Facebook Mini-Guide: 7 Tips They Don’t Want You to Know
We don’t claim to be Facebook marketing experts, but over the years, we have learned more than a few lessons. I want to straight up just give you all a list of 7 rules of thumb that we apply to our marketing. Have any to add? Let us know, and we’ll be sure to give you credit. We’re looking to learn too!
One of the main reasons that we have not advertised as much as I would have liked on Facebook is because, frankly, I am not a fan of their service on the ads side. I think Facebook is a useful tool, but they really only care about the users and the really big spenders. They don’t want to deal with medium and small sized operations and, hence, have the most horrific “customer service” in the industry.
Their entire algorithm on the ads side is designed to (in their eyes) generate the most ad revenue for them. As such, the resulting system does not behave like you would expect it to under different circumstances. Facebook operates their system behind an iron curtain and does not disclose much of how their ads system operates.
Therefore, we have been left to fend for ourselves and figure out the nuances of the beast. We, along with the collective help of many marketers, have gotten (we think) a great idea of the rules of the game. Without further ado, here are our…
Our Mini-Guide: FB Ads Marketing Rules of Thumb
1. Upload 1 ad per campaign.
There is no reason not to upload just 1 ad per campaign. It’s cleaner for the reports generated, making it easier for you to work with the data. When you put multiple ads in a campaign, Facebook quickly decides which ad to give traffic to and which ads to halt traffic by which ad is making them (not you) the most money. By uploading only 1, you avoid this.
2. Make multiple duplicates of the same ad.
We always make at least 3 copies of each ad we upload because they almost never perform the same even if everything about them are identical. You should only do this for broader targeting since you’ll reach banner blindness too quickly with narrow targeting. Once you get enough data, you can pause the poor performing ones and only keep the high performance ads.
3. Facebook checks CTR at around 2,000 impressions.
This is based on our own experience. Facebook will determine, at around 2,000 impressions, whether it likes your ad enough or not. If the CTR is considered good by their algorithm, then they will continue to send you traffic, or else it will discriminate against the ad and not give it the time of day.
4. Bid high to start, then lower bids to replicate a frequency cap.
Of course, Facebook doesn’t have one of the basic functions of a self-serve platform: frequency capping, which means you can’t limit the number of times someone sees your ad, causing banner blindness. How you combat this is bid relatively high to start, and once you have ads with decent CTR’s over at least 20 clicks, start lowering the bid over the next few days.
As you lower the bid, the frequency the ads are shown will lower, as demonstrated by your click volume lowering. At that point, you should have higher CTR’s and ads that last longer.
5. Never actually pause a Facebook campaign.
We have experience many instances where we paused a campaign, and when we started rerunning it, the campaign performed completely differently (or received little traffic due to lower CTR’s). My theory is that Facebook places your campaign given the current landscape that it thinks it’s the most optimal for them.
When you pause and rerun the campaign, Facebook re-calibrates your placement based on how your campaign is running at that time. Therefore, when a campaign does well, and we need to pause, we simply set the budget to $1 instead of pausing.
6. Don’t trust Facebook reported clicks.
Facebook click discrepancy is larger than any other traffic source we’ve used. Base your CTR, CPC, and CVR calculations on clicks you track on your server using your own tracking software. At least rely on a trusted 3rd party to track. We’ve had instances where almost half of the clicks of a campaign were unable to be verified. On the same lines, count multiple clicks from the same IP as 1 click, unless it’s mobile traffic.
7. Edit images that other marketers are also using.
If you decide to use an image for an ad that other marketers are already using on Facebook, be sure to edit it just slightly, such as changing 1 pixel. Facebook seems to retroactively disapprove ads they don’t like. When they do, they likely purge all the identical images on their system. Editing your images slightly will likely avoid this.