You’re feeling a tad bit unwell. For precautionary sake, you paid a visit to the local clinic. You’re ushered into the doctor’s room. She doesn’t speak a word, never ask any question to diagnose your condition. Instead she start jamming injections; one after another. After which, she grab a random fistful of pills and shove it down your throat.
You’re more likely to be killed by the “cure” than the disease. Sounds ridiculous, right? Yet that’s the approach very commonly used to campaigns. “Oh, it’s been more than 6 hours since I had deployed my campaign and there are no leads. I will just randomly tweak the audience, change the ad creative, etc”. The right approach to campaign optimization is very similar to what a good doctor will do; understand and diagnose the problem, form hypothesis, conduct experiments to identify the cure.
Prevention is better than cure
There’s a good reason why professional campaign managers spend so much time conducting pre-campaign planning and research. Thankfully, if you are using Peasy – almost everything except for creative messaging planning is done automatically for you. So if you had skipped through the lessons in the “Lead Generation Starter” segment, please revisit. If your campaign doesn’t spend, not generating any results right from day one; anytime if you encounter the “At Planning” optimization causes in the diagnosis chart (download below), please replan your entire campaign.
Are your suffering from Campaign-ADHD, use the methodology
I have seen a lot of people sitting in front of their laptop, making frequent changes every other minute to their campaign. Let me try and put this into perspective; suppose you’re teaching your beautiful toddler the alphabets. You have laid out few upper case alphabets, and had asked her to find the letter “A”. While she is still ruminating, you threw in lower case characters. Before she can decide, you had dumped in the same alphabets but different color – and now you’d asked her to locate upper and lower case “A” in red.
Your ad are served by an algorithm and AI, whose sole objective is to learn “which audience will find this ad to be most relevant”. By constantly changing your campaign, you are disrupting the AI’s ability to learn. When the AI can’t learn, your ad won’t be served.
The solution to this problem is to take an iterative approach. Define a set period of time for 1 iteration. In each iteration, you have:
- Objective: You will focus on diagnosing and fixing only one leakage at a time
- Sufficient time/data for analysis. If it takes 100 leads to get 1 sales, how can you possibly determine a campaign’s lead quality after 10 leads?
- Form a hypothesis. Use the diagnosis chart (download below) to identify symptoms from the campaign’s metric. Based on the process of elimination, identify the most likely cause.
- Design your A/B Testing. Based on the selected causes, design test which can conclusively prove your hypothesis.