Home > Paid Traffic Sources > Facebook & Instagram

---

Looking for help with targeting and bidding strategy (5)

---

Hi all, I’m new to STM and excited to be part of the group! I’m hoping I can get some advice on targeting and a mock campaign.


- Regarding campaign structure: is the ideal structure to have 1 campaign that contains 1 ad set with several duplicate identical ads inside and repeat the process to create numerous campaigns that follow the same pattern ?

- is optimizing based on “conversions” the best option when starting to test a new campaign ? Let’s assume it’s a nutra campaign with a $40 cpa. Also there are options to use “bid cap” and “cost cap” - I’m curious if you utilize these or do you leave them blank?

-from a targeting perspective, some are saying to go very broad and some say the opposite. Which is the ideal way to test assuming pixel is in place and “purchases” are tracked on the pixel ?

For example, let’s just use skin care as an example targeting Canada.

Here’s my audience target:

Women age 40-65
Married status
Behavior: engaged buyers
Interests; skin scare, beauty, Sephora, aging, alternative medicine

Is this too broad ? Curious if anyone can advise on how they may approach refining this audience to weed out the bad quality users ? Also when I use the Facebook insights tool and type in “skin care” the pages liked by users are also stores like tj Maxx that have high affinity. Does that mean you would target these stores in interests as well even if it’s not directly specific to skin care ?

Advice greatly appreciated !

---

The 'best' or better structures for campaigns comes a lot down to what you're running first, different pixels and targeting options are best for different verticals.

Casino/sweeps - lead pixel, nutra sales - purchase, ecom - Initiate Checkout etc

For skincare, I would say go even broader. You don't need to worry about interests, most people in the world are self conscious for their skin. Age targeting higher is usually a measure to keep quality for advertiser as well, a lot will come down to creatives and also luck of the FB audience pool you receive hence the need for multiple adset testing.

When you have working creatives that is an effective method of scaling yes, although you could have multiple adsets inside the campaign, but for skincare this is a perfect world scenario lol, a lot comes down to the process of running the account....

---

Thanks Maynzie! Appreciate the advice, and that totally makes sense. No need to over think the targeting since as you said, aging/skin care is a concern of a lot of women

---

Hello rawkuspave, nice to have you here. Sorta new here myself, but let me try and answer your question the best way I can.

Yes, thats a good structure, but it also depends on your goal. Are you testing which creative is working best? Then you should probably do 1 campaign, multiple adsets with 1 ad inside of each, so that you are FORCING fb to spend equally on all creatives. Are you testing different audiences? Depends completely on your goal. Usually what I do is create 1 campaign, multiple adsets (different audiences) and multiple duplicates of variety of ad creatives.

Depends on how mature the ad account is. If its a completely new vertical you are going for, I would say go with traffic, PPE to get some cheap data. It can also help you get an idea if your targeting is correctly setup. But yes in general conversions campaign by far are the best. Once you start getting 50+ conversions a week, you will see tremendous stability. For bid cap and cost cap, I usually use it to have controlled CPA and mostly when I am scaling campaigns. If you are starting out, just go with auto bid for most part, because you want facebook to spend as much of your budget as possible to give you the max data.

For an audience I would suggest atleast 1M min size audience. That can help you get lower CPMs overall. But if you are spending like huge money ($2-5k/day) then you might burn that audience real quick and would probably need fresh creatives running in regularly. I feel that targeting is going to be less than 1M in size? But at the end of the day, just test it out. If you are getting good conversions with an audience size of 400k, just roll with it. The end goal here is to just make more money

For high affinity pages, try out different adsets to see how they are performing.

TIP: Run Video view campaigns on very targeted videos that are longer in length. 5-10 mins in length. And then retarget people who have watched it 75% or more. These are potentially your target audience that fb doesn't know about, so that can be a huge bonus for you. A lot of times I have seen traditional targeting to fail compared to VV retargeting. Once you have enough custom audience on that, create a broad LLA, add some broad interests to filter, and use that as well. I am sure you can see how strong this audience would be. Hope this helps

---

Thanks @anshdeb!

Regarding LALA, I'm having an issue...

It looks like I never set up a "Purchase" Conversion value, but the FB Pixel did capture "Purchases" and they're being reported in Facebook accurately.

When I tried to create a lookalike audience, it gives me an error because my Purchases values are $0 and won't let me create it.

I went ahead and created a Custom audience based on "Purchases" instead, and then when to create a LALA based on the custom audience instead / those individuals who made a purchase but I'm not sure if this is the right approach.

Is there a way to handle this considering I did not assign a dollar value to each conversion in the Pixel, but the Pixel still tracks properly the purchase?

---

Home > Paid Traffic Sources > Facebook & Instagram