Yuma 4×4

Media and Communications

Scaling Straight Sales for the Long-run with Native Ads | AWeurope 2018

Scaling Straight Sales for the Long-run with Native Ads | AWeurope 2018

Hello! How you guys doing? My name
is Kelly and today I’m going to talk about my favourite affiliate-friendly
traffic source, native ads. And not just how to run any offer,
but how to run high payout, high cost consumer ,
straight sale products, specifically. As opposed to your average free plus shipping trial offers, your $4.95 trial offers. Now, the bad news with native, is we all know, it’s very difficult, but when you add in straight sales, and high payout products, it adds another level of difficulty. That’s actually okay, because the good
news is that with the right approach, things become much more manageable
and profitable sooner. As this is a crash course, I can’t really cover everything. But I’m gonna go over some of the key takeaways and tools you can use to build a solid
foundation for success with native. Whether you’re about selling skin creams,
or diet pills or really any other higher priced item. So, why should you listen to what I have to say? Well, I’ve been an affiliate for about
six years now, as a solo affiliate and kind of coasted from campaign to campaign for many years. Making decent money here and there. I suspect that many affiliates
are probably similar to me. Eventually, things come to a
standstill, whether it’s for a day, or a week or a month or longer. For me, about two years ago was a good year and a half of drought, and a couple of things going wrong and really gave me some time to re-evaluate what I was doing and how I was approaching marketing
and affiliate marketing in particular. So, I really went back to basics, and
tried to focus on the fundamental aspects
of marketing I’ve been neglecting and to stop chasing all these hot new tools,
and one simple trick type methods. Fundamentals are not a very sexy thing to
talk about. We don’t want to hear these amazing
tricks and secrets. But really, with native, I’m seeing so many people.
They just are copying the same landers and ads, expecting things to work, just by switching around bids or doing these little tricks here and there. I feel like that’s really more
like a hope and pray method. It’s not a long-term strategy.
So with native, I found that, just taking a little more effort to put
more work on the ads and creatives and really working with making better
decisions with your data. I feel like that’s one of the number one issues with native. It’s that there’s so much data to sift through. If you’re trying to test too many offers,
too many landers, sort through all these widgets and
placements. You can really make a mess of things. You can lose a lot of money very
quickly on native. So, why native? Well, it’s very affiliate
friendly, at least for straight sales. You typically can’t run several offers
on Outbrain or Taboola. They’re not compliant there. But you still have Revcontent, Content.ad, MGID and there’s huge scale to native.
It’s not quite Facebook level, or Google level, but it’s definitely up there. You can
definitely reach 4-figures a day pretty quickly, and 5-figures is definitely attainable.
Your ads last for a long time, even though I still like to continually upload new ads. Campaigns lasts a long time. The great thing about straight sales and native is they have really long-lasting campaigns. The current one I’ve been running,
has been going for 9, 10 months now. Not the longest campaign I ever had,
but a long time nonetheless. Account reps on native really make your life easier. That’s one of the things I love most
about native, is how approachable reps are, how much information they can give
to you, and really pretty much make or break your whole career on native.
One of the benefits, actually, is that there’s a high barrier entry, so there’s
less competition. Now, this is surprising with straight sales.
Coming from running trials for most of my affiliate career, or part of it, I would assume,
that going from a cost consumer of $4.95 or $5 jumping up to $50, $150, that
people weren’t just going to buy these straight sales. I think, maybe just
the international market is where it’s at, and it’s like a second best option.
But in reality, people will not only buy one unit, they’ll buy multiple units.
