Guide to Meta Ads Targeting Part 3: Remarketing

[00:00:00] remarketing an audience we like to use a ton is the uh all CA so all custom audience so kind of all the warm audience we can we can create think of kind of thing so um as far as we can go for all of them so visitors past six months let's go over remarketing so remarketing an audience we like to use a ton is the uh all CA so all custom audience so kind of all the warm audience we can we can create think of kind of thing so um as far as we can go for all of them so visitors past six months engagers Facebook and uh Instagram a Year followers there's no date it's just like the the followers all the email list all the subscribers uh even if you have a list of suppress that you can upload legally you could use that video viewers so uh often we'll separate you'll have like in a separate adset the video viewers because that would be sometimes like the bigger audience and it's sometimes not super targeted and we kind of

[00:01:11] put have it too much of a big percentage of the audience uh for for this one so uh for this one often we'll try to separate it uh it could be you know for video views you could do like 3 seconds I think 7 seconds or 15 seconds type of thing or percentage 25 50 75 uh 90 and 100 something like that so if you include it in there I probably go you know at least 75% watch uh just so you have uh decent targeted uh people so it's not like 80% video viewers and just 20% of the rest that's kind of what we want to try to not use uh for these guys so to create pie uh it's kind of the same thing as I wish showing inside the look like audience so for the website uh anything you want to do here in in terms of like all website visitors that's already there just 180 days then um if you want to do uh followers um yeah so whatever the page but um here you can have uh everyone who engaged that would be all of the one who

[00:02:25] um you know click the call to action button message but that'll be included All U all of those will be included in this one that's all the engagers that's a year old we can go back and if you want to do the followers you could just select that one this one doesn't have any dates so that's same thing for uh Facebook and Instagram very similar uh email list then you can connect clavio it's an integration or you can just upload uh so if you want to upload you could just go customer list and follow the steps uh right there um meta always push to have like um some value in there you know zip codes and whole numbers and whatnot I think email just work just fine um for the match trate so yeah either email or both email and phone that's what I would recommend uh for lead if you're uh having a a lead gen client that use lead form that's cool thing that anyone that open the anyone that opened the form or

[00:03:29] uh any that uh submitted the form that could be good and then if you have Leen too but it's on the on the web page usually it will be lead so this one we don't have the the event but if you you have an active event you have leads that you can select there um yeah and then video views so for video views uh right there so that's kind of all the option you can have uh 3 10 15 and then the different percentage let's say do and then you can go back to a year for these twoo and then you kind of have to uh can have to uh select each one by one uh that have enough views and kind of scroll through the pages uh you can go up to a year so it can be quite a long process and you can update it uh you know for sometimes but it's worth doing um and again for these uh ATT 10 to separate it from these so it's not uh you know all video views in in your all war audience and then um one that we see working really well you know each time

[00:04:41] there's a sell and whatnot you know these will usually be both uh the audience that work the best is the the recent one so really recent will be like engagers of past three days so anyone that like comment follow or click the button on the page and whatnot uh on Instagram and uh Facebook so they just they just interacted with with one of your ad past seven days so they visited the website in the past seven days so maybe they they were in a rush or you know they were waiting for payday or something like that so uh you know past seven days is is good to keep the one that are that forgot to to purchase or got distracted or say I'll I'll purchase later because of X reason um and then these could be just together but we like to often uh putting the people that added to cart you say check out in the past 30 days exclude the the purchases of course the past 30 days and you can also share an AB and

[00:05:40] then check out uh from clavio past 30 days or something like that or just recent subscribers of the past week would make sense to just you know something that recent usually works really well because you get the lwh hanging fruit and the people that just kind of were about to purchase but forgot for for some reason and sometimes we'll do ibrid too especially if it's a small um a small brand uh I sometimes I'll I'll put in say check out n cart inside the recent and even the buyer sometimes in there so we have really like a an hybrid of uh people that were that either purchased just recently uh or you know added to cart and then just the recent one so these usually work really well but a large enough brand uh tends toward separating those to so just have buyers by themselves sometimes want to talk differently to to your own buyers or have this special offers or special ads so often it makes sense to to

