Building a growth machine

Growth loops are an experimental approach similar to scientific research which involves defining clear growth hypotheses, testing them and scaling best performing actions, discontinuing the worst performers and repeating the process. Through a continuous cycle of experimentation, businesses can build "Growth Machines"–the aggregate of successful channel, creatives and product tests.

To get started


Think about all the activities you have–paid ads, channels, landing pages, product onboarding flows, checkout funnels and everything in between. Then brainstorm ways you can make changes to each of these. Use the Eisenhower matrix or a similar prioritisation process to decide which ones to do first, then define your hypothesis.

A hypothesis needs to be scientific, including success criteria, and what you will do based on your learnings. For example;

We believe influencers are a good channel to scale our brand. To test this we will run an influencer campaign across channels, measure the cost per conversion, and compare it to our baseline of $xx.
If this approach yields better results than the baseline, we will scale influencers. If not, we will focus on other channels.

It's really important to make sure you have a control to compare your experiment against, and that you control variables where possible. Read more about A/B testing here.

Build learning loops


A learning loop is a process in which you make a hypothesis, take action, gather data, reflect on the results, and then use that information to inform your next steps. This process is repeated in a continuous loop, allowing you to continually learn and improve your understanding of your customers and the market.

To get ongoing value to growth experiments, learning loops can be incorporated by setting up rituals for regularly reviewing the results of your experiments, reflecting on what you've learned, and using that information to inform your next steps.

Learning loops also shouldn't apply solely to marketing. Growth is about more than just the marketing mix–remember that Product is one of the 7 P's of marketing!

It's therefore important to also apply the same basic principles to your product. Take an onboarding flow for example, or a checkout journey–test different variations and find little improvements in conversion rates to iteratively improve your revenue. (More on Conversion Rate Optimisation here)

Customer feedback


Another way to incorporate learning loops is to involve customers in the process by gathering feedback and testing new features with a small group of users before rolling out a change more widely.

Is it ever really over?

No. We can always be better. Growth loops should be a constant in your business. Although there is a point in time where you will have built a money-making growth machine,  there is always room for optimisation and improvement. It's also important to remember that markets change, algorithms change, and user and customer needs, wants and cultures are constantly moving–meaning it's important to continuously try and test new things. You also need to press home your advantage and stay ahead of the competition.