Session 35: A Scientific approach to advertising

In session 35 of A Marketing Podcast I’m talking about executing a scientific approach to advertising. The approach provides you with a way to constantly compare and contrast the many different options and approaches there are to performing advertising, both online and offline.

Session 35: A Scientific approach to advertising

Thanks for joining me again for Session 35 of A Marketing Podcast, sorry about last week, the dreaded flu got me and I was in bed for a couple days.

But today’s sessions, I think, is worth the wait. Especially if you are doing your own advertising at the moment, or thinking about it in the future.

Today’s going to be about a scientific approach to advertising. buts its not a complex as you think. It’s about performing experiments in order to continuously improve the advertising that you perform and to maximize the bang for your buck.

Today’s show will include:

* The structure of a science experiment

* Developing a hypothesis

* Controlled elements in your experiment

* Variables in your experiment

* The experimental procedure

* Analysis and your findings

So let’s get started.

The structure of a science experiment

Remember back in high school when you were asked to prove a hypothesis and had to do so with controlled elements and variables elements.

Let’s review shall we.

All science experiments require the following to develop an insight or develop a fact:

* a hypothesis

* controlled element/s

* variable element/s

* procedure

Developing a Hypothesis

What is a Hypothesis – a supposition or proposed explanation made on the basis of limited evidence as a starting point for further investigation.

It’s the thing we’re trying to prove as true or false put simply.

In advertising this is the hardest part of the process. But the most important as well. Our hypothesis can be virtually anything but here’s an example to get you thinking…

“Is Facebook advertising better for my business than Display advertising through Google?”

or even better

“Facebook delivers more conversions for the same media spend then display advertising through Google”

As you can see I refine the hypothesis to help me establish something more measurable and specific – which makes it easier to prove/disprove.

Controlled elements in your experiments

The controlled element in your experiment will allow you to test different variable elements against each other. You must have at least one controlled element to execute a scientific approach to advertising.

The controlled element, may also require multiple elements such as in our previous example the controlled elements would be:

* the same budget spend

* the same targeting

* the same conversion metric

* the same advertising messaging (if not exactly the same words)

The more precise you can be in making as many things exactly the same the easier it is to pinpoint the variable and determine your hypothesis.

The variable elements

The variable elements are those that differ in your experiments. The variables for your experiment can be many, however you must work to have as few variables as possible in each experiment.

Again using our example, the only variable we want left over is the channel we’re using, be it Facebook or Google Adwords.

Now how do we actually perform the experiment???

The procedure to perform our experiments

In order to execute a scientific approach to advertising your experiment you must have a set procedure to test our hypothesis, by managing your controlled elements, and comparing the variables.

Here’s a simple procedure for our experiment.

1.