Comment la méthodologie Lean startup a-t-elle changé l’innovation en entreprise ?

AI can supercharge the lean startup methodology. Here are a few ways in which I've seen add a lot of value: Supercharged Design (Build): AI can help in swiftly creating MVPs and generating designs or code, quickening the path to market and learning. Exponential testing (measure): AI streamlines A/B testing, speeding up the validation of ideas and the learning process essential to the lean approach. Better Customer Insights (measure): AI can help to efficiently process feedback, and analyze research reports or customer data allowing companies do do faster product iterations. Smarter Decisions (learn): Companies leverage AI for quicker, data-backed decision-making, bypassing guesswork and directing product strategy with precision.

Lean is about learning as fast as you can. Understanding what you need to learn and then figuring out the fastest most reliable and most ethical way to learn that thing and move on to the next assumption, question, risk etc. Bonus points if your learning objectives start with the hardest most unknown and most human centric things knowing that humans are irrational and unpredictable.

In my experience, lean startup is a way to test new projects and get information. The ability to get data and information from the clients is priceless. The problem is that companies don't get the basis of innovation accounting and that makes the implementation of lean startup difficult. As companies we need to explain the method not only to the project developers but to the finance department and executive team that has to make changes and the way they finance the projects and measure results.

The core idea behind the Lean Startup methodology is to create a more efficient and effective path to building a successful business by focusing on validated learning and iterative product development. The key components of the Lean Startup methodology: Start with a hypothesis Build a Minimum Viable Product (MVP) Measure Learn Pivot or persevere Build-Measure-Learn feedback loop: The Lean Startup methodology is an iterative process where you continuously cycle through these steps, gradually building and improving your product based on customer feedback and validated learning.

I think it is great for kicking off a customer centric approach that builds capacity for eventual creation of futuristic consumer products, as a result of anticipating customer behavior overtime.

The build measure learn loop is the process we go through to reduce risk, by focusing our resources on building the things that matter- the things that will allow us to learn we need to learn about our product or service. The faster we move through this loop, the higher our odds of success are as it allows us multiple iterations and opportunities to design the thing that customers want. A helpful tip is to think about it as a learn-measure-build loop and ask yourself the following: - What do I need to learn about my business? (learn) - How would I know that I have learned that thing? (measure) - What do I need to build to most efficiently achieve that metric of success (build)?

The Build-Measure-Learn is the core process of the Lean Startup methodology. It involves: 1. Build: Create a Minimum Viable Product (MVP) based on your initial hypothesis about your product or business. 2. Measure: Collect data and feedback by releasing the MVP to real users. Track relevant Key Performance Indicators (KPIs) to understand how users are engaging with the product. 3. Learn: Analyze the data and feedback to determine to validate your hypothesis. If it's not, pivot or continue with the current strategy based on what you've learned. Continue iterating through the loop to refine your product and strategy. This iterative process helps us make data-driven decisions, refine their product, and adapt to changing market conditions.

The quality of your hypotheses and assumptions are key ingredients to determine the success of a build-measure-learn loop. Make sure you invest time in clearly establishing your riskiest assumptions and associated hypotheses, but also that you prioritize them in the right order: you want to be building, measuring and learning against your riskiest assumptions and most foundational hypotheses first so that you can pivot early if needed.

An MVP can also be characterized as any vehicle used to derisk a risky assumption. For example a landing page is not a full product per se but when used correctly and responsibly it can help answer important questions like “does my USP resonate” “will people pay” “How much will people pay” “what revenue model might work” etc

When creating an MVP, it's essential to ensure that it provides tangible value; otherwise, your idea is unlikely to succeed. For instance, if your ultimate product is a car, developing an MVP that consists of just a steering wheel won't attract interest. However, if you opt to create a skateboard as your MVP, it offers more value in comparison to just a steering wheel.

Here is an MVP landing page example that I recently built for my own startup: - I first worked on a general unique selling proposition (USP). - I then asked ChatGPT to write a press bulletin using this USP as if I was launching a real product. - I used the features and the bulletin language to create an MVP landing page. - I announced on my Slack groups that I was looking for users to test my MVP. - I showed my landing page to several testers, and did their screen recording via Google Meet. - I extracted the subtitles of the session recordings and analyzed the sentiments of the testers via ChatGPT. - With the learnings I derived, I am now working on a prototype to test the features that resonated well with my test group.

A Minimum Viable Product (MVP) is the simplest version of a product that allows a startup to test its core hypothesis, gather valuable user feedback, and learn about customer needs and preferences. It includes essential features to address the primary problem or need, while minimizing development time and cost. The MVP helps startups iterate and refine their product based on real-world user interactions and data, ultimately leading to a more successful and customer-centric final product.

An often overlooked aspect of developing MVPs is the unconscious bias that the fidelity level can introduce during testing. While it's essential to avoid unnecessary time spent on polishing an MVP, it's equally important to consider your testing audience and understand elements that might distract their assessment. For example, when testing with kids, we found that girls paid closer attention to the aesthetics of a MVP than boys - which could often make them disengage from the whole experience rather than isolating aesthetics from function or play. Make sure you have a strong foundational understanding of your audience in order to truly benefit from the MVP process.

An important differentiation to make is between iterating and pivoting. Iterating involves small tweaks or changes to satisfy testing feedback, but a pivot is a much greater strategic move that can change the course of your product development. It is not aways easy to know when to iterate further, or when it is time to make a more significant pivot. This is where it is important to let your risky assumptions guide your decision making. They should be viewed as the legs of a chair, if one is not validated, the whole chair is useless. Make sure these foundations are solid before iterating a three-legged chair to death only to find it is still useless to sit on.

We expanded our company in 2020… Yes. Worst possible time. But, we did it. Let’s go! Be creative. Solve big problems! Thanks - Sheila

"Validated" learning... that is, learning that is NOT based on an assumption or someone's opinion, is crucial to diagnosing when it's time to pivot. You can't manage what you don't measure.

The Lean Startup method not only reduces uncertainty and speeds up innovation but also brings overlooked advantages. It optimizes resources, ensuring startups prioritize crucial features (i.e. the ones with the most impact). The approach fosters adaptability, enabling swift responses to market changes. Continuous improvement is key, while the customer-centric approach improves product-market fit.

The biggest unexpected benefit of lean startup methodology is that moving at pace can force you to prioritize much more effectively and clearly identify your unique and most valuable differentiators. It is about spending time on the most critical features and functions first versus taking on the full proposition too early and thereby diluting your understanding of your most valuable assets and unassailable strengths.

Lean startup methodology relies quite heavily on a shared understanding of progress and learnings in order for the entire team to move at the same pace and in the right direction. This can offer require in-person collaboration where a sort of 'information osmosis' can more easily occur, but often teams are split across time zones and markets, making the close collaboration needed for this process to be most effective. To help support this, teams need to find ways to maximize time apart for iterative learning and to share information and learnings clearly, concisely and often.

I concur that these challenges exist, but a potential solution is to assemble the right team to implement this methodology. It's crucial to have individuals who are highly motivated and passionate to ensure the success of your efforts; otherwise, the efforts will go into vain.