Skip links

Why science is the number one growth hack for your business

“What’s the number one growth hack you’d use to grow my company?”

The most recurring question prospects ask us in sales meetings is also the one we love most.

As soon as the question is popped, a trace of a smile plays across my lips.

It’s the perfect moment for us to cut to the chase and enlighten the person sitting across the table on what makes our agency different from others.

It’s an open goal. I can’t miss.


The room falls silent.

They expect a silver bullet.

I give them an idea, a process, a methodology.

And explain why all the ‘increase traffic with 300% in two weeks’   hacks aren’t working for them.

Growth hacking is a quest for truth

There’s an old legend in which Socrates, the great Greek philosopher, is recognised by the Oracle of Delphi as the wisest man alive.

He proclaims:

“I am the wisest man alive. I know that I know nothing.”

When I get into a sales meeting with a prospect, I know nothing.

I don’t know their product, I don’t know their audience, I don’t know their business model and I don’t know their customer journey. I don’t know their business and neither does any other external marketing agency.

The difference with most marketing agencies: we acknowledge that ignorance and use it as a tool to build growth machines.

There’s no sense going into tactics when you’re still in the dark about a business and the relation to its audience. What works for someone else, isn’t likely to work for you.

Copycat marketing isn’t a thing. Every business is unique and should be treated as such. Blindly implementing tactics is more likely to blow money and harm a brand than it is to drive growth.

Marketing is about understanding and influencing customers behaviours. We aim to make people react to stimuli in predictable ways. For every marketing action, there is a reaction and we want to be able to anticipate those reactions as much as possible.

That makes marketing a never-ending quest for truth. Never-ending because the idea of truth itself is an illusion. As human beings, there’s only so much we can grasp from reality. It is too vast and complex for our brains to fully make sense of, and it changes at unprecedented speeds.

A particular tactic not working for you comes down to a lack of understanding of your customers. Of what they need, of what moves them, of where they hang out, of how they liked to be talked to and of how all of that comes together in a brand.

You’re spending you’re time guessing what might work instead of finding out what will work. Burning money and time meanwhile.

Knowledge is power. The closer you get to the truth about your business’ reality, the more power you’ll have to direct that reality and drive growth.

The tactics you’re so desperately looking for come with the quest. What we call a ‘growth hack’ is the result of a thorough understanding of your target audience and its relation to your business. You find one by slowly making sense of the complex reality that is a business.

That’s where science comes in.

Dropbox’s secret sauce wasn’t the referral program

From September 2008 to January 2010, Dropbox grew from 100.000 to 4.000.000 users.

You probably already know what happened.

Sean Ellis drove virality by inventing the modern referral program: users who got their friends to sign up for Dropbox received 500MB of extra storage, an enormity at the time.

Businesses from industries all over the world were quick to ‘steal’ the referral program, only to find out it wasn’t performing for them as it was for Dropbox.

They copied the wrong strategy.

Dropbox’s success isn’t built on the referral program, it’s built on the process that gave rise to the referral program.

That process is a scientific one, built on experimentation.

Before doubling-down on the referral program, Dropbox conducted small experiments to find out how far users were willing to go for free storage space.

Experiments like these learned Dropbox that users were willing to perform certain actions to gain extra storage. They went from connecting social media accounts to inviting friends and eventually mass-inviting friends through a clever integration with Gmail.

They didn’t discover the referral program as a growth hack overnight. It was the logical consequence of an ongoing loop of small-scale experiments. Experiments that gradually revealed the referral program as a major growth lever for Dropbox.

The simple reason this referral program doesn’t work for you the same way it works for Dropbox is that your customers aren’t the same as Dropbox’s.

Sounds obvious and yet every day again people ask me if I can give them plug-and-play growth hacks to grow their businesses.

This obsession with tactics over process is the number one reason why you’re not achieving sustainable growth and why growth hacking as a discipline has been getting a bad reputation. Your growth hack isn’t someone else’s.

To quote growth hacking godfather Sean Ellis:

“Sustainable growth comes from understanding best customers and figuring out how to find and acquire more of them.”

Similarly to how Dropbox learned its users would be willing to refer friends to get more storage, Twitter found out all of its active users followed at least thirty people and Airbnb found out its target audience was looking for vacation rentals on Craigslist.

Both then turned those learnings into effective tactics. Twitter started showing users interesting people to follow in the signup process to boost activation. Airbnb built an integration to cross-post listings on Craigslist to drive awareness.

The rest, as they say, is history.

These examples come to show that growth comes from a process of relentlessly pursuing the truth about your customers. Not from blindly copying tactics that work for someone else’s customers.

The minimum viable experiment: speed beats perfection

The best marketers are experimenters, not fortune-tellers.

They spend their time finding out the truth rather than proclaiming they already possess it.

They know that their perception of reality is limited and that their assumptions about it are merely assumptions: guesses about how their strategies will appeal to customers and drive growth. They know that these assumptions may be wrong and embrace uncertainty as a tool for better decision-making. They turn assumptions into hypothesis & experiments and create opportunities to get closer to the truth.