I was really surprised to see that majority of the people actually buy in the top tier,
the maximum units they could, with an order. Just like any other vertical, there are pros and cons. Just like any other offer type, there’s definitely things that work to its favour and to its disadvantage. The biggest thing for me was that I don’t have to worry about retention rates and hitting certain rebill thresholds with trials. I hated the idea that I was gonna
get a message from my rep saying, I was having money withheld or I wasn’t gonna get paid when I thought I was gonna get paid. And occasionally, straight sales
can actually outperform trials. This really surprised me as well. When
the advertiser knows what they’re doing, and they put in the effort and time, the EPC’s can be very good. There’s less competition. People assume that there’s more work to straight sales, and there is. The industry seems to be going that
direction, at least this year. Who knows where they’ll be a year or two from now. But I was told, it used to be, straight sales were the primary model, many years ago. And because they’re a little bit harder to work with, you have to approach things slightly differently. You have to put a little more work. It kind of forces you to be a better marketer. One downside is that requires a higher
budget to test. Whenever you jump up the payout, you’re gonna have to spend
that much more per widget or placement, to get results, to see what’s working. And
overall, they’re still less performance than trials. Advertisers are really struggling,
from what I’ve been told, to adapt this model or switch back to it.
To customer service scripts, they have to talk about. So are there any tips or shortcuts
you can use to identify if a straight sale is going to be successful, or have a good chance of being great before you even run it? There’s really two things that I found. The main thing is actually, payout.
As affiliates or advertisers often told, don’t worry about the payout, it’s
misleading. But in this case, if you look at the payout relative to the cost
consumer, it’s a really great indicator of how much work the advertisers put
into optimising their front end and back end. So, the higher their payout relative
to what they’re charging the consumer, the better. The really poor offers are
going to be hovering around that tier one price. You’re going to be $60, $65.
And I can just tell you right off the bat that those offers will probably
not work in a competitive geos. Good offers are pretty close to
your tier two price, maybe $80, $85. And the really great offers
are right around that that price range. So, that tip alone can really help you
cut through a lot of the offers out there. When it comes to ads on native, the amateur type images are definitely preferable to overly professional. I really like to use cellphone photography. Pick up a tripod on Amazon and start taking your own images. One of the benefits of doing this is that you
own a copyright of that image. You can go to the ad network and if you
see someone using your ad on spy tools, you can basically have those ads taken down, without getting too complicated. Normally, just send them an email and
tell them that someone’s running your ads and that’s a big benefit. You can basically lock out or lock in your your ads from people running them. So ideally, you only want to use content you own the copyrights to. It really helps. Try adding call-to-actions to your ad text. Things like, “do this,” “try this,” “watch this.” They get a little more extra clicks. Help out
your overall profitability. When it comes to landing pages, you really want to focus on the content over design. It’s not pops, it’s not flashy. You want to make it easy for people to read the pages, as simple as that sounds. You can also try removing filler content. There’s a lot of assumptions with neutral
pages, in particular. All these videos and testimonials. You can use sites like
Wistia.com and upload your videos and test. And I found, that people weren’t
actually watching those videos that everyone was running and removing them, had no negative effect on performance with that campaign. You want to keep certain elements above the fold, particularly on mobile, the attention span of a
visitor on native is pretty short. So things like your headline, sub-headline,
image, you want to keep that above the top half on your screen or your phone,
so you don’t have to scroll down to get the full intro. Essential scripts. You really want to make sure you always have a back button. For mobile, that ads a good 2-3% extra revenue, just using the same landing page or a slightly different one. You also want to add a exit modal intent. When someone moves their mouse to the top of the screen on desktop, an exit modal will appear.
You can offer them another link to the offer. That’s another few percentage points. You really want to work on becoming
a better storyteller for native, that’s pretty much the key thing I’ve been focusing on. Some resources I found very helpful are Michael Hayes, I believe. Of course, Covert Cash Advertising is a great resource as well. But it’s all about telling the story and capturing someones attention and less about celebrities, and fake comments, I found. At some point you, come to realise, that just copy pasting landers off the spy tools is not going to be even productive use
of your time. You want to start iterating faster and be able to make variations quicker. You might want to look to a content management system, as basic as that may be. There’s WordPress, which I’m sure everyone is familiar with. It’s
very easy to get started with, it’s pretty much the gold standard. There’s so many great plugins. One people may not have heard of, it’s called Netlifly, it’s a hosting platform, plus Hugo.
