Maybe the only rule that’s 100% true about every startup, is that there are no rules when it comes to startups. By it’s definition, a startup is born and has to find a way to develop and succeed in conditions of great uncertainty: you don’t know how the customers will react to your product, you yourself maybe don’t know exactly what your product is capable of under the requirements of the future market, and, probably even more frustrating, in most cases, your future customers don’t even know they need your product in advance. To our luck, there are some smart people out there, that have seen 1-2 startups rise and shine (or rise and fail), and have even build their own companies, so the lessons such people learned along the way can provide insightful tips and strategies that might work or not when you decide to sail such unknown business waters.
Eric Ries know what he’s talking in his recent book “The Lean Startup”, as he himself founded 3 companies, regularly advises startups, large companies or venture-capital firms, not to mention he is entrepreneur-in-residence in Harvard Business School. I devoured his book, and I compiled some of it’s precious lessons in this blog post. This is me taking the time in warning you that this post is long. I tried to keep it short, but the amount of value that this books delivers it’s huge, and I wanted detailed notes for the times I will come back to them later.
This should not impede you reading the entire book, it’s a masterpiece in it’s own, beautifully sustained by real life examples of failures and success stories.
Ries basically starts by defining a startup as a human institution designed to create products and services that uncover a new source of value for its customers, under conditions of extreme uncertainty. If a startup starts with a plan, and perfectly executes this plan for a product they think their customers want, to solve a problem they think their customers have, might lead in the end to a lesson learned on a very high expense. He thinks that planning and forecasting are only accurate when based on a long, stable operating history and a relatively static environment, and startups have neither. After much invested time, money and sleepless nights, a scenario can be that no one wants the product, no one uses the service. Ries uses the term “achieving failure” to describe this process: successfully, rigorously executing a plan that will turn out to be utterly flawed.
What a startup should do instead, is to figure out the right thing to build, the thing customers want and will pay for, as quickly as possible. And this is exactly what the lean startup methodology is planning to help entrepreneurs do: engage in validated learning by using the build-measure-learn feedback loop to drive a successful startup.
If you are here, that means you are interested in startups or business in general, and if you are interested in startups, you have heard/read at least one time until now that an entrepreneurs biggest mistake is to wait the launch to happen when he/she thinks the product it’s perfect. And that the basic lesson with the build-measure-learn loop as well, described below in 6 simple steps:
Step 1: Start by a list of important assumptions about your product, business and customers.
Step 2: Based on these, build a minimum viable product with a minimum amount of effort and the last amount of time.
Step 3: Launch it -imperfect as it is-, not to your average customers, but to your early adopters – those customers who feel the need for the product must acutely.
Step 4: Develop an experiment system and test these assumptions guiding yourself by simple, tangible, measurements.
Step 5: Validate or invalidated these assumptions, and keep what proved to be real.
Step 6: Gather customers feedback, improve, test again, learn as you go and build upon the truths you are discovering.
You start with some assumptions about your product, and based on these assumptions you build a prototype of your product, and you just toss it out there. I know, the perfectionist in you is freaking out right now, but this method has proved to save many new companies a lot of sleepless nights, energy, and money creating something nobody needed.
Every plan you might come up to launch a product, is mostly based on your assumptions about the customers needs or problems, about their reaction to your solution, about their behaviour and so on. You have to break down the big picture, into smaller assumption rather then facts, and start experimenting and testing them. The most important assumptions are the parts of a startup plan on which everything depends on: the leap-of-faith assumptions. the most important ones are:
– the value hypothesis: “tests whether a product or service really delivers value to customers once they are using it”
– the growth hypothesis: “tests hot new customers will discover a product or service”
A minimum viable product (MVP) allows entrepreneurs to start the learning process as quickly as possible and test these vital hypothesis. A MVP is often perceived as low-quality by the customers, allowing you to learn what attributes the customers really care about.
Measuring learning. Innovation accounting.
