Generally a decent strategy is start big and work down to niche if you want to spend as efficiently as possible. Example would be get Google, Apple Search, Meta, and maybe tik tok up and running. When you can buy effectively in these channels, you then look at new channels. Applovin, Snap, Twitter, IronSource, liftoff, etc. Once you can buy effectively on these tier 2 channels or have filtered your partners down to those that you can buy effectively with, you would start looking at tier 3 or niche channels. I would consider Reddit to be a tier 3/niche in this instance. Other less established ad networks or smaller niche websites would fall into this category as well in this example (Pinterest, hobby websites, so on).

The larger networks are generally more efficient thus starting big and working down is a fine general strategy to scale efficiently. Google has traditionally been stronger as a CPI or CPM optimizer whereas Meta has been stronger as a CPA network historically in my experience. If you can’t buy profitably with the tier 1 networks you likely won’t be able to buy profitably with smaller networks. At this point, your targeting and creatives are going to be more important than your choice of channel. Iterating on your targeting and creatives as cost effectively as possible until you find a mix that’s sufficient for your goals (regardless of channel) would be my recommendation.