other Facebook traffic is shares, likes, etc., not advertising.
How do you set it up?
Set up your tracking campaigns.
Start by opening the URL builder developed by Google.
Then configure it as shown in the example.
Copy the link into your ad.
When people click on your ad, you’ll be able to see in Google Analytics what it generates in sales, email sign-ups, and much more.
Easy, right?
The best part is it’s free, and you can use it for all external platforms like YouTube, Instagram, and partners.
let’s take a step up to understand what Facebook is really worth!
Now it gets a bit tricky.
Attribution models – advanced
Attribution models are rules that determine how much credit a given channel gets for a sale. For example, a “last click model” credits all value to the last click before the sale, whereas a “first click model” credits all value to the first interaction.
By studying these models, we can see which channels possibly influence sales – and I can reveal that it’s not always the last click.
By default, Google Analytics measures everything in a last-click model.
That means even if people click on multiple sources before placing an order, only the last click gets credit for the sale.
User flow model
Imagine how many times a user is exposed to your business across different media?
Normal attribution model according to Google Analytics
This means Google only credits the last source.
But is it really that simple? Most people visit many different media before buying something.
My thesis:
There are many different attribution models, and it’s hard to say which are most accurate for your business. What I challenge you to do is look at the different models and see if you find any data that doesn’t add up. Find out what your social media really gives you.
You get a lot of branding – or do you?
When you run Facebook ads, it’s often with branding in mind.
Companies that really understand Facebook marketing – and social media marketing overall – have realized these platforms are mainly for branding and less for direct sales.
Branding online can be measured – the same applies if you run offline marketing.
Google Trends (free)
When your brand starts getting searched for, you can see if there’s an increase in brand searches. You can also compare it with other brands like your competitors.
You can use Google Trends for this.
Below I compared a few grocery stores in Denmark.
Google Keyword Planner (free)
Another free tool is Google’s Keyword Planner.
Here you can easily see how many searches your brand gets monthly and how it has changed over time. For example, if you run TV campaigns for a while, you will definitely see a spike in brand searches – the same applies if a video goes viral for your brand on social media.
How to use Keyword Planner
As you can see, I put one of our own shops in the tool — Made4men.
This way, we can see how many searches the brand gets each month and how it develops.
Why is branding so important?
The great thing about a stronger brand is that it results in higher conversion rates since branded traffic often converts better than non-branded traffic. This means your Adwords and SEO campaigns will perform better if you build a strong brand. Can you see how this fits well with our theory about attribution models?
Wrong data?
Unfortunately, most people find their data completely wrong when running Facebook ads.
The problem is often they don’t analyze or segment their data.
Audience segmentation
I have written a lot about segmentation on Facebook and why it can be a problem.
Let’s refresh.
When you create your campaigns on Facebook, you can choose all the people on Facebook – that’s over a billion.
Is that relevant? No, of course not!
Audiences that work every time
Here are my “secret audiences” that work every time, no matter which company or industry I’ve worked with.
- Visitors to your website (traffic via custom pixels)
- Fans of your page – they already know you
- Email list (upload it to Facebook)
- Lookalike audiences (people similar to your other audiences)
- Competitors
A good piece of advice
The narrower your audiences, the higher the chance of conversion.
You can test broader audiences, but use a low budget and pay attention to whether you reach the right people.
Placements – you take the lead!
Just as segmenting audiences can be challenging, ad placements can also be tricky.
When splitting your ads, remember to choose one placement per ad set – where you target your audience.
The problem with having both Facebook and Instagram as placements for the same ad is that you let Facebook decide where your ads will show – this doesn’t give very precise data.
I divide my campaigns like this:
- Audience 1 | Facebook | Desktop | News feed
- Audience 2 | Instagram | Stories
- Audience 3 | Network | General
- Audience 4 | Instant articles
These are just some of the audiences you can create on Facebook and Instagram, but the important thing is to split them up. With a very clear segmentation, you can analyze your data and turn off placements that don’t perform.
There is often a big difference in which ad types work on Facebook versus Instagram.
Comments