Home > Paid Traffic Sources > Adult Traffic (NSFW)

---

How to Get More Volume, When Testing in a Smaller Geo? (23)

---

Hey guys,

This might sound like a real newbie question.... but I understand that in order to really collect data to be able to make solid decisions and learn this game you have to spend more and get a decent amount of volume.

If I'm running a campaign in a GEO with let's say..... 20M ppl.... and I'm running this campaign on 1 spot on 1 site.... how do I SPEND let's say... 100-200/day to test??

Or should I run simultaneous campaigns in multiple spots on the same site? That wouldn't make sense cause obviously every spot performs differently, right?

So how do I test with significant volume in a quick time period if the GEO I'm running in isn't..... as BIG?

---

The short answer is you can't.

If you're in a small geo it has been my experience you are at least going to need multiple spots to get that kind of volume.

Or you could try to bid really high, but I don't see how that will help you as that data wouldn't be really usable as your CPC would be astronomical.

---

In bigger GEO's i would recommend aspiring to a win ratio (Imps to Click) of 10% max or you will burn your budget otherwise. But in smaller GEO's you would want to aspire to a win ratio of 50%. You don't have much traffic to begin with so you would want to win as much as possible. So highest possible bids that still bring you positive ROI. Open your targeting as much as possible, Target all devices and see what you can do running like that.
Good luck!

---

Yup, as jabong said, in small GEOs you can't really spend a lot in a single spot, since the traffic simply isn't there. You need to target multiple spots in order to drive the volume up.

Yes, different spots perform in a different way, but that doesn't mean that the same funnel can NEVER work in multiple spots. Quite often it does actually, or a small tweak will make it work in a different spot.

Working with smaller GEOs is a bit different from the large ones, you simply have to accept the nature of it and give it the time it needs. Sometimes I need to wait a week to get actionable data and there is nothing I can do about it.

But then again, even the smaller GEOs still have enough volume on the largest sites, for example xhamster or xvideos ... even with GEOs under 10 mio, you should be able to get decent traffic on these mega tubes.

---

Thanks guys, I guess my question was really around how do I collect enough data to test and optimize campaigns in these smaller GEOs?

I guess the answer is "you just have to wait several days for the data to gather"

---

What do you mean by 'win ration' Impressions to clicks of 10% or 50%?

Let's say I get 100k impressions with a banner CTR of 0.2%... that's 200 clicks... what do you mean by win ratio? Thanks!

---

I guess with win ratio he means how many impressions are available for your campaign vs how often your creative is shown.

100k impressions available and your banner is shown 10k times = 10% win ratio
5k impressions available and your banner is shown 2.5k times = 50% win ratio

The higher you bid the more bids you win.

When there is huge volume available (big geos, wide targeting and so on) you can bid lower and still win enough bids to have good volume for your campaign.

But when there is not much volume available anyway you need to run high and competitive bids to win more bids = high win ratio to receive enough volume.

---

Gotcha, that makes a lot sense thanks!

---

One more note regarding this... if possible, do not go over 50% winrate +/-, the higher you go, the bigger cap you need to use so you end up showing your ads to the same users over and over. It's ok to do this with an optimized campaign that runs at decent ROI, but it's not something you want to do from the very beginning.

---

Does the win rate include traffic sectioned off for block buys?

For example... I'm running on a spot with about 1M total impressions/day

500,000 are sectioned off for block buys

500,000 are for bids

I'm currently getting 200,000-250,000 impressions out of that total.

So are you saying stay around where I'm at while testing, or try to get closer to the total 500,000 available for bidding?

I'm on 2/24 capping right now, and if I opened it up would it REALLY make a big difference? I imagine I'd probably get close to 400k impressions... which would allow me to gather data faster.... why do you say to not do that? (in this specific instance? I understand if there is lotsss more volume you could end up spending more money than necessary...)

---

Depends on how they distribute the traffic. In case the block buy get's a balanced rotation, you might want to take a bigger part of what's left. By balanced, I mean that they don't just get the first impressions, but a balanced mix of everything. But you don't really get to know how they distribute their impressions. So you need to test.

I prefer to stay around 50% max, because in order to get more, you usually have to open the cap or not use it at all. Which means you will be showing your ads to the same people over and over again. This means lower CTR, thus more expensive clicks and lower CVR too. You can still do that, if the funnel is strong enough, but make sure you have a lot of banners in rotation so even when they get displayed to the same people multiple times, it's always a new ad that they get to see.

In your particular example, since you're on 2/24 now, I think there is some space to get more volume without harming the quality too much. But don't go for 4/24 or more, try to lower the "hour count" so something like 2/6 etc... In my experience, it's better to show your ad to the same people after some hours have passed, instead of doing it in a row.

