What is Growth Hacking and Why We Need a Hypothesis Testing Pipeline?
What to do if you want to grow faster? You can’t increase your sales plans and advertising budgets indefinitely. You’ll either spend a lot of money, or all the money at once, and the expected effect won’t happen.
As Andrew Chen said, after a certain time, channels exhaust themselves and cease to be effective, many of them do not scale. So this approach will not lead to explosive growth.
Today we are going to tell you how you can grow by leaps and bounds, what Growth Hacking is, or in other words, how to hack the growth of a company.
What is Growth Hacking?
Is the systematic and continuous testing of hypotheses aimed at company growth. Growth hacking occurs through various, mostly marketing, initiatives. These are exactly what are called hacks.
It is a step-by-step generation and testing of hacks that will lead to growth.
At one time, hacks with the referral program “Invite a friend and get a bonus” or content marketing brought companies 60% growth or tens of millions of dollars in revenue.
Examples of famous hacks
The companies we all know were once small startups without users or revenue. Now they have multimillion-dollar revenues and a huge customer base. All because they hacked growth in a way that would drive you crazy.
Airbnb did not have users or a reputation as a reliable service for finding and renting accommodation. They needed to build a client base in a short period and establish themselves in the market.
To do this, they offered users the opportunity to quickly and easily post their homes on Craigslist, a popular online classifieds site, in addition to Airbnb. This gave them access to a huge target audience.
Dropbox made a hypothesis about the referral program: if a user and his friend whom he invited to the service were given 500 MB each, the number of new users could be quickly increased. It was possible to share a link in two clicks and this accelerated growth.
Thanks to this referral the number of registrations increased by 60%. The hypothesis was confirmed, and now we all know what Dropbox is.
Facebook now has more than 2 billion users. This is the number they have managed to achieve with their hack machine.
One hypothesis was an email notification when a user was mentioned in a post or when a picture of him or she was posted. This helped increase the number of returns to accounts.
Why Growth Hacking is necessary?
helps companies achieve multiple growths through constant hypothesis testing. Multiple growths are two, three, five, ten times, and so on.
How fast you grow depends directly on the number of hypotheses and experiments you test. The more tests you do, the more likely you are to multiply your growth.
What it means to test hypotheses
It’s just like in physics class when the teacher talked about proving the theory of relativity. First Einstein formulated the hypothesis and then set up long experiments to prove it. In the end, Einstein found that the results of the experiments contradicted the generally accepted theory, but this only encouraged him to build a new theory based on the findings. And he wasn’t wrong 😛
A hypothesis is an assumption that requires proof.
In the case of growth hacking, that assumption is aimed at rapid growth. And the experiment is conducted to confirm or disprove the hypothesis. That is, the hypothesis is an assumption, and the experiment is the process of confirming or disproving it.
Let’s assume that if we install an online chat for communicating with customers on the website, our sales will increase by X, – this is a hypothesis. Next, we experiment: we set up the chat and monitor the appeals and sales.
Sales increased due to the chat, so the hypothesis is confirmed.
The hypothesis is a risky assumption that the metric will behave a certain way if we change something. Most often it is a statistical experiment.
The whole point of the growth hacking process is to set a lot of hypotheses and quickly test them with experiments.
Why test many hypotheses
You want to grow fast, so you start testing hypotheses. You can test one hypothesis per week. There are 52 weeks in the year, and subtract the vacations and holidays and you have 44, minus a couple of weeks for sick days, meetings, conferences, etc. So we have 42 work weeks per year.
If we test one hypothesis per week, that means we test 42 in a year. Not bad, but experience tells us that out of ten hypotheses only one is successful.
This means that out of 42 hypotheses, only four will burn out in a year. Imagine that each of these hypotheses brought you 3% growth.
That adds up to 12% for the year. That doesn’t sound like explosive growth.
But it’s not that simple, and to test five, ten, or a hundred hypotheses a week, you need a well-oiled testing process: a hypothesis factory, continuous experimentation, and analytics.
How to run a hypothesis pipeline
You will need a hacking team that devotes all its time to generating, prioritizing, testing and analyzing hypotheses. As we already know, the more testing is done, the higher the chances of getting explosive growth.
Who is a growth hacker?
As our invaluable Andrew Chen said, a growth hacker is such a hybrid of a marketer and a programmer.
When asked, “How do you bring users into a product?” he answers: “With A/B testing of the landing pages, virality hacks, and mailing list deliverability.”
The Growth hacker always relies on data and quantitative measurements and concludes only from them. These are the kind of growth-oriented people you’ll need on your growth team.
The selection of the team itself should start with a master – the person who will be responsible for team processes, planning sprints, and managing the testing process.
In a growth team, it’s hard to have clearly defined positions and full-stack specialists are valued because there are many tasks, and it’s too costly to hire each specialist separately. You should have people who can perform the functions of:
- Marketer who knows everything about channels and traffic;
- layout designer;
- Designer, preferably a UX developer.
The main goal of the team is growing, the main goal is rapid testing of qualitative hypotheses.