[Book Summary] Lean Analytics by Alistair Croll & Ben Yoskovitz
Leveraging information to help startups focus on the right metrics in order to embrace growth faster!
If you are familiar with Eric Ries’ The Lean Startup book, which focuses on a sequence of three main pillars - Build, Measure, Learn - in order to build a fast-learning and adaptive environment in a startup, Lean Analytics will tackle the Measure part: what should we be tracking in order to confirm whether we are meeting the expectations or not.
The book is structured in an ordered and easy-to-read fashion. Nevertheless, what makes it super valuable, in my opinion, is that it contains lots of examples that will help you grasp the key ideas behind each chapter.
MAIN LEARNINGS
Why you need data to succeed: Be data-informed, not data-driven
Data, and hence information, helps entrepreneurs remain accountable about how their start-ups are going and avoids them being deluded around intuition.
The author states that founders lying to themselves about assessing their companies’ success is more usual than not, since they need to sell their products/services as much as they can to derive growth.
This, according to The Lean Startup methodology, could end up in a terrible situation: imagine you are not tracking the correct metrics, looking at the wrong charts, or that your metrics are poorly defined. How are you going to know where are you standing towards your goal?
Numbers are important, but they might lose sense if we do not apply our own judgement and business knowledge: be informed by data (draw your own conclusions), but do not be driven by it (do not let a single metric drive your intuition away). Data should always be a very useful tool to help you make decisions.
The Lean Analytics Cycle: Find, Experiment, Repeat
Highly influenced by the Lean movement, the Lean Analytics Cycle couldn’t be less. It relies heavily on a try-and-error fashion that will help you adapt rapidly to the different stages and requirements of your young company by following three crystal-clear steps:
Find a metric that you think is important enough to optimize, and define the features that might affect this KPI
Experiment to improve the specific metric by testing with different features
Rinse and repeat: by looking at our previously defined baseline, we will know whether we can move on to the next phase; or keep experimenting with different set-ups and features.
A good metric should be:
Comparative.
Understandable.
Usually a ratio or a rate.
Actionable.
Finding your North Star metric: Models and Stages
The concept One Metric That Matters (OMTM), also known as North Star metric, is defined in the book as “a single metric that’s incredibly important for the step you’re currently working through in your startup”.
Two key ideas arise from this definition:
The OMTM is mutable, meaning that you need to be adaptive enough to find the correct metric at a given moment in time.
The OMTM is singular, meaning that you need to be specific enough to be able to define and aggregate a single metric that will drive your business efforts for the following phase.
As you might have guessed, OMTMs might vary greatly depending on which business model your company is based upon. The book contains six examples of different kinds of business models, and suggests many OMTMs, even though I will only show the most relevant ones from my point of view:
E-Commerce - Revenue per Customer (a.k.a. Customer Lifetime Value, CLV), since it is an aggregated metric that will take into account many other key figures.
SaaS - Engagement & Churn, since these will dictate the adoption and loyalty of your users.
Mobile App - Average Revenue per (Paying) User, since it maximizes the revenue/engagement relationship.
Media Site - Focusing on what kind of audience you have and the ads displaying system will rule in here.
User-Generated Content - Leveraging the engagement funnel with In-App or Push notifications will drive this kind of businesses.
Two-Sided Marketplaces - Focus on buyers over sellers, at least at the beginning. Creating artificial supply to attract demand, and focusing later on Buyer/Seller growth, Conversion funnels and Ratings.
Now, what’s interesting is that these metrics will mean nothing if you choose to use them on the wrong moment in time. This is where the different startup Stages come in handy:
Empathy: you have identified the problem and a possible solution, but you need to make sure that it is worth solving.
Stickiness: your product/service is out, now you want customers to use it recurrently. Is your product good enough?
Virality: it’s time to make users share your product/service, as in word-of-mouth.
Revenue: now you need to prove that your product/service can actually make money. Reaching a breakeven point as soon as possible is always desired.
Scale: finally, making sure that you can drive a consistent, sustainable and fast-enough growth is key to establish the company as a fierce competitor in the market.
You guessed it right! Each of these stages will have an OMTM that will specifically drive the transition into the following stage.
Now, you know 1. what model you are using, 2. at what stage you are in a particular moment in time, and 3. you actually know the metric/s you need to track in order to succeed.
But one particular question may arise in here, though. How do we know we are doing good enough?
Benchmarking your metrics: Draw Lines in the Sand
Drawing lines in the sand, or setting baselines for your metrics, will help you in two very important, different ways:
Letting you know that you are playing the game. If your metric has already been optimized as much as possible, is probably time to move on to the next one; if you are well below the competitors average, you definitely need to be aware of that.
Letting you know what game you are actually playing. Considering an up-to-date and field-specific metric is crucial in order to succeed. Baselines may (and will) vary greatly between different models and markets.
“It’s easy to get stuck on one specific metric that looks bad and invest considerable time and money trying to improve it. Until you know where you stand against competitors and industry averages, you’re blind. Having benchmarks helps you decide whether to keep working on a specific metric or move on to the next challenge.”
As stated in the book, thanks to the guys from Startup Genome and their wonderful Ecosystem Reports, that collect key metrics from thousands of startups and scale-ups, we learn that being average is usually not enough to thrive in this highly competitive market. And you definitely need to know whether your metrics are mediocre or top-notch in order to act accordingly.
CONCLUSION
This book covers the process and the intuition behind a Lean-driven analytics mindset in a startup, and actually most of the advice makes a lot of sense. Lots of use cases mentioned in the read - most of which I did not mention in the summary for the sake of brevity - will help you connect the dots between the general ideas and the key concepts.
Nevertheless, it might feel a little bit repetitive and difficult to read in case your business model does not align with any of the ones mentioned in the book, since a great part of the read falls upon the six models and the five stages mentioned previously above.
Do you fancy receiving a book summary every week right into your email inbox? If so, do not forget to subscribe, it is completely free!
Great summary! Waiting for the next one.