It’s a very nerdy option. There’s a lot of technical configuration to learn. But it’s completely free, there’s no servers to worry about and everything is hosted on CDNs. When you’re getting ready to launch, this
is the biggest issue I think people struggle with on native, is having a control of data set to work with, versus just throwing your money out there, and hoping that something sticks. And trying to sort through all these different widgets and placements. So, it’s better to really constrain the amount of data you’re working with in the beginning. Here’s some of the best options
I’ve found for native. For starting your own, whether it’s a whitelist or blacklist.
The very best, the very top is, if you have a good relationship with
your account manager, they’ll often give you whitelists of someone who might
have run a campaign months ago and stop spending. And those whitelists
can often be immediately profitable. You have to work on developing a good relationship with your account manager for that to happen. But if you’re not there, your data is not there, the next best option is using data from your own campaigns. Good widgets tend to be work for a wide variety of verticals in campaigns. Often times, the traffic source provides you whitelists. Those are a good place. You just want to limit the amount of widgets you’re testing to maybe less than 100, depending on how much money you have to spend. One of my favorites are
premium placements or brand sites. These are very competitive,
highly priced placements. You can often tell very quickly whether a new campaign or offer has even a slightest chance of working. And this is my favourite method for testing whether a campaign has the room, the scale, the 4 and 5-figures is working with brands. If you can get the breakeven on brands, I found, you automatically have a 4-figure day campaign. if you’re working with a smaller geo, you’re probably not gonna be running whitelist or blacklist, you’re gonna be running network. And you also have the option of going with one site or one large placement. I don’t recommend this because you could get a bad performing widget or placement. You really want to test a wide variety or at least a smaller subset of widgets in the beginning. As far as devices, I found both mobile and desktop perform well. Tablet is underperforming, it’s still profitable. So if you want to start a campaign you, want to do one for each device type. And if you want, you could omit tablet, to start with. I like to bid high, with a minimum budget set. Just to secure the ads, to get a good amount of impressions and clicks. So the algorithm deems it is worthy of being served. Now just be careful with overspend. Don’t try to launch or run a network campaign on $100 a day and expect it’s only gonna stop at $100. It’s gonna overspend. I also like to launch in the afternoon of the geo. And then, once it’s reached that initial cap of budget, we’ll just keep lowering it by 10%, 20% over hours, or maybe the next day, in accordance with performance.
You want to look for how many landing page clicks you’re getting, what your CTR is like. And just kind of get a sense of
what your data is looking like. If you already have an optimised campaign, and you want to test new ads, instead of raising the bids on your existing
optimised campaign and messing everything up, I like to test ads in a new campaign at higher bids. And then once I see that they’re getting traction and working, I’ll pause them or duplicate them over to my optimised campaign. Data is a a unique challenge on native,
with all the moving parts and all the widgets. And when you’re trying to test a lot,
you have to be cautious of jumping to conclusions too soon.
This is one example I pulled. You look at this, and I think most people would say, yes that’s a successful test. The bottom landers are performing. And you would expect that, if you were to cut out the worst-performing one, that your ROI might go up by a considerable amount. That’s 40% difference. But I’ve found, with straight sales, because the conversion rates are much lower, there’s just a whole lot of volatility in a day to day. And you really want to run tests for extended periods of time and not get trapped in thinking or going down a route of thinking just because an A/B test shows a winner, that you actually have one with
very few date data points. 52 conversions is not a whole lot of data to work with. So, I ideally, like to keep a campaign running, test landers, until there’s at least 20 or 30 difference.
20 or 30 conversions difference between a given lander. That typically means, I’m running several hundred conversions to different landers. One thing I also like to do, just to help make things easier with tests is I’ll break out each successive tests into versions in my tracking platform. So version one is the initial one. And if I take the winner, I will name that
the control in the subsequent tests. So, it makes it very easy to go back through
your your history of your campaign and see how performance has changed over time. Instead of dropping your landers, and adjusting the weights on the fly. So, once you’ve found something that works, there’s really only three methods
basically to scaling. You can go with a whitelist, can go with a blacklist only campaign, or run a network, and we can go with a
whitelist campaign that’s being fed new converting placements by a blacklist. You can also of course scale international geos. And what I’ve seen, is that, a lot of affiliates have success by making many separate whitelist campaigns. Just for many different verticals, different geos. And that’s one way to get a lot of spend per day. Account managers are your best friend,
non-native. If you’re not friendly with an account manager, you don’t have one, get
one ASAP. They will make your life so much easier. The amount of information they provide is, I would say it’s better, spy tools are practically irrelevant when you have a good account manager. I like to ask them how much someone is spending per day, and more importantly not just what
they’re running, but how they’re running it, is probably the most valuable thing on native.
