- Your Content Should Sell
- Posts
- How to understand you ad data
How to understand you ad data
Establish your data review process like Goodo Studios
Matthew Gattozzi
May 23, 2024
Your Content Should Sell is a free a newsletter that gives you practicals about creating content for your brand. It’s a must-read among DTC marketers and business leaders.
Was this email forwarded to you? Subscribe here
You have finalized your ads and you drop them into Facebook Ads Manager ready to make you money.
This is just the beginning of the journey.
Because once those ads start to run, there are still more decisions to make.
We aren't going to talk about the technicalities of media buying today, but we will talk about using the data to make creative decisions.
You can have all of the opinions of what you think will work for an ad, but you don't decide.
The market decides.
They tell you if the ad is good or bad with their wallets.
Our goal is to make ads that make money, not make us feel good inside.
There will be a lot of times the ads that work aren't ones you originally thought would work. It wasn't only until the data showed these were winners.
Data is the most important part of our content creation process and creates a feedback loop to improve the ideation stage.
Setting up your report
When looking at Facebook ads, we set up a report in ads manager. Here is exactly the report.
Ad name
Ad delivery
Results
Cost per Result
Amount Spend
CPC
CPM
Hook Rate
Hold Rate
Average View Duration
Unique Outbound CTR
Adds to Cart
Website Purchase ROAS
What does the data report say?
Ad Name
I like to go by ad name so I can see the data from the ad level.
Ad delivery
Ad delivery is just letting you know if the ad is active or not.
Results
The number of purchases this ad brought.
Cost per Result - amount spend/results
This is to get an in-platform customer acquisition target.
Amount Spend
Amount spend is how many ad dollars is going to this ad. I usually sort the whole report by amount spend so that we can see the highest spending ads from top to bottom.
CPC (Cost per click) - amount spend/clicks
This metric is helpful to see if the ad stays efficient with more spend. When an ad starts to become “tired” you can see the CPCs jump up.
CPM (Cost per mille) - cost for 1000 impressions
Similar to the CPC, I use this metric as a health check to the account and ads overall.
Hook Rate - 3 second views/impressions
Of the people who viewed the ad, how many people watched this video for at least 3 seconds.
Hold Rate - ThruPlays / 3-second video plays
Of the people who watched the first three seconds, how many are watching till the end.
Average View Duration
The average time spent on watching the video.
Unique Outbound CTR
The percentage of impressions who are clicking on the link to go to the website. (Change type of CTR if you are also running META Shops)
Adds to Cart
The number of people are adding the product to cart.
Website Purchase ROAS
Seeing the return of ad spend - so if you put a dollar into Facebook, and get $2 back, then you have a 2 ROAS.
If you have put $1 in but get $0.50 cents back, you are at a 0.5 ROAS.
Now let’s talk about how to read this data and how it helps your creative decisions.
If you have found this helpful so far and aren’t subscribed to the newsletter, can you take less than 10 seconds to add your email?
This newsletter is free to join and free to leave.
Average benchmarks for these metrics
A good hook rate:
If you have 30%+ of impressions watching at least 3 seconds of your ad, it means the beginning made people stop scrolling and watch.
Something caught their attention that made them want to learn more.
This metric can help you understand your targeting towards customers based on what was said or shown in the first three seconds.
A good hold rate:
If you have a hold rate (ThruPlays / 3-second video plays) of 15%+, it means that the retention of the video is good.
It is a great sign when watch time is up, and people are engaged with your content.
This can help you see what is being shown and said in the video and could help you understand how to communicate with customers in the future.
A good click-through rate:
If you have a CTR of 1-3%, you are doing something that is getting people to say yes, I want to click and learn more.
This metric can help you understand if the messaging and visuals are compelling people to want to buy.
How to use these metrics to make creative decisions
Hook and Headline
The most important part of ads is the headline of a static or the hook of a video.
Who cares about second number 20 on a video if people never make it past 2 seconds?!
The hook of a video sets the expectation of the video.
The same is true for a headline of a static ad.
To start analyzing, I just look at the beginning.
Did we get people watching past three seconds?
If not, then I would focus there first in terms of making edits and creative decisions.
Since statics don’t have watch metrics, you can only base it off of CTR, so if the CTR is sub 1%, I would focus on changing up the headline.
If hook metrics look good, then let’s move to a little bit later in the video and statics.
Body of ad
For a video, if the hook is looking good, I then look at hold rate and average view duration.
I can start to identify when people are dropping off when viewing the ad.
If the average view duration is 6 seconds, that means most people are dropping off at that point in the video.
