Why this way?

[sumaiyakhatun26]

This in turn means that a really good test idea is to set up ad sets that target similar groups and interests separately.

Of course, appropriately selected – for example, according to the model I discuss in the episode on how to select interests in Facebook advertising .

Because it may turn out that similar groups that you use will achieve worse results and you will not know about it.

A niche, well-selected interest of an appropriate size is better than a similar group based on all visitors to your website.

Why?

Because if your website is getting a lot of low-quality traffic, people who don't france rcs data convert and don't delve too deeply into the site (which is a common problem, for example, when generating traffic through SEO channels or B2B mailings), it suddenly turns out that such a group, theoretically the best semi-automatic targeting method by the creator of the algorithm itself - Facebook - will work worse than manual interest selection.

So the machine won't always be better for us, because it's only as good as the quality of the data it's based on. And at this stage of our business, they're not always good enough.
So test how interests vs. similar groups work for you. Our experience shows that interest-based groups convert very well, even in cold traffic, especially in e-commerce. Something that many advertising specialists seem to forget or stop taking into account over time.

YouTube Artur Jablonski

#4 Locations
In the vast majority of cases, the best choice is automatic placements . I also recorded a separate episode about this, to which I refer you for the argument "for".