Home > Paid Traffic Sources > Adult Traffic (NSFW)

---

Best workflow for adult campaign optimisation? (6)

---

Hi all,

New here, started running on adult (Exo to start with) a few days ago. I made a newbie adult mistake and because I had no experience where was a good place to test placements, more or less went "let's go on ALL OF THE PLACEMENTS!" and then watched my money disappear faster than you can say "where's the pause campaign button!?!?"

heh

Anyway, a hundred quid later, I backed away from that method and just picked the top 30 or so sites that I thought might work. I was about -20% ROI today on a spend of $100. To be expected with no data behind me, and I would be in profit if I was to discount the two or three major click eaters that generated zero conversions (yes, xhamster is one of them)

So, I was wondering, what's a good way to start optimising, other than just blocking the obvious non-converting sites? How much data do you allow to build up before you go, "no, that's a bad one", "or yes, let's invest more in that"?

Once you've got your blacklist of 'bad' sites, do you apply that to all your campaigns from there on in, or would you start from fresh and regather new statistics? I know from my years in non-adult that different offers in different verticals will convert completely differently on a site, but adult offers are all going to be a similar theme - you're not going to advertise car insurance on there - so is a site bad-converting for adult offers full stop, or does it vary?

Also - mobile landers a good idea for adult? Research tells me most seem to direct link. That's what I have right now, all mobile traffic redirecting to rotated offers. Is it worth investing time into mobile landers? I know the obvious answer is "go test it", but was just wondering what the general consensus is before I start investing time trying to hunt down conversion rate increases that may not exist.

Any advice or suggestions greatly appreciated. Apologies if this has been covered before.

Aside from that, I can easily see that there's a lot of money to be made in adult, purely from the techniques and advice I've already seen here, so thanks! If I can be hovering around break even after a couple of days, and I've barely got a library of banners or landers yet (2 and 3 respectively), I'm pretty sure I can make this a nice little (big) earner. Once I've got the basics down and finished making up my landers and banners, I'll post up a follow along.

Cheers all

---

Hi mattp81,

Take my advice with a grain of salt. I've been learning mobile adult marketing for mere 30 days.

Everyone has a different approach. My current approach for new offers will be to run ads on top20 or so sites/placements (no RON) initially with low bids just to test how much traffic I can get and which sites are potentially good. If EPC is close to CPC, the same as CPC or higher than CPC I'll invest more in these sites. If there are no conversions while other sites convert well with a similar number of visits, I'll block non-performing sites (use statistical significance calculators to be sure). In general, you should spend at least the payout amount on each site to test if it's going to work. The longer you wait, the more certain you are which sites are good are which ones are bad. However, I think that sometimes it's better to make decisions with 80% confidence than wait a few days more and lose more money in the meantime.

You lost money probably because your bids were too high, not because you targeted all placements. You can't lose money fast if you set very low bids because you won't get a lot of impressions - which is good if you want to test an offer slowly without blowing your money too fast. I made the same mistake as you and lost $400 partly because of it. Now I know that setting high bids at the beginning is not a good idea unless you want to test really fast (and can afford it).

I think it varies because adult offers are still different. Some sites will convert better for porn subscriptions, some sites will convert better for adult dating offers or webcams. They also convert different for different countries, carriers, OS, etc. Moreover, you have to keep in mind that 300x250 ads and 300x100 ads also convert differently so if you test 300x250 ads on on site, you can't automatically assume that it's a bad site (you also have to test 300x100 and if you want to segment it even further, test different placements).

I actually asked a pretty similar question here: http://stmforum.com/forum/showthread...ing-vs-landers

From my limited experience mobile landers can greatly increase your conversions. However, you'll also lose a lot of time and money looking for the one that works at least as well as direct linking. So far I tested 7 landers for one offer (some were complete time wasters - no conversions) and now found one that will probably perform as well as direct linking. This means that once I tweak it a bit it may convert better than direct linking and then it will be worth the effort.

