If you’ve previously run Facebook ads, or have watched some of the content to learn to do so, you may know about the 20% rule. If you don’t, here is what it is; many people recommend to bump your budgets by 20% a day in order to scale your ads without ruining or reseting the optimization.
This isn’t broscience as there’s a “last significant edit” column in the ads manager and any bump in budget greater than 50% triggers the last significant edit and resets the optimization. This is even more trouble-some for CBOs which you often really want to scale as they have several ad-sets and audiences that are ready for larger budgets after you’ve proven your original thesis.
Alex from GetNotissed has worked a simple work around for CBO scaling which seems to work in most cases and I’ll explain that in a bit. The 20-50% budget raise without triggering reset is a guideline given directly by Facebook. However, the fact that we associated a time-window with it was how we perceived that guideline. In other words, you can do multiple 20% raises each second to reach your desired budget in a minute instead of doing 20% raise/day.
So you can go from spending $100 per day per CBO to $500 per day per CBO, without triggering a reset, in 1 minute instead of 9 days as long as you do multiple edits of 20% each. Be sure not to directly raise your budget from $100 to $500 which obviously will trigger the reset.
In my personal test, the theory worked great but I’d still not advise making extreme budget raises using the 20% per second rule.
I like to scale my Facebook ad campaigns with manual bids. One of the the tough decisions is to identify what is the optimal bid with regards to the best combination of number of sales and profit per sale. In short, getting the best return on ad spend (ROAS).
One manual way that I’ve used in the past is to start my ads by placing a bid that would result in 1.00 ROAS. For example, if the cost of goods sold is $10 and I’m selling the goods for $30, I’d start by placing a manual bid of $20. What this translates into is that I’m willing to break-even to initiate the learning phase for the ad-set.
Then I reduce this bid by 10% everyday until my ROAS keeps getting better and stop when the spend starts going down. This has helped me identify the right manual bid in the past. But there’s one drawback and that is the auctions change everyday and I only run the top down bid-identifying strategy once. So my manual bid may not be the most optimal manual bid everyday in the future.
However, since the launch of campaign budget optimization (CBO), there’s a simple solution to this problem. You can create multiple ad-sets with different manual bids and place them in a CBO. So if you’re selling that $10 product for $30, you can may be create 5 ad-sets in a CBO with a bid of $13, $14, $15, $16 and $17. The CBO will automatically choose the ad-set that’s likely to get the best results and each day a different ad-set with a different bidding may be getting the sales for you.
Facebook launched campaign level budgets in the mid of 2019. Initially, I was skeptical but I’ve started to like CBOs a lot. By using campaign level budget, I can now test 5-10 adsets in the same budget that I needed before to test 1 adset.
Facebook simply spends higher budget on the adsets within a CBO that are more worthy of my budget and spends lesser budget on adsets that are more likely to burn cash.
There is always a risk of missing out on a potentially winning adset but the reward overshadows the risk. In addition, you could still define minimum spend per adset within a CBO to ensure that each adset gets a bare minimum spotlight. Although, I generally advise against that.
My only problem with CBOs thus far is the organizational structure. Prior to CBOs, I only had to create 1 campaign per product. My campaign could then have hundreds of adsets.
Now I’ve to create multiple campaigns per product with each campaign grouping similar adsets together. Because of this, I’ve to create 10s of campaigns per product. The downside is I can’t group together data for 1 product without using filters which is just an added inconvenience
If Facebook introduces something which is above the campaigns level only for sake the of categorization, I’d really like that.