More on this book
Community
Kindle Notes & Highlights
companies that grow the fastest are the ones that learn the fastest. The more experiments you run, the more you learn.
data analysis and insight gathering, idea generation, experiment prioritization, running the experiments, and then returning to the analyze step to review results and decide next steps, in a continuous loop.
The cycle is managed by a one-hour weekly growth team meeting to review results and agree on the next week’s set of experiments to implement.
the growth lead should clarify the role to be played by each team member and how they are expected to work both individually and collaboratively to support the team’s work.
data analyst to share the results of the initial analysis done, and the growth lead should present the key growth levers, the North Star metric, and the area of focus or objectives for the team.
set the goal for the volume and tempo of experiments to launch each week;
Next, the team will need to uncover what the aha moment is for people who are using the app, and what it is about them and their usage that differs from those who don’t.
growth lead works with the marketing and data team members to think through the app’s growth equation,
take the next week to brainstorm and percolate ideas for what experiments to run in the first cycle;
marketing expert on the team conducts a set of user surveys and a set of interviews.
STAGE 2: IDEATE
As Linus Pauling said, “The best way to have a good idea is to have lots of ideas.”
growth lead should set up a project management system to coordinate the submission and management of ideas, as well as the tracking and reporting
description must include the why—the rationale behind the idea—and the how—a recommendation of the type of test to be done,
HYPOTHESIS: Like in any other type of experiment, the hypothesis should be a simple proposition of expected cause and effect.
METRICS TO BE MEASURED:
The scope of measurement includes how many people use the new feature and also the feature’s impact on their purchasing behavior,
STAGE 3: PRIORITIZE
ICE score system, with ICE standing for Impact, Confidence, and Ease, as a way to organize all the ideas
TIR system, which stands for Time, Impact, and Resources.3 Another system is PIE, for Potential, Importance, and Ease.
STAGE 4: TEST
USE A 99 PERCENT STATISTICAL CONFIDENCE LEVEL: Many tools automatically set, or let you set for yourself, the confidence levels for tests. Common levels are 95 percent and 99 percent. While the difference of four percentage points may not seem like much, it is actually quite significant from a statistical point of view. A 95 percent confidence level means that a “winning” test can still be wrong 5 percent of the time. That means that 1 out of 20 tests that come back as winners could actually be losers. At 99 percent confidence, however, that number of false-positive tests drops to 1 out of
...more
written up in a test summary,
The growth meeting is held on Tuesdays, which provides the team with a day at the beginning of the week to finish some of the requisite prep work.
15 MINUTES: METRICS REVIEW AND UPDATE FOCUS AREA
10 MINUTES: REVIEW LAST WEEK’S TESTING ACTIVITY
15 MINUTES: KEY LESSONS LEARNED FROM ANALYZED EXPERIMENTS
15 MINUTES: SELECT GROWTH TESTS FOR CURRENT CYCLE
each member to give a brief overview of the ideas they’ve nominated, which is followed by a brief discussion among the team about the merits of those experiments.
5 MINUTES: CHECK GROWTH OF IDEA PIPELINE
Before Sean and the team at Dropbox implemented the referral program, the company was spending nearly $400 to acquire each new user and the premium subscription price was just $99.
you should not launch into a full-court press for large-scale customer acquisition until you’ve achieved product/market fit—i.e., until you’ve determined not only that you have a good product, but that your product is compelling to its target market.
language/market fit, which is how well the way you describe the benefits of your product resonates with your target audience, and channel/product fit, which describes how effective the marketing channels are that you’ve selected to reach your intended audience with your product,
The term language/market fit was coined by James Currier (who we met in the introduction) to refer to how well the language you use to describe and market your product to potential users resonates with them and motivates them to give it a try.
“1,000 Songs in Your Pocket.” Rather than spend his time trying to differentiate his product from others on price or features, in other words, Jobs understood that the core value, the magical aha experience, was carrying your entire music library around with you anywhere, all the time, totally hassle free.
It’s the Real Thing” and “Just Do It” so impressive. Each of those phrases is so simple—there is nothing particularly poetic or even distinctive about the language—and yet they are both powerful and memorable.
“Find a Date,”
“Help People Find a Date,” and sure enough, users started sharing invites with their friends, even sending them to married people because, after all, they can help their single friends with finding dates, too. The service added 29 million users within 8 months of making the single change.
as you plan your first hacks to test, start with language, and then go from there.
Marketers commonly make the mistake of believing that diversifying efforts across a wide variety of channels is best for growth. As a result, they spread resources too thin
to put “more wood behind fewer arrows.”
There are two phases in which to home in on your best channels: discovery and optimization. In the discovery phase, the growth team should experiment with a range of options, and this does not mean trying all sorts of things haphazardly to see what sticks. Channels must be researched thoroughly, then prioritized down to a few to target for experimentation,
three basic categories: viral/word-of-mouth, organic, and paid.
Let’s say the scores come out as follows:
to be truly viral, the product must have a viral coefficient (or K-factor) of greater than 1. This means that each new user who signs up brings in one or more new people to the product. Yet achieving this degree of virality is extremely rare, and when achieved it’s often for only very brief periods of time.
virality of any product is controlled by three factors: payload, conversion rate, and frequency, which can be expressed by the simple rule: VIRALITY = PAYLOAD × CONVERSION RATE × FREQUENCY Payload is the number of people to whom each user will likely send the promotion (or link, widget, etc.) at a time.
CREATE AN INCENTIVE THAT’S IN SYNERGY WITH YOUR PRODUCT’S CORE VALUE
product/incentive fit. This was the genius of the Dropbox referral program.
