Traffic arbitrage not going well :(

[wrasse123]

I am sending traffic to a site through Facebook. I am getting $0.09 per website click for my campaign, and each user is going through 7-8 pages per session, but I am still losing money or just breaking even.

I am using Adsense, native ads and display ads to monetize the impressions.

Any guidance would be appreciated.

 

[wrasse123]

I have worked with the placement to no avail. Could it be the number of ads? It's quite intense right now. The thing is the same placements used to work before but dont anymore

 

[Deleted member 813988]

You can get cheaper clicks from Facebook, that's for sure
What's your total CPM?

I would first figure the formula of CPC/CPM/Pageview so you know where to head next.

 

[arlind-]

What ROI you getting when paying $0.09 cpc?

 

[Deleted member 813988]

also, how much traffic do you buy? is it just hundreds or thousands (and during what period of time)

 

[bigg_capp]

The Facebook algo change has made arbitrage that much harder; that being said, you should be able to get much cheaper clicks.
I used to get .008 clicks, now I'm getting .018. All US traffic. It's gone up more than 100% since June 4th.

That being said, you should easily be able to get lower than $.09. I would work on that end instead of the ad end.

-Capp

 

[wrasse123]

@se900se On Facebook I am targeting only US desktop users only. I am using manual bidding and paying per click. Should I be getting lower than $0.09 per click?

Though I don't have the exact metric for the earning as I am using multiple ad networks, either I have to get cheaper clicks or more $ for impressions. The thing is I don't think I can do any better on either front unless I cramp even more ads which I don't think is wise.

@ arlind Breaking even or making up to 40% profit on good days. Also if I try to scale profits drops as a lot of impressions go empty.

 

[bh9889]

Keep optimizing to keep your CPC down. Also if you don’t already, try making slide articles.

Top 20 worst tattoos ever - make the user click next for every picture so you get more impressions.

 

[Deleted member 813988]

@se900se On Facebook I am targeting only US desktop users only. I am using manual bidding and paying per click. Should I be getting lower than $0.09 per click?

Though I don't have the exact metric for the earning as I am using multiple ad networks, either I have to get cheaper clicks or more $ for impressions. The thing is I don't think I can do any better on either front unless I cramp even more ads which I don't think is wise.

Yes, you can get it way cheaper (at least 50% cheaper).
By total CPM - I mean that if you combine CPM rates per all networks what would that be?

This way I would be able to calculate 'success' formula for you that shows what kind of performance you need from traffic (considering it's price) to make a profit.

There's a number of ways to adjust both CPC and user behavior towards higher earnings, but you need to have a plan/goal first.
So let's say your Total CPM is $10, then you look minimum at 9 page views to breakeven. Your current total CPM would.

 

[wrasse123]

@se900se I advertise on Facebook daily but only scale when I see a profit. But I am testing every day. Spending around 5 k or more or less monthly depending on the ROI.

@bigg_capp Yes, I know its gotten more expensive I can get CPC of 0.08 but can't maintain that CPC while scaling.

 

[Deleted member 813988]

@se900se I advertise on Facebook daily but only scale when I see a profit. But I am testing every day. Spending around 5 k or more or less monthly depending on the ROI.

I got that part, there are not so many people who scale when it's loss
So that's on average 50k+ people and 300k+ pageviews?
Are you running DFP?

 

[wrasse123]

@bh9889 Yes, I use paginated articles.

@se900se I can get cheaper CPC but then the page views go down. Cheaper traffic equals lesser page views per session at least for me.

 

[bh9889]

I personally think PPE ads on

@bh9889 Yes, I use paginated articles.

@se900se I can get cheaper CPC but then the page views go down. Cheaper traffic equals lesser page views per session at least for me.

Have you tried running PPE campaign instead of CPC?

 

[Deleted member 813988]

@se900se I can get cheaper CPC but then the page views go down. Cheaper traffic equals lesser page views per session at least for me.

I'd suggest to stop worrying about the traffic price for now but work towards increasing CPM rates and your whole ad setup.
You're not playing lottery and can make it work your way. The point of traffic arbitrage not to have the cheapest clicks, but to have that kind of performance from traffic to get higher payments.

 

[wrasse123]

@se900se Yes, little more than that. No, I am not using DFP presently I am using waterfall right now. The ad network I work with does use header bidding. Please see a sample page metrics here https://imgur.com/a/moxjVMB

 

[wrasse123]

@bh9889 Yes, but I get better results with CTW.

 

[Deleted member 813988]

@se900se Yes, little more than that. No, I am not using DFP presently I am using waterfall right now. The ad network I work with does use header bidding. Please see a sample page metrics here https://imgur.com/a/moxjVMB

Looks like you have multiple images per page + pagination (judging by the 20 sec load time and 3-5 minutes per user).
If that's so - you may want to add fewer images per page, more pages.

Headerbidding - is not waterfall and works differently (waterfall where networks go one by one depending how they setup, headerbidding where networks bid for each visitor), so I guess you have headerbidding and pay a company to manage it for you if you not dealing directly with ad servers and programmatic coding (you can set it up yourself and not pay anyone for that part of your earnings).

No matter if that's headerbidding or a waterfall you can add additional networks that do not work with headerbidding or waterfalls, like video ads ($5 CPM +/-) and 100% viewable ads, those that only appear for certain kind of traffic but they pay a lot ($15 CPM).

 

[wrasse123]

@se900se Yes, I am using multiple images per page as it makes the page longer so I can fit more ads and the user stays on the page longer thus increasing the session time.
If I use one image per page my bounce rate increases and the page per session goes down as well.

Yes, I am aware of the difference between header bidding and waterfall My apologies for not framing the answer well.

Yes, we work on a revenue share basis for header bidding. I don't have the numbers to signup with DSP's directly for prebid

 

[Deleted member 813988]

@se900se Yes, I am using multiple images per page as it makes the page longer so I can fit more ads and the user stays on the page longer thus increasing the session time.
If I use one image per page my bounce rate increases and the page per session goes down as well.

Yes, I am aware of the difference between header bidding and waterfall My apologies for not framing the answer well.

Yes, we work on a revenue share basis for header bidding. I don't have the numbers to signup with DSP's directly for prebid

Single image probably would not be a good idea, especially if a user sees 1 of 999 , but if you use 5 now (just example) - you can try 3-4 and check the result, you may want to set up a separate slide/pagination post to test it (can be the same post as you use now, but with a different structure).

The numbers on the screenshot are good to join some of the DSP's and you have something to start with, but I would do that only in case you know what you doing/know how to do the setup.

 

[Deleted member 813988]

Depending on how your FB ads setup - check in your GA what users reach the final page of your article and get more of those ads/sources.

If you using GA only - you may want to use additional analytics to track user behavior, helps to adjust ads, positions, type of ads etc.