I can rewatch the video and start to ask questions.
Did I fulfill on my expectation I set in the hook? Is there something confusing or boring within second 3 to 6?
If the hold metrics aren’t very high, that means you grabbed the attention of the viewer but underdelivered. You want to be careful of over-optimizing for the hook that you don’t have people watch after the hook. This view data, can be super helpful to identify areas of improvement on the body of the ad.
The key is to watch the ad back.
This data by itself isn’t helpful if you aren’t actually watching back the ad for what is being shown and said in the ad.
Most people don’t do this - don’t be that person.
Conversion
Let’s say you have good hook and hold metrics, we look to the end.
Are we getting clicks and conversions?
The job of an ad is to create demand and direct that demand to the store.
I like to look at CTR, adds to cart, and ROAS together.
I will caveat that depending on your complexity of business, in-platform ROAS isn’t great, and you might have a different north star metric you use. You can switch out ROAS for something else, but you need a target.
You have to be able to assess, does this ad help you hit your target? Yes or no?
So if the ad is not getting clicks but had good view metrics, that might mean that the ad is entertaining but not selling. You need to sell, so an ad optimized for retention but not selling is not a good ad. If the CTR is low, there might be something in the body of the ad that is not compelling people to take action.
If you are getting a good CTR but not conversion, that might mean life after the ad, aka your website, is not a good experience. Sometimes the ad is great but the website is confusing or not optimized to sell. I won’t be talking too much about this on the newsletter, but the point is that you can have a non-converting ad because of the website, not the ad itself.
I like to look at adds to cart because it has a little signal for purchase intent. People putting items in carts means they are seriously considering price and the purchase.
At the end of the day, you need to rewatch the ad to see if the lack of conversion is because of the website or ad.
Most likely the ad is not doing enough to build demand, so opt to improve the ad.
Why can these metrics be lying to you
People like to flex metrics to make the ad seem good, but there is always more to the story than meets the eye.
A customer journey is not linear
What were the ad placements?
Audience Network placements can skew metrics. Were they watching because they wanted to or because they had to for the sake of playing a game?
So when you see really great metrics, dig a little deeper to get the full picture.
You could have a high hook rate but end up attracting the wrong cohort of customers.
You might have gained someone's attention but at what cost.
Did you say something so shocking to get people's attention that instead of attracting the 40 year old man that the brand is built for, you have 18 year old boys watching?
You could have high retention metrics but low conversion.
Maybe you focused too much on attention that you never sold the product.
Oh and to top it off, the ad is great, but because the website experience is not good, the ad doesn't convert.
Is that the ad's fault?
Instead of focusing on improving what happens after the ad, you blame the ad for not selling so you change a really good ad because the ROAS wasn't good.
Metrics can be a guide to help understand creative strategy and ad creation.
But because the customer journey is not linear, and every brand has unique rules to their account, you need to be careful of taking your metrics face value.
You need to dig in a bit more to make sure you don’t go the wrong direction.
We work with one brand that is very niche so we know the hook rates will be low, but the goal is very high retention on the hold rate. When we have low hook but high hold rate, we know that we got the right audience watching. The consideration window is a lot longer so we know the lag of impact is two weeks. We won’t turn ads off right away because they need time.
We have had other brands, mainly on Amazon so the Facebook ads metrics might look lower but Amazon is up 30%. Data can be complex and tricky.
Use data as a guide, but remember you always need to find what is true for your brand. Unfortunately I have to be a bit more broad on the newsletter to speak to many types of brands and businesses, so please take this as general guidance.
The scientific method
In grade school science, you learn about the scientific method.
Define a question
Gather information and resources (observe)
Form an explanatory hypothesis
Test the hypothesis by performing an experiment and collecting data in a reproducible manner
Analyze the data
Interpret the data and draw conclusions that serve as a starting point for a new hypothesis
Publish results
Retest (frequently done by other scientists)
The same is honestly true for advertising.
Claude Hopkins, the grandfather of advertising, wrote a book call Scientific Advertising.
So you get data back, but now it is time to go back to step 1.
This, my friends, is how you get your feedback loop complete and your content system in order.
We have come to the end of the 4 part series!
These newsletters took a ton of time to write but it was the most fun creating them.
We will get back to shorter newsletters next week.
Can you let me know what you want to learn next?
Until then, keep creating!
Matthew Gattozzi
Get a free assessment with Goodo Studios
My team and I create hassle-free creatives for consumer brands. If you are thinking about getting content, let’s set up a chat.
|
|
