If you prefer to listen, podcast version of this article, click here.
Do not break what is already working
How-to articles and guru advice are an issue today. This issue extends to all forms of advice, including the answers we get back from AI.
In ecommerce marketing, a lot of this kind of "How-to" advice exists. It might sound something like this:
- “Your ads should be X type of campaigns”
- “You should use lots of design and images in email campaigns”
- “You should use broad targeting in your ads and let Meta figure it out…”
The problem with this kind of advice is that, when people hear it, depending on how much they trust the source, they treat it as true. That is dangerous.
Marketing, especially paid marketing, is simply a collection of ideas that we may or may not choose to implement. These ideas have no data attached. We simply do not know which idea will bear the most fruit.
Therefore, at EssTwo, we see all ideas as experiments. A marketing experiment is an idea with unknown results; we must test it, watch the outcome, and then either kill it or expand it. We call this cycle the Winner Cycle at EssTwo.
Any advice you get from your paid marketing, be it from a guru, article or ChatGPT, is just another experimentation idea.
In my experience, however, people treat "how-to" guidance as an experiment that has already succeeded. And thus, they DOUBLE DOWN and put a lot of money behind an unproven experiment.
If I were to summerize that, I would say this: How-to advice, from any trusted source, leads people to spend a lot of money on unproven marketing ideas. Instead, they should systematically integrate these ideas into an experimentation framework.
Treat all How-to advice as an experiment
In the previous section, I argued that how-to guides lead people to spend money on ideas as if the idea had already proven itself. I went on to discuss that one should treat any idea as an experiment.
But what does it mean to treat an idea as an experiment?
If your store is not new, then it means you already have data in your business. Maybe you have run paid traffic in the past, maybe you did SEO, maybe you have lots of customer reviews. No matter the source, these are all data points. And you can use any data point as a marketing idea in paid traffic.
Every time you collect this data and form paid traffic ideas, you will accumulate a list of what I call "credible marketing ideas".
What makes them credible?
Good question. These ideas are credible because they are born from your business data. They are backed by data, or, said differently, they are backed by customer behaviour.
When you get an idea from a random guru (even from me!), this experiment is called a "Non-credible marketing idea" because it isn't born from your data. Now, a non-credible idea can still work. But we are in the business of probability. An idea generated from your own data is more likely to work. Especially if it comes from paid traffic data.
So, the question remains... When you get a paid traffic idea from any source that is not your own business, add it to the bottom of your "experiments to try" list.
How to determine if how-to advice is a good experiment
By now, we have established that all marketing tips and guidance, from any source, is a potential experiment. And, as a non-credible idea, it should be placed at the bottom of your list.
So, what should be at the top of your experimentation list when it comes to paid traffic?
This is a very important question. We work on it every day with our clients at EssTwo.
When I run paid traffic experiments, I will always test ideas born from previous paid traffic data. Second, I create new paid traffic campaigns based on highly responsive email campaigns.
There is only one exception to the "external ideas are at the bottom of the list" rule. And that is when an idea is newsworthy.
For example, I work with a hair care brand. They use 100% natural ingredients. Some of their competitors do not. There was a time when a specific ingredient was banned in the Western world because it causes permanent skin damage (because when you treat your hair, it's unavoidable to not get some on your face and skin)
Anyway, we ran paid traffic campaigns discussing this issue and how "our products were natural, and didn't include this". Very relevant, very timely.
I will leave you with a question: what ideas keep recurring in your business that you can turn into paid traffic campaigns?