Thread: Day 31: 3 Optimization Approaches

In the last lesson, we covered the TESTING process. Next step is OPTIMIZATION, which will be the focus of this and the next 2 lessons.

This lesson provides a high-level, general overview of optimization, while the next two lessons will go into the nitty-gritty.

Let's get right to it!



***********************************

EXPLANATION

So in the last lesson, you've done a lot of testing. You've:

-Tested some offers to find one that has decent conversion rate.

-Used that offer to split-test landers and identify a winner.

-Used that winner lander to split-test more offers to find a winner.

-You may even have cut some budget-zapping placements in the process.


If your campaign passed the "Campaign Worthy of Optimizing" test as described in the previous lesson, definitely follow this and the next two lessons to optimize it further - as it has a decent chance of hitting green.

If your campaign did NOT pass that test, it MAY still stand a chance at reaching profits with further optimization, but keep it paused for now, go through all the lessons on optimization first, THEN decide whether you want to run the campaign further.



*****

Keeping Our Eye on THE GOAL

First of all, let's revisit our ultimate goal:

The goal is for the optimized campaign to make enough daily profits to justify the effort required to keep it running.

Minimum Daily Profit:

You want each campaign to make at the VERY least, $5/day in profits, to even justify keeping it running.

At a minimum, you need to check stats daily to make sure the campaign is still profitable, and you'd probably need to cut a placement here and there. If the campaign is making peanuts, you'd be much better off spending your time on testing more offers/landers for the same or a new campaign, or scaling another profitable campaign.

When you're more experienced and have bigger campaigns running, you may not even want to babysit the $5/day campaigns. But for a newbie, $5 campaigns will give you the motivation you need to keep hustling.


Minimum ROI:

ROI would be a less important metric which I'm not going to discuss too much.

However, I would suggest keeping your ROI above 10% at a minimal.

Pop campaigns tend to have volatile performance. It can be profitable one day but not the next. And a higher ROI can provide a cushion to keep the campaign from constantly dipping back into red.

Another consideration is the fact that some of the tools you're using - such as Voluum which charges by event, or your tracker server which needs to be upgraded with the level of traffic - will cost differently depending on the amount of traffic. So the more traffic you run, the higher the cost, and by working that into your ROI you can ensure those costs are covered.

Also, you have a finite amount of time and budget, so if the ROI is too low, then you'd be better off investing that money into a campaign that makes better ROI.

However, if budget isn't a bottleneck for you currently, such that you're not having to decide which of several campaigns to fund, then I would suggest to base campaign decisions more on potential daily profits and less on ROI.



*****

The 3 Optimization Approaches

There are 3 general approaches you can take when optimizing a campaign:


1)Blacklist Unprofitable Traffic

If it looks like a large-enough portion of your traffic is profitable enough to give you acceptable daily profits at acceptable ROI , you can keep cutting unprofitable traffic (placements, OSs, browsers, whatever else you're able to exclude from your targeting) until you're profitable.

This is known as the "blacklisting" approach.


2)Whitelist Profitable Traffic

Same situation but different approach: If it looks like a large-enough portion of your traffic is profitable enough to give you acceptable daily profits at acceptable ROI, you can target the profitable traffic (placements, OSs, browsers, whatever else you're able to include in your targeting) to lock-into profits.

This is known as the "whitelisting" approach.


3)Test Bids

If it does NOT look like enough of your current traffic is profitable-enough, then you should not rely on blacklisting or whitelisting to get to green.

In this case, whitelisting won't be feasible, because there aren't enough green segments to target.

And blacklisting won't be feasible either, because as we've agreed before, cutting unprofitable traffic when there's not enough profitable traffic in the first place, will NOT leave you enough profitable traffic! (Digging through a pile of shit that has no pony in it will not give you a pony.)

In this case, you'd need to introduce new traffic you're not currently targeting - for example by increasing your bid to trigger traffic from higher-quality placements - in order for campaign performance to improve.



*****

Important Notes

1)Unfortunately, I won't be able to give you a step-by-step on exactly how to optimize a campaign. There are so many factors that can combine to give an infinite number of situations, and I can't think of a way to come up with a one-size fits all approach.

You'll just have to do your best while running your first campaigns. As you run more campaigns and gain more experience, you'll get better at deciding whether a campaign is worth optimizing or not, and then optimize the worthy ones in an efficient manner.

However, to prepare you for battle, I will do my best to provide a general, basic approach in the next lessons.


2)Always keep in mind to focus on testing offers and landers first and foremost! ESPECIALLY offers!

I know I'm beginning to sound like a broken record, but improving your funnel by testing offers and landers (particularly offers) will make MORE of the total traffic profitable, so that with further optimization, you'll end up with more profits.

Whereas if you don't test enough offers and landers in the beginning, you will put a ceiling on the amount of profitable traffic, such that more cutting (and more spending) would be necessary to reach green, and the resulting profits would be limited.

Plus: Testing offers and landers is often less expensive than cutting placements etc. I've made that very clear in the lesson on how to cut placements.

So, when viewing campaign stats, always be asking yourself the question: "Is my offer+lander good enough for me to focus on optimization next, or should I test more offers/landers first?"

In other words: "Is my offer+lander likely to reach at least the minimum daily profits + ROI when I optimize it further?"

This question, though, may not be easy to answer. (We'll go into how to estimate minimum daily profit in the next lesson, which will help.)

When you're a new affiliate, as long as you've tested at least 3-5 landers with at least 1 offer as described in the last lesson, to arrive at a winning lander+offer, then go ahead and focus on optimization - because learning is the most important task at hand.

If, after optimizing a campaign for a while, you find that it "isn't going anywhere", i.e. either you hit green but the resulting profits are small, or you're not able to reach green at all, then you'd know for sure that your offer+lander aren't good enough - which means you'd need to test more offers/landers before optimizing again, or test bids, or both.

Worth mentioning: If you need to test more offers/landers, don't start from scratch! Split-test new offers and/or landers against your currently-best offer/lander! And apply the same placements blacklist so you'd be using the best traffic to test the new offers/landers! Chances are you'll save test budget this way.

And this is why I recommended sticking with a couple of geos in the beginning and running campaigns in them again and again - so you wouldn't need to start from scratch.



***********************************

The next two lessons will cover the first two approaches - blacklisting and whitelisting. The lesson after that will cover bid testing. You can probably read all 3 lessons in 1-2 days, but the actual implementation may take you a few days or more.



Amy