We continue our series of useful materials, and today we turned to the Affiliate Research Laboratory with a question on how to optimize a campaign. At the entrance to the research center we were greeted by an eccentric man in a strange suit. “Oh, sorry, we thought you were going on a children’s excursion. So the singing dwarfs and ironic punishments can be canceled.”
Our guide changed into a regular lab coat and led us around the building, answering our questions as he went. “Most of the rooms on this floor are testing grounds for affiliate campaigns. Agree, before optimizing something, you need to make sure that it works at all. You cannot improve a table if there is no table yet. Why did I mention the table? We are scientists, we love hard facts, and also tables, lab coats and careless hairstyles, everyone knows that. So look behind this glass, don’t worry, it’s shockproof, do you see how the test subject is wasting his campaign budget on optimization without testing? A terrible sight. Of course he is doomed. His campaign will have to be scrapped soon.”
“Using absolutely scientific methods, and without testing on animals, we have established that before you start a campaign, you need to test it for at least a week. You should get conversions. The more of them, the more successful the testing. It sounds simple, but in reality this is where the devil lurks. What is your budget? How long are you willing to incur losses without getting results? But, as a scientist, and I am a scientist, look at my lab coat, I can authoritatively say that the absence of a result is also a result. This is the essence of testing, collecting information.”
In one of the testing rooms there was a bang and yellow-green smoke poured out. “Let’s go up to the second floor, directly to the optimization zone. We inherited this training ground with pits, ropes and an obstacle course from the previous owners, don’t pay attention to it. In fact, here we are doing budget calculations. I can show you a large board with formulas, to give your words more weight, but I will be grateful if you save us all time and take our word for it. So, the optimal optimization budget… Hahaha, that’s funny, optimal optimization. I need to tell that at the water cooler. Sorry, in general budgets are always different, and there are no ideal approaches. But if this answer does not satisfy you, then the optimization budget should, on average, be 10% of the expected profit. And remember that everything depends on the result of conversion testing.”
“And this brings us to a conversation about statistics, and for this, let’s go down to the basement, where our best statistics specialists sit in a windowless room, in inhumane cramped conditions. All statistical indicators of your campaign matter. Conversions are most important, but they also depend on a lot of other variables. Clicks and views without conversions can already tell you a lot. There are even crazy people using probability formulas, but we keep them separate so they don’t scare everyone else with their laughter during a thunderstorm.”
“Let’s go back to the hall, it’s hard to breathe here and it smells like a zoo. So, testing, optimization, statistics, all this gives us information. That’s the whole point. The more information we have, the more successful we will become from campaign to campaign . On the top floor we have specialists who have been through many campaigns, they just sit and smile. We even stopped turning on the lights for them. They have so much information that they can optimize any campaign. We all need to strive for this. Remember, negative results are sometimes more important than the positive ones. This is the building foundation for your future successful campaigns. And bring your children to the excursion, we have singing dwarfs and ironic punishments.”
