Lean Analytics · Benjamin Yoskovitz, Alistair Croll · 2013
The Lean Startup focuses on quickly learning about your greatest risks in pursuit of a scalable and repeatable business model. Lean Analytics aids in tracking progress and deciding what to concentrate on next, as part of the build-measure-learn loop.
Entrepreneurs have to be relentlessly optimistic. They create a reality distortion field of positivity to push through the odds stacked against them. Analytics is the necessary counterweight of realism.
Good metrics should inform our next move and change our behaviour.
Steps in the Lean Analytics Cycle:
- Choose a crucial metric to improve.
- Set a realistic target, using industry benchmarks as a guide.
- Determine how to enhance the metric.
- Find out what good users do, make it easier
- Brainstorm ideas, test them, implement the best
- Measure the impact of the change on the metric.
- If the change fails, pivot, try a different approach, or conduct more discovery.
- If the change succeeds, select a new metric to improve.
The Lean Canvas is a great way of consistently articulating your hypotheses and verifying it with real customers:
- Problem: Identify real problems that people know they have.
- Customer Segments: Define your target market and the messaging that will help you reach them.
- Unique Value Proposition: What is the clear, distinctive, and memorable way to explain how you're better or different?
- Solution: Can you solve the problems effectively?
- Channels: How you’ll deliver your product to customers and get payment in return.
- Revenue Streams: One-time or recurring? Direct or indirect transactions.
- Cost Structure: Identify the direct, variable, and indirect costs.
- Metrics: Do you know which numbers to track to understand if you're making progress?
- Unfair Advantage: The force multiplier that gives you a greater impact than others
There are many data analysis pitfalls: assuming clean data, not normalising data, excluding/including outliers, ignoring seasonality, ignoring size in growth reports, metrics that cry wolf. Avoid overwhelming with too much data without actionable insights.
The Lean Startup is the process you use to move toward and achieve your vision. Early-stage founders aren’t building a product, they’re building a tool to learn what product to build.
The Lean Analytics stages suggest an order to the metrics you should focus on:
- Empathy: I've found a real, poorly met need in a reachable market
- Stickiness: I've figured out how to solve the problem in a way that they will accept and pay for
- Virality: The users and features fuel growth organically and artificially
- Revenue: I've found a sustainable, scalable business with the right margins in a healthy ecosystem
- Scale: I can achieve a successful exit for the right terms
Choose the One Metric That Matters (OMTM). The one that’s crucial given your current stage. Success lies in focus and discipline. Optimising your OMTM squeezes that metric so you get the most out of it, but it also reveals the next place you need to focus your efforts, which often happens at an inflection point.
Draw lines in the sand. Establish a clear target number for success metrics, and be ready to reassess if it's not met. Success can be defined by what your business model needs or by comparing to industry norms when the business model is not yet clear.
To decide which metrics you should track, you need to be able to describe your business model simply and just think about the really big components.
Business growth comes from improving one of these five knobs: selling more stuff, to more people, more often, for more money, more efficiently
Not all customers are beneficial. While some are valuable, others can be distracting, resource-consuming or harmful.
The Problem-Solution Canvas helps you maintain discipline and focus on a weekly basis. What’s the problem, how do you propose to fix it, and how will you know if you succeeded? Use it to home in on the key problems you’re facing. Agree on and prioritise your problems.
- Current status: key metrics you’re tracking compared to previous period
- Last week’s lessons learned: What did you learn, what was accomplished?
- Top problems: List and prioritise the top problems
For each problem:
- Hypothesised solutions: list the possible solutions. Rank them. Explain how you think they’ll solve the problem
- Metrics / Proof / Goals: List the metrics you’ll use to measure the solutions, list any qualitative proof you’ll need. Define the goals for the metric.
There is a normal or ideal for most metrics, and that normal will change significantly as a particular business model goes from being novel to being mainstream.
