Information on digital marketing and online advertising has become full of noise. Learn how to build masterful campaigns just by slightly changing your approach.
If you want your advertising to work, don’t burn the witches. In 1589, Anne of Denmark sailed to Scotland to meet her new husband, King James VI. During the crossing, a violent storm forced her boat to cruise away, and delayed her arrival by a few weeks.
King James was convinced that witches had summoned the storm to sink his wife’s boat. So began a series of famous witch trials called the North Berwick trials.
People used many advanced criteria to find potential witches. Including having red hair, a strange birthmark, or being left-handed. This resulted in over 70 people tortured and burned at the stake. Now, you see, “intuition” or “common belief” is not always the best bet.
The Problem With Digital Marketing
Why am I telling you this? Because in today’s digital marketing world, “intuition” and “common belief” is the norm. Shady best practices and so-called experts have led to confusion and poor advertising.
“Make that button green! No, make it blue, green is so overrated! Wait, we should put some hot women in our new car commercial, I’m sure it will help the sales. Nah, we must put a baby face, that’s the real secret.”
That’s pretty much how it works today. People making assumptions, and then wondering why the results are not there. May it be in digital marketing in general or in online advertising.
People come up with an idea that sounds good. Next, they write a 200 pages business plan to prove it can be profitable. Then they launch the product only to find out nobody wants it. Because they were too smart and their idea was too good to look at the data.
Without data, you can’t isolate the cause and effect relationships of things, and so you can’t understand why some things work, while others don’t. And when that happens, people see witches. They start thinking marketing is all about luck, that it is some sort of obscure dark magic. Though it isn’t.
“A blind pig can sometimes find truffles,” said David Ogilvy, “but it helps to know that they are found in oak forests.”
The Cure to Digital Marketing’s Aches and Pains
Surprisingly, no one seems to use them. Few online advertising theories are backed with solid evidence. And most market research is actually more something like market guessing. So advertising effectiveness drops, and costs shoot through the roof.
If you want your digital marketing and your online advertising to work. Or if you want to do effective market research, you must have proof. And if you want proof, you must test your theories using scientific methods.
Unless you test your theory, one variable at a time, on a statistically significant sample, you don’t have proof. You have opinion.
In most occasions, it’s always worth looking at the best in the world, to see how they do things. Not only in your own industry, but also across industries. Here we will talk about both: Formula 1 and Google.
Learn From the Best #1: Formula One
Formula 1, or F1, is considered the Queen of motorsports. The 20 best drivers on the planet, driving for the 10 best teams in the world. All competing with each other on the smallest margins you could possibly imagine.
In F1, a mere 1/1000th of a second gained per corner, results in 0.015s gained on a 15 corner lap. Enough to win or lose pole position. Or on a 70-lap race: 0.825s. More than enough to win or lose the race.
The battle is fierce and only the best of the best can taste victory. No more than 3 of the 10 teams have won races in the past 5 years: Red Bull, Ferrari and Mercedes (Jan. 2019).
And during those years, Mercedes has won 74 out of the 100 races. It is also the only team that has won at least one race in each year. So we can assume they know what they’re doing. And what they do is testing.
Every part of the car has sensors. They measure virtually everything, down to the wheel gum. Is it worn at the right angle? If not, adjust it. There is no guesswork involved. No “I saw a YouTube video where the guy said we should paint the front wing yellow cause yellow paint is lighter.”
What they do is they isolate every part of the car. Then they take one part, and they tune it. Then they go on track and test it. Back in the garage, they look at the data, adjust it, and go back to testing. And so they iterate, until they find the best setting for that specific part and for that specific track.
Learn From the Best #2: Google
Let’s go back to the digital marketing and online advertising industry with Google. Google is the leading search engine in the world. It is used for more than 80% of searches from desktop computers, and 95% of searches from phones and tablets.
If Google has reached such a dominant position in its market, it’s certainly not a fluke. There was a debate at Google once, about which color ad links should be. In that sort of situation, people would usually go to the marketing manager who would choose the one he considers best. But not at Google.
To settle the argument, they split tested 41 different shades of blue. They assigned each color to random users, and measured their profitability. The best color was chosen based on hard data, not opinion. And the new color also happened to increase Google’s yearly revenue by $200 million. Seems like a good deal. I am sure that I have convinced you of the importance of data. But how do you get good data?
Getting Good Data for Digital Marketing, Online Advertising and Market Research
When Google decided to improve the color of their links, they didn’t try a whole new interface at the same time. What if Google had changed the color of their links, plus the whole interface at the same time? How could they know where the money came from? Was it the new color, or the new design?
Logic, right? Yet, when people go about measuring things, they tend to change several key components at the same time. They don’t isolate the variables, and they base their entire work on false data. To measure something, you must keep other parts constant. Otherwise you will not be able to know what produced the outcome you observed.
You must also use a sample large enough. Take a dice and throw it five times, you may have five 6s in a row. That’s what people do. They do one thing, it works, and now they strongly believe this is the right thing to do, and the rest is bad practice.
That’s like thinking just because you threw five 6s in a row, the dice always falls on 6. Without a large amount of data, the data you have may not be representative of the reality. To be as close to reality as possible, you need a lot of data. There is no point in killing your Facebook ad because no one visited your website in the first hour. You don’t have enough data.
But remember, large data doesn’t mean good data. You need it to be neutral too. If you want to know whether maths are important, don’t ask the maths teacher.
If you want to sell to the general public, ask the general public, not the specialist. Otherwise the answers will be biased and your data worth nothing.
Conclusion: Your Online Advertising and Your Career on Steroids
I will leave you with this quote from Claude Hopkins:
“Again we come back to scientific advertising. Suppose a chemist would say in an arbitrary way that this compound was best, or that better. You would little respect his opinion. He makes tests — sometimes hundreds of tests — to actually know which is best. He will never state a supposition before he has proved it. How long before advertisers in general will apply that exactness to advertising?”
This was in 1923. So, how long before advertisers will apply that exactness to online advertising? And marketers to marketing? It seems they never learn. But that can be your opportunity to gain an unfair advantage and win by a large margin. If you think it’s time to stop burning the witches, and start producing actual results, check out TCLoc’s Master Program.
Don’t wait to apply, seats are limited.
- David Ogilvy, Ogilvy on Advertising, 1983
- Claude Hopkins, Scientific Advertising, 1923