3 Things You Must Know Before Scaling A Campaign


Often times scaling a campaign isn’t as easy as it’s made out to be. The first route is usually adding more traffic sources into the mix, or by duplicating your efforts over more campaigns, with different targeting, different angles, or by running your campaign over more than one traffic account on the same traffic-source.

What are the common roadblocks here?

Well, every source that say, may have the same format of ads (for example, native), won’t always perform as well as another source for a multitude of reasons, such as the available demographics, placements or overall traffic quality (some sources are really bad at vetting their publishers and are subject to a lot of fraud). It may just be that the source in question requires more testing of variations to your winning ads and landing pages in order to ‘crack’ that source. An example is landing page sizes on pop sources. Try tweaking the viewable size of the page and see how each variation performs on said traffic. It’s wise to find out what the pop dimensions are too, as many sources vary. If the source is break-even to slight loss, I’ll continue testing tweaks, so long as I don’t go into the red too far, until I’m in the land of the green, where the tree’s sway, the daffodils flourish, and I can frolic in the meadows of profitability.

Add More Countries?

Scaling into other geo’s (geographical regions) also tends to come with a new set of issues. A lot of people purporting scaling to other geo’s should be as easy as 1-2-3 are pulling your leg, I’m afraid. Some of the common problems I’ve faced are typically incorrect or poor translations (even from expensive translators on onehourtranslation.com), traffic sources that perform better than other’s with certain geo’s  (for example, one source worked great for me in the US, but sucked royally in France and Japan).

Why would this be? Different publishers, different placements & different audience come to mind. To note, it’s also potentially easier for these publishers to utilise bot traffic to increase their RPMs.

I have a solution for traffic sources that would help filter out a lot of the bullshit – pretty surprised they don’t do it already (greedy bastards) – block colocation & hosting companies and only allow valid ISPs through. That’ll minimize the problem a bucket-load.

Death by CAP

Looking to scale infinitely? Yeah, calm down soldier. Caps are a ceiling, imposed by the advertiser on the network, of how many leads they are able to handle on a daily/monthly basis, and they can stifle your ability to scale a lot. I absolutely LOATHE caps. I always seem to get surprised by them whenever I start scaling up, depositing a lot of money into traffic sources etc… Complete nightmare. I’ll go from 50 to 100 sales a day, and no one bats an eye. I go from 100 to 250 sales a day and all of the sudden the alarm bells start ringing and a soft-cap which I never knew about is imposed on me, at 150 leads a day (for example), which, if you blow over, can cause the advertiser to hard-cap the network. The difference between a hard and soft cap, from my understanding, is in the case of a soft-cap, any minimal spill-over, will still get paid out, but in the case of a hard-cap, any accidental spill-over will be on the network or affiliate and they’ll eat a loss. Obviously it’s in the best interest of the network to avoid a hard-cap, as it ultimately means less revenue for the network. The advertiser will either stop firing conversion pixels after xx leads a day, or they will only pay what they set the maximum cap at, if the specified cap is reached and exceeded.

My only tips here is to have backups in place, if you can find any, that still convert (normally not as well) and can keep your campaigns above water. Else, just pause traffic and resume when the clock strikes 12 (check with your network which time-zone they abide by) or when the month restarts.

Managing Workload

As a one man show, still, I only ever outsource bits and pieces when I need the extra grunt work, but don’t (at this current point in time) have any full-time employees. This creates a drastic issue, and that is, scaling over to 5 different sources, means 5 x the workload for optimizing, which can be a daily chore and take FOREVER. I had a situation last year, where I ran a campaign over 13-14 countries on 4-5 sources and had to optimize every 1-2 days to keep it profitable. Looking back, it would have been really convenient to have a full-time employee optimizing for me all day so I could focus on other, more important things, like finding more traffic sources.

Let’s have a conversation…

What are the main issues and roadblocks you run into when trying to scale your campaigns? Leave a comment below and if it’s something I’ve dealt with and found a way to handle it, I’ll try my best to provide some tips.

