What Is Growth Marketing?

If you doubled your ad budget but revenue didn't grow nearly as much, the problem probably isn't the ads. It's what happens after them. When visitors don't sign up and sign-ups don't come back, ad spend becomes water poured into a leaky bucket. Growth marketing is the practice of breaking this "after the ad" stretch into numbers and fixing it through experiments.

What is growth marketing?

Growth marketing is a way of creating growth by using experiments to improve conversion metrics across the entire customer journey, from the moment someone first discovers a brand to the moment they buy again and recommend it to others. It doesn't stop at bringing people in with ads. It takes responsibility for making sure those people stay, spend, and bring in others.

The idea spread in 2010, when Sean Ellis used the term "growth hacker" in his blog post "Find a Growth Hacker for Your Startup." He defined a growth hacker as someone who treats growth as the only yardstick and judges every activity by its impact on growth. Ellis was the marketer behind the early growth of Dropbox, Eventbrite, LogMeIn and others.

Growth marketing is not a channel. It's a way of working.

Is it the same as growth hacking?

They share the same roots, but they're used a little differently. Growth hacking often refers to the clever tactics early-stage startups used to add users quickly and cheaply. Growth marketing is closer to that mindset turned into an operating system an organization repeats every week.

How is it different from performance marketing?

Performance marketing is one part of growth marketing. If performance marketing asks "How much did it cost to bring this customer in?", growth marketing asks "How much did this customer leave behind?" That's why the two can look at the same ad results and take different next steps.

For example, when ROAS drops, performance marketing changes creatives and targeting first. Growth marketing, at the same moment, first checks how many people churn within seven days of signing up. If the problem is the first experience rather than the ads, no amount of ad tuning will bring the numbers back.

How does growth marketing work, step by step?

Growth marketing runs on two frameworks. The AARRR funnel decides what to look at, and the hypothesis–experiment–learning cycle decides how to fix it. You find the single stage of the funnel that leaks the most, fix only that stage through experiments, and repeat every week.

The AARRR funnel: see where it leaks

AARRR comes from "Startup Metrics for Pirates," presented by investor Dave McClure in 2007. It splits the customer journey into five stages and attaches one question and one set of metrics to each.

The stages run top to bottom, but you usually fix them in reverse. Growing acquisition while retention is low just pours more water into a leaky bucket. It's safer to check activation and retention first, and only then increase the acquisition budget.

The experiment cycle: fix one thing at a time

The ICE score and the weekly experiment cadence were laid out by Sean Ellis and Morgan Brown in their 2017 book Hacking Growth. What matters is not the success rate of experiments but their speed. Even a failed experiment leaves a lesson: "This isn't the problem at this stage."

What does a real success story look like?

The most frequently cited case is Dropbox's referral program. When ads didn't pay off, Dropbox switched to giving free storage to both the person who sent an invite and the person who accepted it. Within 15 months, sign-ups grew from 100,000 to 4 million, and 35% of daily sign-ups came through referrals.

They tried ads first, and it failed

Dropbox started with the textbook moves: a launch at a tech conference, search ads, a PR agency. But search ads cost $233 to $388 to acquire a customer, which didn't work for a $99 product. Affiliate marketing and display ads didn't deliver either.

They tied the reward to the product's value

The key to this case is that the reward was storage space, not cash. The reward itself made people use the product more, and folder sharing, the way people already used Dropbox, made invitations feel natural. In AARRR terms, fixing the Referral stage lowered the cost of Acquisition.

What CREATIP has seen firsthand

CREATIP runs integrated marketing communications for FURSYS, a full-line furniture company, and continues to contribute to its revenue growth through branding.

For DESKER, we handled performance marketing, testing multiple combinations of ad creatives and targeting before running ads with the mix that best fit the brand. As a result, conversion rate, click-through rate and return on ad spend all improved year over year, helping expand sales volume.

For iloom, we run SNS-based video campaigns and content to improve the brand's image, and continue to build relationships with customers through content. One approach lifts conversion through experiments; the other creates reasons for customers to come back through content.

Where should your team start?

The first step isn't a new tool or a reorganization. It's filling in your funnel numbers with the data you already have. If you don't know which of the five stages leaks the most, you can't prioritize any experiment. A realistic goal for the first month is not results, but "running one experiment all the way through."

Only the biggest leak, one thing at a time.