How you feel about a metric changes when you know what’s normal or best in class for your type of business. Before investing time and effort into a metric, understand your position against competitors and industry averages. Benchmarks guide whether to continue working on a metric or tackle the next challenge.
Optimisation efforts usually have diminishing returns, indicating when it's time to shift focus to a different metric.
Achieving local maxima and seeing diminishing returns on improvements can serve as a baseline and indicate when to shift focus to other areas.
Sketch your business model and key metrics:
Subscribe Button Quick Links
12 sources of insights for Product Discovery · Article
Non-Obvious Signs of Early Traction · Article
Fooled by Stickiness (DAU/MAU) · Article
7 Categories of Product Strategy Questions · Article
We're going to pick one thing from this list, and fix it by the end of the day · Tweet
The typography of books · Article
Predevelopment Activities Determine New Product Success Robert G.Cooper · 1988
This article focuses on the predevelopment activities of the product innovation process: those often ignored steps that precede the actual development of the product.
First, we look at the mounting evidence that identifies gaps in how many industrial firms handle the predevelopment steps. The evidence also reveals that new product success and failure is often decided before the new product project even enters the product development phase.
Second, we discuss ways that managers can, should, and have improved the effectiveness of these early and crucial stages of the innovation process.
The paper's reference to early "pre-development" activities can be seen as a precursor to what we now refer to as "product discovery" techniques. Product discovery has evolved significantly over time, transforming from fairly simple market research and concept definition to a robust and multidimensional process. Today, it involves rigorous user research, data analysis, iterative testing, and validation.
Key insights from the paper:
- Predevelopment activities critically influence new product success.
- Most product failures can be traced back to inadequate predevelopment steps.
- Essential predevelopment activities include market analysis, product concept definition, and project evaluation.
- Market research and understanding customer needs are foundational for product success.
- A strong market orientation and a superior product concept are key success factors.
- Early project evaluation should assess technological feasibility and market potential.
- Successful new products often exhibit a high performance-to-cost ratio and strong launch support.
- Effective predevelopment involves idea generation, preliminary assessment, and detailed concept definition.
- Financial considerations and resource allocation in early stages are linked to product success.
- Systematic predevelopment processes, including rigorous screening and market assessment, improve product outcomes.
/w=3840,quality=90,fit=scale-down)
Book Highlights
Write down what the product as soon as you can guess. He begins with 1) internal combustion engine; 2) four wheels with rubber tires; 3) a transmission connecting the engine to the drive wheels; 4) engine and transmission mounted on metal chassis; 5) a steering wheel. By this time, every student will have written down his or her positive identification of the product as an automobile, whereupon Scott ceases using features to describe the product and instead mentions a couple of user goals: 6) cuts grass quickly and easily; 7) comfortable to sit on. From the five feature clues, not one student will have written down "riding lawnmower." You can see how much more descriptive goals are than features. Alan Cooper · The Inmates are Running the Asylum
Simpson’s paradox, a phenomenon in which a trend appears in several groups of data but disappears or reverses when the groups are combined. Chip Huyen · Designing Machine Learning Systems ·
The fact that you are not present means that you cannot ask questions as they arise, such as, “Why did you click that button?” You must provide all the questions you'd like the user to answer in advance—so you need to give more thought and attention to detail to how you word the questions compared to moderated testing. Dan Olsen · The Lean Product Playbook ·
Routine tasks such as forming up a column of march or deploying a skirmishing line were standardized and everybody was trained in how to do them. Today, they include things such as forming a road block, and are called standard operating procedures or SOPs. They are very useful because they create uniformity and therefore predictability where that has high value. They enhance efficiency by enabling these tasks to be carried out at speed with little supervision. Stephen Bungay · The Art of Action
/w=3840,quality=90,fit=scale-down)
Quotes & Tweets
If you ask other people about every decision you make, you're going to end up doing exactly what everyone else does and getting the same results that everyone else gets. Shane Parrish
Know precisely what you want, Determine the cost of what it will take to get it, Don't bargain over the price. James Clear