– Andrew, aka Andy D

If you’re looking for further Affiliate Marketing Guidance, check out StackThatMoney. Best community of experienced marketers from around the globe, exclusive meetups, follow along’s, tutorials and the knowledge of a thousand sun-gods.

Can’t Decide on Tracking Software?

I’ve recently switched over to a new tracking platform called Thrive by the guys over at iPyxel which I love. It’s still in development, but is constantly improving and making strides, and the best part about it is it can be self-hosted. The offer a 30-day trial and it’s $99 a month thereafter which is well worth the investment.

Views: 21159


Young, driven, health-conscious affiliate marketer providing you with tips to improve your health and fatten your wallet.

6 thoughts on “3 Things You Must Know Before Scaling A Campaign

  1. Great Post Andrew! I’m running on pop traffic right now and the thing I’m struggling the most is finding a winning lander. One day I could find my winning lander doing well and then the second day it just tanks. There’s a lot of factors I’ve been thinking such as:

    – Blacklisting Placements x2 spend(This is after I found the winning lander)
    – Find Translator to translate the page to local slang(OneHour could be good, but probably have to hire 1-3 translator to split test results in how they translate in local slang).
    – Trying to go direct with the offer. I’m working with the affiliate network right now. And I’d say I’m generating a consistent 30-50 leads a day. I’d like to establish a good relationship with the advertiser therefore getting better payout bumps and more rapport in what the advertiser wants out of my leads. I can’t get a hold of the advertiser even after trying to find their contact info on their Terms and Condition page.

    My plan right now is to do this:

    1) Running campaign(w/ winning lander and offer) using day-parting(14 hrs) this week. The reason I’m using day-parting is because I have a budget I have to work with and that I want to run a campaign specifically at this block of time to measure results in a consistent manner.
    2) Blacklisting placements
    3) For every two days that I run the campaign, I’ll implement new changes, ie new landers to test, changing up the translation with OHT. Discover what targeting options are doing very well and then tailor my lander to call out those options.

    1. Hi Simon,

      Thanks for the great comment!

      Definitely sounds like you have a good optimization strategy there. Regarding pages, a few other things you should always keep in mind are page-weight (the smaller, and faster a page loads the better), testing out pages that are responsive, versus pages that aren’t, testing whether or not having everything aligned to the far left works better than centering everything, and testing out numerous other ‘scripts’ and techniques to get your page ctr and cvr up.

      The biggest issue, however, in a lot of cases, is the offer just doesn’t work for pop. It might work well, or well enough on say FB traffic, and your rep will tell you it’s smashing, but it just doesn’t seem to convert well enough to work on pop.

      Just ensure, whatever you do, you track your changes. Try multivariate testing, at some stage, instead of simple A/B testing and you should yield some interesting results.

      1. I noticed that about the offer too. I’m promoting a sweeps SOI offer and it’s considered one of the highest earners on the network. It may show a good epc and good volume rate, but at the end of the day it’s determining which traffic type is generating the greatest volume on that offer.

        I read something about the multivariate testing on Finch’s premium post, I’ll have to revisit it. Thanks!

  2. I hate caps…there I said it. I feel better about myself now.

    Question, when hitting caps, what’s your strategy in reducing bid prices to increase margin? If I can only sell 100 widgets a day, and am running at call it a slim ROI @ 20%, how do you start increasing margins? …or conversely, do you just start looking for back-fill offers?


    1. Hey Jason!

      Interesting question there. I think it’s very subjective to the type of traffic you’re working with, and how it affects the quality and delivery. Lowering bids isn’t always going to work to reduce costs, as it may also result in a sharp decline in leads.

      I think split testing this is a great idea. You could also set up multiple ads at different bid levels to see how traffic quality is affected, even though you’re technically competing against yourself.

      Definitely great if you have a backup offer in place, but if not, trying to maxmise your ROI makes complete sense.

Leave a Reply

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