Debunking Myths About Facebook Ads

On this occasion I would like to talk about some common myths that exist in the industry when running ad campaigns on Facebook and Instagram. I hope you find it useful.

While it is true that Facebook prefers clean ads with images with almost no text, it is also true that since mid-2016 any ad that had more than 20% of text within the image would certainly be rejected. However, this is not the scenario that occurs today.

This rule today exists a more flexible way. The ad is given a rating based on the amount of text in the image: the category can be "OK", "low", "medium", and "high". If the image has a lot of text, Facebook will send an alert warning that this may affect the reach of the ads.

AdEspresso ran a test where they ran a campaign with an ad that had a lot of text. As you can see from the results, the performance was not bad.

However, it is important to know that although an ad with a lot of text can be shown to users, in times when there is a large influx of advertisers ready to advertise, (such as Cyber Monday, Christmas, Mother's Day and other important campaigns) the Lower quality ads will be affected, compared to those that do meet Facebook's standards.

While it is true that Instagram attracts a high percentage of young users, it is important to keep up with the growth of the platform to have a more objective view. According The Verge Instagram has reached 1 billion users and according to the latest information provided by the platform, there are at least 800 million monthly active users and 500 million who use the application daily. This shows that such a number of users cannot be only Millennials, but that the age distribution is much broader than we think. If we compare these figures with the 330 million active users that Twitter has, or the 200 million of Pinterest (which is the same number that Snapchat has), we can understand that the 800 million users on Instagram is a much larger and richer audience. .

To this we can add that Instagram Stories has 300 million active users daily, which makes it even more attractive to attract all kinds of audiences.

CPM is the cost per 1,000 ad impressions. It would make sense (keeping all variables stable) that having a low CPM is better, as long as it generates more ad distribution at a low price. The problem is that it is very rare for all variables to remain stable in Facebook ads.

Every time an ad is run, an auction enters and the price paid depends on how many advertisers are bidding to reach the same audience and how much they are willing to pay.

It's a common myth to think that lower CPM is better, but let's look at what low CPM really means.

When we ask Facebook under CPM we would be saying "Give me the cheapest possible views" which Facebook interprets as: "Help me find the cheapest views by showing the ad to a junk audience, which no one else wants to reach"

That audience is cheap for a reason: Nobody bids to reach them because they never take any action or click, or fill out form information, let alone make a purchase on a site.

When you optimize a traffic campaign to achieve impressions, this is the audience you are going to reach. When you optimize for video views it is even worse, since your ad will reach people who have not even turned off Auto-Play on videos, (which most heavy-users do).

So usually when a client expects to hear that the CPM is very high I usually answer that it is very low.

This is how we can see it in this example carried out in the United States this year: Campaigns optimized for link clicks were analyzed, which had a CPM below 8 USD, compared to the CPM of 14 USD in a campaign optimized for conversions. It is obvious that users who are more likely to buy are more expensive than those who only click on the ad.

A quality audience costs more, because it is in high demand and this happens because this audience clicks more often on ads to buy things.

In Google Ads, the price you pay depends on the bid, how much competition exists and the Quality Level. This is why it is commonly assumed that Facebook works the same way with the Relevance Score, replacing the Google Ads Quality Score.

This is a mistaken belief, Facebook itself explains it in the help section for businesses: if your ad has a low relevance score but generates results there is no important reason to change anything. Perhaps improving the creative piece and segmentation can generate lower costs but we do not recommend making changes just to raise the relevance score, if we recommend making changes to generate more results and at a lower cost.

With the exception of businesses such as restaurants, where no one is going to order a pizza at six in the morning, it is not recommended to select specific times for our ads to display. The Facebook algorithm knows that most of the products and services that are advertised do not depend on an hour to be consumed. For this reason, the budget "meaningless" is not spent throughout the day, on the contrary, Facebook shows the ads during the hours when the audience is connected

When you log in to the ad manager, in the account Overview section you will find the breakdown of how the hourly budget was spent.

It is unnecessary to stop the delivery of ads at night because Facebook already takes care of that, it can also be a mistake to think that a key time to show ads is Monday through Friday from 9 a.m. to 6 p.m. During this time, users are busy with their daily chores and perhaps they are more open to browsing social networks and websites on weekends.

It is also important to let the ads run throughout the day, seven days a week, so that the Facebook algorithm can optimize delivery at the most convenient times.

Sometimes we may think that through our analysis of all ad sets, we can give a higher bid to the best performers and a lower bid to the lower performers. That is, we think we can make better decisions than the Facebook algorithm.

Unfortunately if a click costs $1 and we lower it to $0.50 we cannot expect that there will be no consequences. We will probably achieve some isolated results, but the overall volume of results will most likely drop considerably. I recommend this reading. Facebook Ads Bidding Experiment: Automatic or Manual? The Surprising Winner  in which an experiment is developed in detail to find out what really works better, whether automatic or manual bidding.

Taking this campaign as an example, logic would tell us that we should have turned this campaign off a long time ago.

However, even though this campaign had a very high frequency, it also generated an income of 165K with an investment of 14K.

It is important that we keep an open perspective, since in this case income is more important than frequency. The secret of success is not segmenting the campaign to the same users for too many days. Facebook has daily frequency cap tools. For example:

- In the news section a person will not see an ad on the same page more than once every six hours.

- In the same way, a user will not see an ad from the same advertiser more than once every six hours in the Instagram feed.

- If the location is Instagram Stories, a person will not see an ad from the same advertiser more than once every 12 hours.

A good option for retargeting campaigns is to leave them configured so that users do not see them for more than seven days. Although the frequency caps will allow users to see the ads several times but spaced apart, they will see them for a limited period of time.

Are there any other myths that have been demolished? I would like to read your opinions and experiences.