There’s no clear cut, one size fits all way to grow your business.
I assume you read a lot of other marketing blogs besides this one. I also assume you get really excited when you read about a new growth hack or unique marketing strategy.
But how do you know it will work for you?
You don’t.
But, if you treat marketing and growth like a science, you’ll have a lot more success.
The first part of treating marketing like a science is taking all the information you consume (from blogs, videos, courses, etc) with a grain of salt.
You never want to make assumptions or allow information you consume to make assumptions for you.
You want to test everything.
Get the best resources and guides you need to grow your business - right into your inbox, twice a week.
This doesn’t just apply to marketing, either. It applies to product validation, product-market fit, advertising, and conversion optimization.
Never assume one tip, strategy or thing will work.
Instead, take a scientific approach to your marketing. This is where data becomes your friend.
The Feedback Loop
Through testing and data, we can create a feedback loop for ourselves and our marketing.
Here’s what a marketer’s feedback loop might look like:
1. Form hypothesis
Here’s where you form the basis for your test. What will you be testing and what do you think the result will be? For example, in one of my Facebook ad campaigns, I might theorize that if I switched the images of my ads from an abstract image, to an image of a face on it, it would get a higher clickthrough rate.
2. Set goal / benchmark
Next, is to set a goal or benchmark. If I’m trying to get a higher clickthrough rate, as in the example above, I might use my current clickthrough rate on my ads as a benchmark for me to improve upon. If this is my first test, I would use the clickthrough rate from this test as my benchmark.
Set a goal, too. If I’m currently getting a 1% clickthrough rate on my Facebook ads, I might set a goal to double it to 2%.
3. Perform a small test
With this new hypothesis formed and a goal set, it’s time to test.
I say small test here because I don’t want to commit to my hypothesis. Again, I don’t want to assume anything. Instead, I want to test my theory while committing as little as possible (while still getting a result I can measure) so that a failed test won’t hurt me.
In this case, I might set a small budget on this ad campaign, as well as only let it run for a day or two before pulling the plug and moving onto the next step.
So instead of changing all the images of all my ads to an image of myself, I’m only going to be testing it for a short time on one ad.
Similarly, whatever you’re testing, try to do it as a split-test or one-off test. If the sample size is too small (say, there’s not enough traffic on your website) you might want to let the test go on longer or put more resources behind it to get better data.
4. Measure results
Once you run your test, measure your results.
This is why tracking everything is so important.
Whether it’s through Google Analytics, heatmap software, or the analytics in the tool you’re using, whatever you’re goal is, it should be easy to measure and track.
If I set a goal of doubling my clickthrough rate but have no way to accurately measure it, what’s the point?
From here, compare the results to the benchmark. Improvement? Your theory was correct. Worse result? Your theory was wrong. Either way, don’t involve your ego. Separate it from your marketing. You shouldn’t care whether or not you’re right, just whether or not what you’re doing can make you and your business more money.
Remember, even discovering something doesn’t work is a win. The point of testing is to not confirm your theories, it’s to see whether your theory is right or wrong. Don’t allow yourself to become a victim of confirmation bias.
5. Adjust / improve off of your results
From here, it’s just a matter of taking your learnings and looking at your results and making adjustments to your marketing, as well as implement your learnings.
This is your feedback loop. Now that you know what does or doesn’t work, you can adjust and continue with more experiments.
Rinse and repeat.
Without a feedback loop like this, your marketing might wind up like a chicken with its head cut off. If you don’t know what’s actually moving the needle, how do you know what to focus your time, money and resources on?
Track Everything
Not only should you be tracking the data from which you’re using to measure your results, you should also track your marketing and your experiments.
Once you come up with a hypothesis, test it, and learn the results, track it in a document like this:
Why?
Firstly, you can take your learnings (from a nice summarized document) and transfer them into other marketing campaigns.
It also allows you to keep track of your results, so you don’t repeat things as well as create a template from which you can create optimized marketing campaigns. If you’re working with a team, it ensures your team doesn’t duplicate your testing efforts, as well as take your learnings and apply them to their projects.
Lastly, it forces you to focus on what works and throw out what doesn’t.
When you start creating a feedback loop for yourself and use it to improve your marketing, you’ll slowly begin to fine tune your marketing into a machine that performs really effectively and efficiently by being very focused.
There’s likely a more elegant way to keep track of all your experiments, but if you or your team aren’t using something like Trello or Asana, keep it simple. A Google Spreadsheet does the trick.
Keep It Simple
It’s worth noting that taking a scientific approach to your marketing is meant to help make your business much simpler, not complicated.
The point of this is to allow you and your business to focus on what works. This shouldn’t be cumbersome or get in the way of your marketing.
This is just a process and approach to starting a new marketing campaign or optimizing one that might not be working as great as it should be.
Remember, never assume anything, especially not in marketing.
While you might think doing X should increase Y, test it, measure it, and then use that data as rule of thumb for the rest of your campaigns. And even then, remember everything is contextual. Just because an experiment proved something about your Facebook ads, doesn’t mean it will work with certainty for your Twitter ads. Hypothesis, test, track, and repeat.