---

Thanks @matuloo, makes a lot of sense!

---

Hey, great question.

If you are new to Adult its very hard to pick test campaigns. I understand that, because I have been there.

After some experience, you will be able to identify test campaigns quickly. I usually pick campaigns that I know work decently, and has good volume. And I have test campaigns running almost everyday of the week (Almost).

Start obviously by doing one geo in one traffic source. Choose the geo that is working really well for the aff network, and that has decent volume and smaller payouts. That means, no Ireland(Low Traffic), no Malta (Low Traffic), no Nordic (Higher Payouts). Pick something like Switzerland, Austria, Canada. These geos are not cheap but has good offers, low competition, decent payouts and good amount of traffic.

Then you test the fuck out of that geo. Start all spots in the geo.
Like you said, performance is different for each spots. NTVs has a different CTR and Bids compared to Top Right for example.
But the particular Angle-Image combination should work similarly for all spots.

So lets say you want to test 10 banners: These 10 banners have different images (I1, I2, I3 ... I10), but same angle (A1). Name them well. A bit like A1-I1, A1-I2, Ai-I3 .etc

Test 5 banners in each spot. And run them in +5 spots, total volume of these campaigns must be like $100 a day. You can easily achieve $100 a day with +5 campaigns.
Test the remaining 5 banners in another set of +5 spots. Now you have $200 a day of adspent.

Then next day, you optimize by checking performance of each banner in all spots they are running. Cut the worst in all the campaigns they are running and keep the rest. Let more data accumulate in the winning banners.
Wait for one more day, and cut more.
In 2-4 days, you would have found one of two things - All your banner sucks, a few of them are good. If it is the latter, go print money and send me 10% profit share haha.

Lastly, banner testing is guesswork. You are making guesses with your creatives and if all your guesses are wrong, you can never win. Even if you are Alan Turing or Ramanujan, and can understand complex patterns and optimize data in your sleep, you will still not win. You have to guess right. And the difference between a novice and an expert marketer is the ability to guess faster and more accurately. A novice needs 100 banners to find a winner, while an expert only needs 25. But both have to test to find the winner.

And in Adult Media Buying you will need winning banners in a weekly basis to keep scale, because there is ridiculous banner burnout, audience burnout and competition with teams that spies everything.

This is one of the strategy that I use for testing, but I have other strategies too

---

It means you need to calculate how much are you getting from the total amount of proposed traffic and how much are you winning

---

You could group several small geos together into one campaign.

---

I've done this a few times, with countries that speak the same language... for example Austria and Switzerland. However, quite often, even countries this close have different preferences, some ads worked great in AT for example, but bombed in Switzerland. Some placements converted well with AT but not so much with CH ... I don't think this approach really helps much. At least based on my tests.

Did you have a different experience? Maybe you did something in a different way?

---

Yes sorry, it might be different. This is a thread in the adult forum, where probably the ad is the most important thing.
I'm rather speaking about my experiences with pop traffic (but there is also a lot of adult traffic included). Combining smaller geos gives me faster data about for example what placements (website IDs/zones/sources, whatever) to block. Thanks for asking for a clarification Matuloo!

---

This makes total sense, the worst placements usually suck across multiple GEOs, so targeting multiple countries to spot these suckers earlier is a valid option for sure.

---

I have noted that, if I restrict the cap then I have to bid little higher. What is your take on this?

---

Yup, with a limited cap your average CPM will be higher, because the "first' impressions are more expensive. That's caused by the fact that most affiliates run with limited CAPs... once they get served, there are some impressions left that nobody is bidding for and those will be sold for a cheaper bid.

So the more open the cap is, the lower the average CPM will be.

With networks that offer dynamic bidding, it's easy to manage, you simple set a bid and the algo will charge you based on what impressions you're getting served. In case of networks that do not offer dynamic bids, you need to run several campaigns in the same spot, with staggered bids, each of them with tight capping.

---

Got it, thanks for the tip

One more doubt to clear. For example if I create two campaigns in the same spot. Should I use same banner for both the spots?

---

Depends on the particular situation, in case you're just after more impressions at a lower price, you can use the same banner... so it will be the same as running with one banner on a network with dynamic bidding. But to maximize performance, you might want to use a different banner, to reach higher CTRs. Whenever I do this, I at least try to use different ads to see how they perform. But I had situations where I had 5 staggered bid campaigns in the same spot with exactly the same banner

---

I have always seen low LP on this spot for other geos as well.


I think so too. Will try with other offers this time.

---

Home > Paid Traffic Sources > Adult Traffic (NSFW)