Moreover, landers and direct linking perform different on different sites. For one of my offers, on one site direct linking is three times better than a landing page while on another site there are no conversions with direct linking and there are conversions with a landing page (with less visits to a landing page).

You have to keep in mind that the fact that most affiliate marketers direct link doesn't mean that it's the best way to go. Many affiliate marketers are simply lazy and they don't optimize their campaigns as much as it's possible. If you want different results, do something different. Being innovative will sooner or later reward you.

Please do it! There aren't many mobile follow alongs here and it would be great to read another one.

---

thanks for your reply. I think the problem with my burning of cash initially was that I targeted so much that my money combusted before I had enough data to make a statistically sound judgement on 95% of them. I had hundreds of sites that had like <10 clicks, so can't possibly make a judgement on site viability, but add it all up and it amounted to quite a bit of £. Well, for a new-starter anyway!

Since I posted, I've had a couple of ideas on how to improve workflow and make it more efficient. What I like about adult is there's so much traffic you can get your data quickly, although it's costly. In my days in travel, you had to wait days to get this kind of data - whereas now I can get it in hours. This makes me excited!

CPM not for me right now. A brief experiment showed that eCPC was WAY above minimum CPC bids, so I switched to bidding just above min CPC. I don't know how people are being profitable on CPM. Maybe my banners are crap, I don't know right now. Something to look at later.

Things I've noted so far from my campaign is that mobile is initially converting twice as well as web. Will be interesting to see how this pans out as my stats grow. Will definitely do a follow along shortly. On a slight tangent, I didn't realise until I looked at my nginx logs that about 40% of traffic from Exoclick "web" traffic is actually mobile. I asked them about this, and got a pretty bizarre answer that didn't really make any sense, so I set up a catch-all mobile redirect for my web campaign and saw saw some conversions, which was good.

In terms of innovating, I'm a Ruby/Rails developer/Sysadmin by day, so once I've got my optimisation workflow set manually, I'm going to translate it into code and let a server do it for me instead. I think it should be simple enough. I'll release it if it works well

---

That's why you should only focus on top20 or so sites so that you won't have hundreds of sites to check. Avoid RON (run of network) traffic to prevent this problem.

Yeah, that's one of the things I love about this kind of affiliate marketing.

Exoclick underreports many clicks (in some cases they didn't report 90% of clicks my tracker reported) so eCPC isn't as high as they claim. You should get the number of clicks from your tracker and the number of impressions from Exoclick and then calculate CTR and eCPC (it takes time but it's the only way to calculate your CTR and eCPC accurately). I also started with CPC but now I prefer CPM because with good banners you can go well below minimum CPC bids (and get much more traffic).

Great, sounds interesting! Would you be able to create something that will pause and start campaigns in Exoclick for you? Exoclick doesn't offer time parting so the only way to display ads only at profitable hours is to pause/start them manually.

---

Everyone has a different approach, thats very key as what works for me may not work for you and others the same, but I find less attention to specifics and K.I.S.S allows me to keep the profits coming in without fizzing out my head being a slave to stats

My approach is

Offers, Landers, Ads. Put up campaign on a Top right placement, and a Vid A or B placement.

Day 1 if statistical significant run best offer
Day 2 if statistical significant run best lander for that offer
Day 3 statistical significant run best creatives for that lander/offer combo

Hey presto you've got the best match, if you're not in profits yet keep working on the creatives

---

Maynzie - thanks, I like simple!

mstm - re: coding up day parting, although there's no API for Exoclick (that I know of) the interface does seem to use simple GET parameters to manage campaigns, i.e. /campaigns.php?id=1234&action=pause or something like that, so would be relatively simple to code a basic day parting pause/unpause tool. I'll take a look once I'm fully up and running

---

Home > Paid Traffic Sources > Adult Traffic (NSFW)