Home > Paid Traffic Sources > Native

---

revcontent bidding (4)

---

HI All,

after running a lot of facebook campaigns, I am taking the move to get some campaigns profitable in native and expand my arsenal!
I am running on RevContent.

I spied for an offer, noted the top 5 placements to create an whitelist modded an LP got a campaign up.

$100 in I am at -50% ROI. which I think is pretty good from what I am reading. I am looking for some help around the bidding.

a) would you ever whitelist to ONE widget if it converted.
b) what positions would you bid for to receive decent traffic and see if something is working.

essentially thats all for now

thanks

---

Not unless it was a huge widget, eg a brand widget.

You need to test more offers though. Just because an offer gets a few conversion across a list of widgets, doesn't mean for example that those widgets are the winners and to go all in on them. It can be deceiving, and the overall ROI is generally the better. Exceptions being tier3 traffic which has a lot of junk.

I don't even think about this. I just test on a sub-set of traffic like a whitelist or brands or push, and find what offer and vertical is doing best. Bid high at first, then lower then as conversions come in, until you find a sweet spot.

---

Hi Dude,

Got a few questions to make sure I am on the right track.

We worked on a lot of pop campaigns and recently moved into native. So I might be missing some of the more obvious native knowledge.

I am using MGID.

Running Nutra in Tier 3 Geos in local language.
I get that most of the traffic is from streaming / file sharing sites.
Any tips for these GEOs will help.

Or, Would you suggest testing Tier 1 / 2 Geos on Native for better performance and quick results ?





I tried running only on few whitelisted widgets that showed conversions. And as you said it didn't work out.


When looking for performing widgets - are we looking for widgets with positive ROI or go with higher conversions and play with bid later ?


How much would you spend to test : a ) the offer - 5x, 10x, 20x etc payout ?

b ) each widget before you decide to cut it ?



Bid high on the teaser and then reduce the bid on non performing widgets ?


If I am getting decent traffic from a widget, will increasing the coefficient on that widget help in any way ? I fear it might increase my costs without any positive results.


Regarding landers - if we are getting conversions with negative ROI.
Would you suggest working on improving the landers or trying new landers for increasing the conversion ratio OR focus on finding better peforming widgets.



Could you please explain the part about 'brands or push'.

And how do we determine which sites the widgets are hosted on to know if it's a brand or a 'junk traffic' site.

---

I was referring to Tier 3 traffic on Revcontent, not tier 3 countries, meaning the traffic doesn't convert as well and gets removed from tier1.

What applies to Revcontent doesn't necessarily apply to MGID.

Regardless, stick with smaller geos if it's your first time on native and learn what works first before taking on bigger countries.

Depends on the the stage of the campaign, early & unprofitable vs later & break-even or profitable. Early on I run on traffic I know does well. This comes with experience on a particular source, but your rep can give you a whitelist or blacklist to start if there's too much traffic to sort through.

I don't focus on individual widget spend too much. It doesn't make sense to spend 5x payout on every single widget for a $50 or $100 offer. You want to look at overall traffic performance to judge if an offer/campaign is good enough to run at scale.

You can also use effective cost per click of offer to cut placements faster. If an offer overall say gets $2 epc, then if you know it costs say $10 to get that offer click from one particular widget, that widget isn't likely to work (eg $30 spent and only 3 clicks on the landing page).

Bidding high in the beginning is good because it sets up your ads for better chance of receiving a good CTR via the platforms algorithm and also exposes you to better quality widgets that may convert better. Of course once you have some converting widgets you can play around with their individual bids.

You should always having a split test or multi-variate testing (if you have lots of conversions) running on landers once you find the winner. It doesn't take long to find and test landers on native - everyones been running them for a while most likely.

Those are targeting options on Revcontent.

Overall I'd advise you to rely more on your rep than spy tools to tell you what's working if you can establish a good relationship with them.

---

Home > Paid Traffic Sources > Native