I ask them how their campaigns are set-up. Are they running blacklist,
whitelist? What’s their campaign set up? I try as much as I can to duplicate not only what they’re running, but how they’re running it. You also get first access to features and placements, API’s. That can be hugely profitable, if you’re the first one on a given placement or a new feature. Great help with native is, they can really help you troubleshoot approval issues. It’s always a common complaint I hear with people with affiliates. It’s that it’s so hard to get things approved on native. A lot of times, they spell it out very clearly in their terms on their pages. But, if there’s any issues you can usually text your account manager and they’ll get it sorted out. It’s kind of a common conception, that
there’s a ton of bots on native. And I thought this was true as well, when I
first started with native, two years ago. The traffic quality was just absolutely
terrible on someone that sources web content. But what I found, was that once I started using an API to actually look at the amount that I was actually spending on a widget, I was not actually paying for the bots. It was all these third-party spy tools, or VPNs that were were blowing up my tracking server, and sending me all these these clicks and visits. So the important thing is that you don’t want to make any bad decisions with bad data. You really want to try to eliminate as much as you can. Views coming into your campaigns and
then clicks on your landers. And one way you can do this, is with a service called CloudFlare. They will come to tech bots and block them at the website level
before they even reach your campaign. I set it to, essentially, off security. You can also try low and see how your ad spend is affected, if there’s any difference. On the lander, I would just use JavaScript base links, or you can kind of hide a link and then trigger the actual one with JavaScript. And those things combined with an API, will really cut out pretty much majority of bots you’re seeing
in your tracker reports. You can, of course, also filter out IP ranges. And I like to redirect wrong geos. So basically, same campaign with just a
different lander on my tracker. So all the data I have, is actually from
that particular country. I want to make sure that I’m making the
right decisions, based on good data. With native, I find, as much as optimisation,
and as much as automation helps, it’s really not necessary if you have a good lander, and
offer, and got good widget placements. But it definitely helps.
So one of the rules I’ve come to really like is instead of testing a widget at
two to three times the payout of the offer, I’ll look at the clicks I’m
getting from my lander to the offer. So I look at the effective cost per click to
the offer. Just like you’d have a guaranteed EPC for some offers when
you’re testing, you also want to look at your effective cost per click to that
offer, so you can really rule out a lot of bad. Which is this why, early on, instead of spending $200, you could spend $20. And if that widget has no clicks or very few clicks, you can cut it right away and be confident that it’s probably not going to work. Offer is number one priority. Locking that down first, before your landers, is essential. You don’t want to be testing too many things at once, especially with straight sales and native. I like to limit my test with landers, beginning to two at a time, but I’m really just focusing on the offer. With automation, you can block out placements, like these high, effective cost per click widgets. And, I also like to remove very low CTR widgets that might affect my overall campaign algorithm performance. It’s also gonna really depend on
the platform you’re working with. Whether they allow bids per domain, or bids per site. I also like to move my widgets from my blacklist, as I get the diversions onto my whitelist. One of the best things about running
Native, and straight sells is, as I said, that campaigns last a long time. And this
allows you to really take the risk of trying new things. I’d really encourage you to focus on campaigns
where you can learn new skills. That’s kind of how I approach things, nowadays. I don’t like to run campaigns if I can’t apply what I’ve learned or I can’t learn something new. I have to keep moving forward. That’s it. Hope you enjoyed the presentation.
And, thank you.

Leave comment

Your email address will not be published. Required fields are marked with *.