The best marketers will stay away from trying to show off with tactics when a prospect asks them for their number-one growth hack. Instead, they’ll lay out the process and tools they’ll use to find that number-one growth hack and all of its friends. They seek power by pursuing truth.

The beauty of experimentation is that you always win because you always learn. Whatever the result of your experiment: you’ll understand the relationship between a business and its audience better. You’ll find out ways to improve existing tactics plus maximise your odds to discover completely new ones. The only way to lose is to not experiment as you’ll stay stuck in the dark guessing whether something works or not.

I call that ‘maybe land’.

All that knowledge stacks up. Jeff Bezos attributes a big part of his success to running experiments.

“If you double the number of experiments you do per year you’re going to double your inventiveness.” – Jeff Bezos

The best part is that Bezos’ assessment is inaccurate here.

If you double the number of experiments you do per year, you’re going to more than double your inventiveness.

Think about it. Every experiment widens and/or deepens you knowledge. New knowledge allows for better experiments, better experiments allow for richer insights and richer insights again allow for better experiments. Rather than double, your inventiveness will grow exponentially. This is the compound effect.

Speed of execution is key here. The goal of experimentation is to learn and stack up knowledge quickly, not to be perfect and look pretty. If you have a new idea to acquire customers, you should test it in the real world as quickly as possible before designing a whole strategy around it.

This also has to do with resources: why spend huge amounts of time and money on Instagram ads if you have no clue if they will work on your audience? As soon as an idea hits you, you find a way to put it out in the real world as fast as possible and see what happens.

It doesn’t have to be good, it has to be good enough to validate or dismiss your assumption underlying the idea: a minimum viable experiment.

Turn assumptions into experiments

To prospects, we like to describe the growth hacking process as a journey of discovery.

You strand ashore a desert island.

No, this isn’t one of those “if you could bring two things to a desert island, what would they be” kind of things.

You need to find out how to get water and food. Where to find it, how to capture it, how to store it.

On the first day, you spot a boar in the woods. There are two scenarios.

Scenario number one: you assume

Boars will be your main source of food. You make a spear, hunt down a boar and have enough food for a week. It went well. You don’t know how to conserve the meat but you’ve seen quite some boars and assume there’ll be enough for you to survive. However, hunting down boars with a spear takes a lot of energy, you’re not getting enough vitamins from just the meat and you have to catch a new boar every week because you can’t store the meat. By the time you realise you should have been looking for new sources of food, better gathering techniques and storage methods, it’s too late.

You die.

Scenario number two: you experiment

While hunting for boars, you explore the trees for fruits and observe monkeys to see what you can and can’t eat. You carry a spear but you’ve also set some traps. You spend a week only spear-hunting, a week only trap-setting and a week combining the two to see what gets the best results. You try out different ways of storing food. You spend your days learning about your environment and find new sources of food, better gathering techniques and storage methods.

You survive.

The analogy has its limits, but the essence is the same.

Google Adwords may work for you today, but there’s no guarantee they will work just as well tomorrow. Customers, tactics, channels, technologies, competitors: they’re changing constantly, at unprecedented speeds.

You can either continue blindly running the same Adwords every day or you can test new keywords, put up different landing pages and experiment with Facebook ads, Instagram giveaways and referral programs meanwhile.

In scenario number one, you die. In scenario number two, you survive.

And thrive.

It all starts by translating your assumptions into hypothesis and setting up experiment. In order to help you get started and first of all translate your vague assumptions into measurable hypotheses, we wrote a blog post and developed the experiment card.

The scientific process is your most powerful marketing weapon

You don’t need to be Einstein to practice science. Science is about pursuing curiositu, acknowledging ignorance and relentlessly trying to bridge the gap between the two. The scientific process is designed to achieve exactly that. It will remove the guesswork from your marketing and will turn your company’s growth into something scalable, predictable and repeatable. It goes something like this:

  1. Set goals
  2. Generate ideas
  3. Design experiments
  4. Execute
  5. Study data
  6. Rinse and repeat


Step 1: Set goals

Always start with the end in mind. Good experiments originate from the goals you’re trying to hit. Are you looking to drive more traffic to your website, reactivate idle users or increase revenue from upselling? Simply ‘drive more traffic’ isn’t going to cut it. To design effective experiments, you want to break down your objective in smaller, more manageable parts  –  following the SMART criteria: specific, measurable, assignable, realistic and time-related.

Let’s assume you run an e-commerce business with 14.000 monthly unique website visitors and an average spend of $7.3, resulting in $1.226.400 annual revenue. To get to $2.000.000 by the end of next year, you’ll need to either grow your website traffic to 22.832 monthly unique visitors or increase average spend per visitor to $11.7. Your goal could look like this: We want to increase website traffic to 23.000 monthly unique visitors within 90 days  – weekly growth rate of 13%. Here’s how that aligns with the SMART dimensions.