So you’ve got your MVP out there and you get feedback from the customers. Based on this feedback, you will want to improve your MVP, logically. These changes can prove successful, or not – the questions are: How do you know that the changes you’ve made are related to the results you are seeing? How do you know you are drawing the right lessons from those changes? Eric Ries introduces the term “innovation accounting” to save the day and validate your startup growth model.
Every product development, marketing or other initiative a startup undertakes should be targeted at one of the drivers of it’s growth model, also called baseline metric: conversion rate, sign-ups, activation rate, trail rate, customer lifetime value etc.etc. Set out with one clear, relevant, baseline metric you want to improve, a hypothesis about what will improve that metric, and a set of experiments designed to test that hypothesis. The most important thing in this equation is the metric you choose to look at when running an experiment. Ries distinguishes between vanity metrics and useful metrics. Vanity metrics are gross usage numbers, gross customers numbers, registration numbers etc. The problem with these metrics when running an experiment, is that they don’t offer a clear cause-and-effect inference.
For a metric to be useful, it has actionable, accessible and auditable.
Actionable metrics – have to demonstrate a clear cause-effect, and what actions would be necessary to replicate these results.
Accessible metrics – are cohort-based reports that basically says that “among the people who used out products in this period, here’s how many of them exhibited each of the behaviours we care about”.
Auditable metrics – data has to be credible, so it’s always better to double check data with conversations with real customers.
Best way to test a hypothesis, is to conduct a split-test experiment and use cohort metrics. A split-test experiment “is one in which different versions of a product are offered to customers at the same time. By observing the changes in behaviour between the two groups, one can make inferences about the impact of the different variations.”. It also helps figure out what customers want and don’t want.
What happens if the experiments you conduct prove that a vital hypothesis is false and you have to make a major change? In this case, a startup has to find the strength to pivot – “a structured course correction designed to test a new fundamental hypothesis about the product, strategy and engine of growth”. This proves sometimes to be incredibly difficult, as one of the very fundamental traits of an entrepreneur is to madly believe in his business idea. It requires a great deal of strength to admit that the product, as you envisioned it, it’s flawed and required drastic change.
Types of pivot
Zoom-in Pivot = a single feature of the product, becomes the whole product;
Zoom-our Pivot = the whole product becomes a feature in the new product;
Customer Segment Pivot = the product solves the problem, but for different customers as initially thought;
Customer Need Pivot = the problem the product solves is not important for the customers;
Platform Pivot = a change from an application to a platform, or vice-versa;
Business Architecture Pivot = a change between two major business architectures: high margin & low volume and low margin & high volume;
Value Capture Pivot = refers to monetisation and revenue models and can take many forms;
Engine of Growth Pivot = a change between the viral, the sticky and the paid growth models (explained below)- it requires change in value capturing as well;
Channel Pivot = a change in sales or distribution channels;
Technology Pivot = a different technology that allows a way to achieve the same solution.
Engine of growth
When testing the hypothesis, we are in fact testing our engine of growth: how will our startup achieve sustainable growth? Ries distinguishes between three engines of growth:
The sticky engine of growth – build to attract and retain customers for long term and rely on high customer retention rate. The company grows if the rate of new customers exceeds the churn rate (the fraction of customers in any period who fail to remain engaged with the company’s product).
The viral engine of growth – growth happens automatically as a side effect of customers using the product. The speed is given by a high viral coefficient= how many new customers will use a product as a consequence of each new customer who signs up.
The paid engine of growth – each customers pays a certain amount of money for the product over his/hers lifetime as customer. We deduct the variable costs, and have the so called customer lifetime value (LTV) that can be invested in growth by buying advertising. If an add that costs 100$ brings 50 new customers to sign in, the cost per acquisition (CPA) is 2$. If the product has an LTV created than 2$, the business will grow.
Ries then moves on and explains some lean manufacturing tools that can easily applied to startups, but the explanations are based on examples, and it would be inappropriate to explain them here. I highly recommend getting your hand around this book and see for yourself how lean startup has worked in the past.
I am already looking to my next read on startup grow, and can’t wait to tell you all about it! 🙂 Until then, you can get a teaser in the video below.
Until next time, dare to play with your ideas!