29 Aug What Is A Growth Experiment?
It’s no secret that buzzwords get tossed around in the marketing field like hotcakes. Every now and then things change and marketers all around get hyped. Growth hacking has been around for some time and it’s high time your marketing team embraces its power. Every business wants to grow, and with all the data being tossed around in the cloud, there is huge potential in approaching marketing from a new window: growth experiments.
But what is a marketing growth experiment? It is a series of tests that help marketers validate certain assumptions and eventually grow. It’s a short-term cycle used by marketers to collect data and find out what is the most effective way to grow a business. If you look at the traditional process of marketing, it’s safe to say it’s rather stiff. Marketing managers get together at the end of every year to create one-year marketing plans. What comes next is a series of big investments and execution plans to meet business objectives and KPIs. Growth experiments are an alternative approach which can give a whole new perspective to your business. You might realize you don’t even need a marketing investment to grow your business. [link to McKinsey’s report]
So what if we say you can change that whole process? In this article, we will thoroughly cover how growth experiments can be planned, executed, and repeated. At Refreshworks, we are hardcore believers in accepting failed experiments and building on successful ones, because all experiments lead to significant results somehow. It’s all about the mindset – but more on that in our How to Adopt A Data-Driven Culture post. We have started our company to help businesses learn more about their target audience and create tailor-made content that can be optimized over and over. And this is how we do it:
1. Set your objective:
As opposed to the traditional marketing funnel, the growth hacking funnel is composed of 3 A’s and 3 R’s: Awareness, acquisition, activation, retention, revenue and referral. When you spell that out, you get an AAARRR, which is why we call this the Pirate Funnel. We recommend you choose only one metric per experiment and focus on it.
2. Research the target audience:
Segmenting your target audience to personals based on psychographics or consumption moments is one way to do it – but the fault is that this can still be assumption-based. Adopting a data-driven approach to learn more about your target audience can result in insights you never thought of! Here are some ways you can research:
Hard data: through analyzing your database or using machine learning algorithms
Soft data: through reviewing case studies, usability testing, consumer behavior analysis, geographic or channel research
3. Develop your strategy:
The strategy to be formulated here will be derived from the objectives, choice of channels or platforms to test on, and the decision to go with organic or paid reach. Once your marketing growth experiment strategy is in place, you can then formulate a series of hypotheses, such as “Target A and Target B will convert easier on Facebook than on Instagram”. There are so many ways to formulate hypotheses and we’ve compiled a list of them here.
4. Create your content:
While creating content for an experiment, it’s important to keep in mind that only the elements to be tested are variables and all else should be rather controlled. This will allow for a clearer reading of the results and allows you to attribute different results in a clearer way.
5. Distribute your media:
Although you might be tempted to share your media on multiple channels and share your work with the world, it’s critical to focus on the experimental nature of your campaign. Media distribution should be done only on the channels that you’ve selected in your strategy. It’s good to keep in mind that each channel has its guidelines and so your content should be optimized accordingly. Those beautiful cinematic horizontal videos are no good for an Instagram story!
6. Data & Results:
To assess your data, you first need to ensure that it’s statistically significant and this varies from one test to another. Once you analyze the results and cross-check them with your hypothesis, you’re ready for your next series of experiments! If you’re curious to know what we can do with your data, book a meeting here.
7. Repeat the loop!
Simply make a decision which tests to continue with and which experiments to kill by looking back at your objectives.
Now that you’re more familiar with the process itself, let’s go through the two main types of experiments. At Refreshworks, we can support you with Learning Experiments and Optimization Experiments. Do you ever wonder how your target audience or customers think of you? Ever speculated what kind of communication you should address them with? We believe those days are long gone. With a quick survey, you can collect data about your audience and tailor-make content for them. Follow through their user journey by tracking their clickthrough behavior. Optimize. Question more. Try new things. Repeat.
There’s no doubt that the growth mindset is scientific, but it’s also truly focused on having an open conversation and understanding the psychology and behavior of every individual. With these tests, you can learn so much about what works and what doesn’t. So, what do you say? Are you ready for take-off?