  1. Specific: increase website to 23.000 monthly unique website visitors.
  2. Measurable: 13% per week.
  3. Assignable: We , the marketing team.
  4. Realistic: Up to you and the context of your business to decide what is ‘realistic’.
  5. Time-related: How much time are you giving the team? Good goals are set between 30 and 90 days.

Base your objectives on the pirate funnel to align your experiments to stages in the customer journey. Driving website traffic is just one example of how you could trigger growth. Every stage of the pirate funnel offers opportunities in this respect. Be smart about which stage of the funnel you’re going to grow first: it doesn’t make a lot of sense to double traffic to your webshop if 95% of your visitors leave without spending. Data can help you to prioritise objectives.


Step 2: Generate ideas

After the ‘what’ comes the ‘how’. This is where you come up with ideas to reach your objectives.

In this case: to increase traffic with 13% in a week. Try to find a balance between ideas you base off data from past experiments and totally new experiments. That way you’ll be able to improve existing tactics as well as maximise your odds to discover new ones. Let’s say past experiments have proven Instagram ads to be an effective traffic driver. You can try to improve those past experiments in terms of media, copy or targeting to gain a bigger increase in traffic. Another idea is to start following influencers on Pinterest because you’ve seen competitors are doing it you and want to find out if it would also work for your business. More ideas:

  1. Team up with other e-commerce businesses for a big giveaway and share email lists and Facebook audiences to cross-promote it.
  2. Create a chatbot that helps people choosing the best gifts for their mother, father, sibling, friend etc.  Write a blog post about it, share on social/in relevant communities and post on Product Hunt.
  3. Start shipping stickers with orders to increase offline exposure.

Come up with as many ideas as possible, then select the ones you want to focus on in the upcoming weeks and put the rest in your backlog to revisit later.


Step 3: Design experiments

Turn ideas into experiments by identifying:

  1. The hypothesis to validate.
  2. The variable to test.
  3. The metric to measure.
  4. The criteria to base success off.

For Pinterest that would look something like this: As soon as I’ve confirmed that following influencers on Pinterest works, I can run further tests to determine the characteristics of the influencers most likely to follow me back and pin my products. The more experiments I run, the more I’ll get to know Pinterest as a social platform and the more I’ll be able to eventually turn it into a traffic machine. Stay true to the idea of the minimum viable experiment.

  1. Don’t waste time on perfecting copy or design. Implement whatever is need to validate/dismiss your hypothesis.
  2. Don’t burn your monthly budget on one ad campaign. Start with a fraction and scale up as it gains traction.
  3. Don’t make ten changes at once: test one element at a time to learn its relative weight in relation to the other elements.
  4. Don’t argue about which ad looks the best: A/B test all of them in the market and the best one will automatically emerge.


Step 4: Execute

Time to unleash your experiments into the world. This is about doing the work. Work hard, but smart. Use tools to facilitate and automate non-creative work. Growth teams like to draw inspiration from Google’s design sprints to plan experiments, manage time and assign tasks.


Step 5: Study data

Arguably the most important stage of the process is also the most overlooked one. About every marketing agency these days claims to be ‘data-driven’ but few of them really are. It’s not hard to see why. Marketers typically don’t like numbers. They chose arts, culture and creativity over science and math in school. Now that the age of Mad Men is over and everything can be tracked, the world demands them to trade in gut feeling for numbers, but the skills to read data aren’t there and the shift doesn’t happen overnight. Still, the promise of ‘data-driven’ sells and so they make it, because they’re marketers.The whole point of the scientific process is to discover new knowledge and internalise it to gain power over your environment. There’s no sense in running experiments if that new knowledge is then ignored, overlooked and not captured. Take the time to look at the results of experiments and try to understand the surrounding ‘why’ by combining quantitative with qualitative data. Here are some guiding questions:

  1. What were the results of the experiment?
  2. How valid was the initial hypothesis?
  3. Why are the results what they are? Try to understand the whole story, not just the occurrence.
  4. Are there ways to segment or combine data to reveal new insights?

About 80% of your experiments will fail, but that doesn’t mean you’ve failed. The more data you gather, the better you can align your assumptions to reality and the better the ideas you’ll come up with after. We’ve been here before, but this whole idea can only work if you take the time to internalise the learnings from your experiments.


Step 6: Rinse and repeat

Congratulations, you’re smarter now. Now use that to become even smarter. Align goals, scale successful experiments, adjust failed experiments, come up with new ideas, design better experiments and harvest new learnings. To then do it all over again.


Scientific marketing is never ending! 

A never-ending quest for the truth. There’s always more things to test, more experiments to run, more knowledge to gain. You can never win the game but the more you play, the better you get and the more you leave competitors behind.


Ready to remove the guesswork from your marketing? At The Growth Revolution, we use scientific methods to find and drive sustainable growth for your business. Reach out & let’s see how we can help you with your experiment design.