[00:06:40] separate them and then uh what you can do too for remarketing is you can do a product or collection specific so again you can select just the the product ID of uh people that view a certain products or collection URL just a URL that someone that went went on that collection and you can also do some fancy stuff with the the number of time they saw this product or the the time spent we usually don't mess a lot with those but um it can be it can be a good cool thing to to explore um especially if there's there's enough data um that's that's a key part there and then uh exclusion so exclusion if you want to go kind of stone coal would be excluding all of our warm audience right there so uh if you have a prospecting audience or uh interest or look likes and whatnot if want to make sure to really go after uh people that don't know the brand or you you you can of have to exclude those but you know

[00:07:43] one thing to keep in mind when you do that is sometimes you'll restrict the algorithm from making a purchase so it tends toward excluding all of those when um you know you're spending a lot or we have high frequency um stuff like that because uh often you know the algorithm should be smart enough to even if it's it goes a little bit in inside the the warm audience for prospecting I don't think it's a big deal you know what matters is the results for the client so yeah that's kind of the the idea here if you want to go Stone Cold uh either you know the brand is is not growing new customers or it's spending a ton um I would then go with excluding all the worm otherwise you can exclude just the buyers or not again sometimes you know buyers will uh comments good good things about their experience that'll bring social proof uh but sometimes you know it's cleaner it looks better uh from a client side to if you exclude the

[00:08:39] buyers so yeah I put it or not here because uh sometimes it's it's good to keep it open just so you get that uh that feedback that if customers uh can comment uh that can that can help quite a bit so that's kind of what I have in terms of um audiences is there is there any question or specific You' like uh me to go over for the last few minutes leave it open to you guys the only thing that I would say just I should have jumped in earlier with this but sure the for the ASC campaigns um kind of on topic of that I just for everybody's kind of so everybody can get a sense of how we test typically I find that when we're testing and we're trying to find a winning audience it's not the best to have the ASC or the Advantage Plus option selected because you don't necessarily know if Facebook is using your target audience or if it's stepping outside of your target audience so you don't know if you did the right thing or if the

[00:09:51] algorithm is making your choices seem better than what they actually are so typically I'll test with like a 10 % look alike like Jonathan was saying and then once that produces good results then I'll create a new campaign with the same audience and turn on the Advantage Plus for like scaling opportunities yeah I think that totally makes sense um yeah that's a very good point to um to mention start because it does work really well it's just you don't know if you're doing the right thing or if Facebook's doing the right thing right yeah and that then where yeah starting with without the advantage plus too just for the for the reason you mentioned too and and also for the you know if it's a new brand or they don't have a ton of span you kind of want to just start kind of limiting a little bit the the algorithm kind of give it Direction but not too much and then go from there yeah I guess that brings up

[00:10:48] another good point that a lot of the times when a business comes to us their algorithm has been trained to do something that's not necessarily of the benefit of the client because they're coming to us because results weren't great in the first place which means that their previous agency was doing something wrong a lot of the time it's they've trained their algorithm to go after a retargeting audience that's what I'm finding a lot recently and so they're they've been running this setup over the past year that's like 75% retargeting just because the agency wants to show the client good results on platform and then we get handed this account and it's trained to do the opposite of what we want it to do so that's just a total side note but letting clients know that when they bring us account that's an account that's in a bad standing it's going to take a little bit longer for us to turn the ship around

[00:11:46] basically yep that's a good point uh often too they'll I'll get I'll be trying to link clicks or engagement or um you know often they're they have like sort of personalities or they have like patterns so so they're used to spending you know $50 but now they want to scale to $500 per day and the account is kind of just the algorithm is not used to that and that's when it will produce like kind of shaky results when you try to scale too fast um yeah they kind of have personalities or you know Based on data too um you can kind of see it from these three about 11 months ago that's when I posted I'm like why I'm moving almost all my clients off pmax um and then it's funny is like then like nine months ago um that was like when we show on pmax whichwich is standard shopping that was n month I mean almost on the same timeline as you're like hey this thing's starting to become less valuable the same thing that I saw