Cost Cap Bidding is incredibly powerful, and very often used by people who wanna see some success… But it’s also almost always used terribly improperly.  

Cost cap Bidding is a way of telling Facebook please don’t give me people they cost more than that. This is the data that you already have inside your account as well as Facebook‘s information on billions of people in the world. Cost cap Bidding relies upon this information to effectively predict the future by taking informed risks.  Now this sounds amazing, and can absolutely help you control your price that you pay. However very often this is used as the primary method of delivering consistency and scaling results.   While this on the surface makes sense, there’s also a reason why it’s highly unstable, bringing diminishing returns, and struggles to spend, it forces you to hack your page and make adjustments constantly, which keeps the ad set an a learning phase, which prevents you from having any confidence in the outcomes.  So, why does Cost Cap have work… Cost Cap works, Here in limited fashion, because you are restricting the amount of choices that Facebook can make. This restriction allows spend only in the places of highest conference to achieve the specified goal. The problem is this is an incredibly unsustainable ask.  If you’re telling the system to go after only the top 10% (Cost Cap bidding), of the 5% of the audience it actually thinks is likely to Convert (Conversion Optimization), by the time you’re in week two you are reaching far less than 1/10 of a percent of your audience. Now, if your audience updates and expands, or if you’re using very big audiences, you can mitigate this decline.  Very often people double or triple down on this effort to “hack“ the system and relying on multiple predictive elements to see success. Cost cap bidding, Is one, lookalikes and seven day optimization windows are others.  Remember, a lookalike is an audience where you are restricting the available choices in hopes of predicting somebody’s behavior because they appear to behave in a way that is similar to some source audience.  Remember, seven day click, or 7day click and one day view, is using data to again try to predict what someone’s going to do in a weeks time.  If you were to lay a cost cap bidding, which deprecates very quickly, on top of a look-alike audience that is a low risk gamble, on top of a seven day click optimization window which I asked you to not only protect somebody’s behavior today but for the next week… Is there any shock that the returns get bad really fast?  So how are you supposed to use Coat Cap?  Properly used, Cost cap bidding provides a source for incrementality when times are good, and stability when times are bad.  Cost cap bidding, Should never be even close to half of the money that you’re spending, and should have a very specific short term goal because it will turn from an asset to a liability very fast.  Cost cap bidding, is a tactical solution to a short term problem.