Home > Paid Traffic Sources > Native

---

widget bids (14)

---

HI all,

i am confused by something . Take an offer that pays $30 perfect for native.

when starting out on revcontent, you take your AMs advice and use the blacklist.

you then cut placements that are x2 or x3 the payout.

here is where I am confused - (in general with AM).

revcontent has thousands upon thousands of widgets. To test out this $30 offer could cost you 30K based on this methodology.

Are people also cutting on High CTR, Low CTR, X impressions no conversions etc....

how are people managing costs so they dont spiral out of control - a lot of people will say test different offers, how can you possible do this when it costs potentially 30K to test one?!

advice please

---

You're right it doesn't make much sense. What if the payout was a $100 or even $500? Not only are there thousands of widgets, but many of them will never even reach 2-3x spend of offer payout. It's essentially a very basic optimization algorithm that's only suitable for low payout offers.

So you have to go based on the next best thing - offer EPC and your effective cost per click (eCPC). Alternatively for e-com that would be add-to-carts.

In this way you optimize your ad spend to block sites that have eCPCs well above your EPC, and you can also block sites faster that have say zero or few clicks after $10-20 spend etc.

This is all assuming you've established a baseline EPC to work with though.

---

thanks!

that is interesting and something that I have not thought of. So essentially if you have an EPC that is say $1 you would just cut the placement at say x5 or x10 of that?

would you ever cut placements based on the EPC of a network? i.e. if an aff network tells you the EPC is .8

---

Exactly. If you had an EPC of $1, then cutting placements that have a $5 eCPC after say $10-20 spend instead of waiting to spend 2-3x offer payout would be a more efficient use of adspend.

And no I'd never go off network EPC. What someones cloaking on facebook isn't going to perform the same as with whitehat native.

---

Hello,

How do you find out the more likely EPC and eCPC?

---

As for the eCPC, you need to run for a while to see what you are actually paying for click. If it's a value that seems to high for the offer payout, it's pointless to continue with such placements. As thedudeabides mentioned, when the cost per click is higher than your earning per click, you cannot profit. Since you don't know what the EPC will be like at the beginning, you need to make a qualified guess, that is NOT POSSIBLE to make without some experience or reliable data from the AM.

That is why these higher payout offers are not suitable for fresh newbies. Experienced affiliates already know what kinda CVR they can ROUGHLY expect for such offers ... so by doing some match, they can set an EPC to aim for. You need to set some tolerance here still, since the final result will depend on the optimization.

It's very helpful when you have an AM that understands the various traffic types and can give you rough numbers that relate to the traffic type you are working with.

---

So if I am understanding correctly - very roughly speaking a 30$ offer, IU should expect to spend around 500-1k just to see if the offer even works?
and then if it did, optimize accordingly.

---

Using the initial information you have from the affiliate network for the offer you are pushing, as well as your current CPC you will be able to make an approximate calculation of what performance thresholds to expect in order to break even. With this in mind, you don't necessarily need to keep on spending up to $500-$1000 before you start optimizing your campaign.

---

With good targeting, you should know much sooner than that.

Best case is to get a custom whitelist from your rep. I've had campaigns be instantly profitable that way, but you're going to have to build up a relationship with them for that to happen.

Second best would be to build one from spy tools. Third would be brands or push traffic. Fourth would be a given category.

Fifth is probably just going with one particular brand site or large site. Smallest possible spend/day that way but the traffic is not going to be necessarily representative of what works across the entire network.

You should try and test at least 2-3 verticals that way before picking one to put more money and focus into. Doing so will increase your chances of getting a winning campaign sooner.

---

Thanks, some really killer responses on here and I cant tell you how appreciative I am (I can just type it! )

from revcontent I have a blacklist the the rep gave me which should help.

Where else I am consfused, say I choose a big site, one that i have found from a spy tool (say breitbart)

If i go for a top 5 position it will just hammer my budget. How much do positions matter to conversions. i.e. if I am managing my budget in position 14 will that still convert (hard question i know) or do I need to be in position 1-5 to see th ose conversions. If that is the case, how do I rrestrict the budget to less than 100 per day with revcontent?

thanks

---

Positions are important for ROI. You generally don't want to be 1st position because that likely means you're overpaying.

I don't even think there is a position 14 for revcontent. Lowest I've seen is like 8. Your ads will stop serving at some point if your ad performance is that low, because it's not making the site owner any money.

If you want to restrict your budget to less than 100 you could start out on 1 or a few widgets I guess. I don't like to approach native that way though.

I definitely wouldn't recommend starting out with a blacklist campaign. Unless you're doing a small Geo youll get crushed most likely. Blacklists are for when you have a solid campaign and want to scale more.

---

I Just reread the thread. Regarding a whitelist, so I did this - I built up around 15/20 widgets that I found on placements in adplexity. I did get a conversion out the park however the cost was prohibitive when i was in position 1-8. How many sites would you recommend having in a whitelist and would you build a whitelist per offer?

---

Some great golden nuggets here, thanks @thedudeabides.

---

Any progress @aady83???

And yea there were alot of good info in here thanks to @thedudeabides!!

- Jeremiah

---

Home > Paid Traffic Sources > Native